From 8730e575053fae727644f672dfc1578cc6a77846 Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:12:04 +0200 Subject: [PATCH 1/8] Fix typos and change order of sections --- manual.md | 228 +++++++++++++++++++++++++++++------------------------- 1 file changed, 123 insertions(+), 105 deletions(-) diff --git a/manual.md b/manual.md index ca5c5cc..0dda021 100644 --- a/manual.md +++ b/manual.md @@ -1,12 +1,16 @@ # Manual -Welcome to Promod's manual. Here you will find all the information you need to run this software. This program is a project for the Structural Bioinformatics subject and the Introduction to Python project of the Msc in Bioinformatics of Universitat Pompeu Fabra. +Welcome to Promod's manual. Here you will find all the information you need to run this software. This program is a project for the Structural Bioinformatics and Introduction to Python subjects of the MSc in Bioinformatics for Health Sciences at Universitat Pompeu Fabra (Barcelona). + + ## What is Promod? -Promod is a python tool whose goal is to form macrocomplexes of molecules starting from the interacting pairs of chains that will form the complex. It is compatible with protein chains, single and double DNA strands and RNA strands. It has several parameters available to play with according to the users needs. +Promod is a Python tool whose goal is to model macrocomplexes of molecules starting from the interacting pairs of chains that will form the complex. It is compatible with protein chains, single and double DNA strands and RNA strands. It has several parameters available to adjust according to the users needs. + +In essence, the builder approach is similar to a genomic assembler: it seeks for similar chains in the different PDB files given as input and overlaps them. The builder places pair by pair all the possible chains into a single model. + -In essence, the builder approach is similar to a genomic assembler: seeks for similar chains in the different PDB files given as an input and overlaps them. After that, joins the chains in a single model. Pair by pair the builder puts all the possible chains inside a single model ## Background and scientific explanation @@ -20,7 +24,7 @@ Regarding the structural properties of the proteins, there are three main catego In this work, our aim is to develop a software that builds protein macrocomplexes using as input pairs of protein-protein interactions. Interaction with other molecules, such as nucleic acids, is also considered. To do this, our program is based on homology modelling. Therefore, the homology of the different chains is analysed, tridimensional structures are superimposed and energy levels of the final models are considered to propose the best possible solution. Another option would have been template-free modelling, but the required computational resources and the steps of evaluation and refinement excluded this possibility, because of the limited means and knowledge that our team has got. -Here we propose Promod as a first step approach to protein-protein and protein-nucleic acid complex modelling. It is distributed as both a Python package and a standalone application to be run in UNIX-based systems. Several parameters can be tuned to adjust to the user's needs. The final result is a single model with the best option according to the algorithm that is explained below. +Here we propose Promod as a first step approach to protein-protein and protein-nucleic acid complex modelling. It is distributed as both a Python package and a standalone application to be run in both UNIX-based and Windows operating systems. Several parameters can be tuned to adjust to the user's needs. The final result is a single model with the best option according to the algorithm that is explained below. @@ -49,6 +53,8 @@ Lastly, it is important to consider the final energy levels of the complex. A go After that, it is possible that the position of the atoms inside the model is not the most adequate. Sometimes, despite having evidence of interaction between two chains, collisions or clashes appear when they are joined in the structure. Taking this into account avoids impossible models, such as those with chains crossing with each other, which will be thermodynamically unstable and unlikely to happen. + + ## Features The most striking features of Promod are: @@ -59,6 +65,8 @@ The most striking features of Promod are: 3. Graphical user interface (GUI), with the same functionalities as the command line interface (CLI). + + ## Implementation @@ -66,6 +74,8 @@ The Promod package is implemented in Python3. It strongly depends on the Biopyth The formatted documentation of the package includes dependencies and installation instructions, examples of use and a full tutorial with sample code. Each function has got its own documentation that instructs on the particularities when importing them to be used independently. The users could learn all knowledge of the library by looking up the detailed documentation. The main functionalities are further explained in the tutorial section. + + ## Discussion Promod is a basic tool that relies on information from both the sequence and the tridimensional structure of pairs of interacting molecules to build a final complex model. As we can see in the analysed examples, it is successful with small complexes and it can even handle nucleic acids interaction with proteins. The MODELLER final step adds an additional refinement of the proposed model, returning to the user a good quality structure. All this methodology is built in an easy-to-use package, well documented and with a GUI to save unpleasant command line work for the standard user. @@ -78,6 +88,8 @@ As we can see, template-based methods are leading the in silico modelling techni A large number of information is nowadays available from high throughput experimental techniques and a lot of structural bioinformatics software has been developed to integrate these data. The best performance of in silico techniques has been achieved at the tertiary structure level of proteins. Nevertheless, quaternary structures are the ones responsible for the majority of biological functions and their knowledge is essential to disentangle protein interaction networks in both physiological and pathological scenarios. It is clear that hybrid approaches, joining atomic-level experimental structures, database information and the newest machine learning techniques, will give promising results in the near future. + + ## Future perspectives As we have already mentioned, there are diverse strategies for modelling PPIs. The development of refined algorithms to perform this task is far beyond our current knowledge in structural bioinformatics. Therefore, if we had time and resources to expand our project, we would focus on improving the user experience and the accessibility to existing data. @@ -90,106 +102,125 @@ Another functionality that could be feasible to achieve is the automatic downloa Finally, it would be ideal to develop integration options of biological information to help build better complexes, in line with trending research in the field. We are not aware of Python packages that would allow to directly implement this kind of work, but maybe external tools are capable of doing it. However, a broad review of the latest literature and a challenging programming development would be needed to know the most promising approximations. + + ## Conclusion Although computational approaches for building macromolecular complexes are far to be perfect at the moment, a lot of effort is being made to develop strategies to overcome the pitfalls in this field. We have provided a software that, despite not using any novel strategy, achieves notable results when using already defined structures. Further steps using machine learning approaches and a broader range of training data would improve the performance and confidence of this type of programs. Finally, the integration of different types of biological data will also be a key step in the progress of in silico modelling of protein-protein interactions. The achievement of better methodologies will definitely have an impact in applied fields, such as drug discovery, biomedical research or food industry. -### Input -The input is a folder with two or more files: each file contains two proteins, a protein and DNA/RNA strand or 2 single DNA strands (which may or may not form a double strand), which represent an interaction between those chains. Two structures are considered to be interacting if the minimum distance between them is in a range of a few Argmstrongs (1-10) . However, in lesser distances, forces between atoms are strong enough to produce changes between the chains and force them to adopt a different conformation, an thus, a realistic model must keep the residues at an adequate distance to consider it correct. -### Output +## Bibliography + +Chang, J., Zhou, Y., Qamar, M. T. U., *et al.* Prediction of protein–protein interactions by evidence combining methods. *International Journal of Molecular Sciences* **17**, 1946 (2016). doi: 10.3390/ijms17111946 + +Ding, Z., Kihara, D. Computational identification of protein-protein interactions in model plant proteomes. *Scientific Reports* **9**, 8740 (2019). doi: 10.1038/s41598-019-45072-8 + +Hayes, S., Malacrida, B. , Kiely, M., Kiely, P. A. Studying protein–protein interactions: progress, pitfalls and solutions. *Biochemical Society Transactions* **44**, 994-1004. doi: 10.1042/BST20160092 + +Liu, S., Liu, C., Deng, L. Machine learning approaches for protein–protein interaction hot spot prediction: progress and comparative assessment. *Molecules* **23**, 2535 (2018). doi: 10.3390/molecules23102535 + +Nealon, J. O., Philomina, L. S., McGuffin, L. J. Predictive and experimental approaches for elucidating protein–protein interactions and quaternary structures. *International Journal of Molecular Sciences* **18**, 2623 (2017). doi: 10.3390/ijms18122623 + +Sarkar, S., Gulati, K., Kairamkonda, M., *et al.* Elucidating protein-protein interactions through computational approaches and designing small molecule inhibitors against them for various diseases. *Current Topics in Medicinal Chemistry* **18**, 1-18 (2018). doi: 10.2174/1568026618666181025114903 + +Keskin, O. , Tuncbag, N. , Gursoy, A. Predicting protein-protein interactions from the molecular to the proteome level. *Chemical Reviews* **116**, 4884-4909 (2016). doi: 10.1021/acs.chemrev.5b00683 + + -The output is mainly one pdb file with all the possible chains joined in the selected folder. However, if the optimized option was selected, several pdb files will be created in the same directory. These are different approaches made by modeller to optimize the energies of the model and the files it used. The model will be saved as final_model.pdb -If the model is required with minimum energy, and thus the -optimize flag is used, then several files will be saved. Modeller will save some mid-step pdbs as 'final-model.DXXXXX.pdb' and the fully optimized will be saved as 'optimized.pdb' ## Tutorial ### Installing dependencies -In order to make Promod work, there are some dependencies that must be installed before executing the software: python3 and Biopython +In order to make Promod work, there are some dependencies that must be installed before executing the software: Python3 and Biopython. -If you're using Windows, you can download python3 from [its website](www.python.org). If you got Linux, there's high chance that you already got it installed. +If you are using Windows, you can download Python3 from [its website](www.python.org). If you have got Linux, it is likely that you already have got it installed. -Independently of the operating sistem you got, you can install biopython using pip. +Independently of the operating system you run, you can install Biopython using `pip`. ```{sh, eval=F} pip3 install biopython ``` -### Instalation +### Installation -The recommended way to install promod is by using pip, exaclty with the same command as we did with biopython previously +The recommended way to install Promod is by using `pip`, exactly with the same command as we indicated with Biopython above. ```{sh, eval=F} pip3 install promod ``` -You can install Promod easily by downloading it from the github repository and running the next command in the downloaded folder. This is, for now, the recommended install way +You can install Promod easily by downloading it from the Github repository and running the next command in the downloaded folder. That is the recommended installation procedure. ```{sh, eval=F} pip3 install . ``` -Alternatively, you can also run the setup script installation commands. +Alternatively, you can also run the `setup` installation script, as follows: ```{sh, eval=F} python3 setup.py install ``` -### Hands on: how to use promod +### Hands on: how to use Promod -Promod has a command line interface which explains briefly how to use it before executing it. +Promod has a command line interface which explains briefly how to use it before execution, when using the `-h` flag. -```{sh} +``` promod -h ``` -There are 3 mandatory commands while the rest are optional. The mandatory ones are related to the input files and output folder, necessary for the program to work, and they are *-i, the input folder; -o, the output folder; and -f, the fasta file of the sequences*. All of them will be covered in this manual. -It also has a graphical interface wich accepts the same paramenters but graphically with message boxes and graphical stuff. It's an independent executable, so the CLI one can be used for automation without confusion. The comand is simple: +There are 3 mandatory commands, while the rest are optional. The mandatory ones are related to the input files and output folder, required for the program to work. They are `-i`, the input folder; `-o`, the output folder; and `-f`, the fasta file of the sequences. All of them will be covered later in this manual. -```{sh eval=F} -promod-tk +It also has a graphical interface which accepts the same parameters, but graphically with message boxes and graphical aid. It is an independent executable, so the CLI can be used for automation by itself. The command to start the GUI is simple: + +```sh +promod-tk ``` -#### Paramenters +#### Input -##### Input folder +The input must be a folder with two or more files: each file containing two proteins, a protein and a DNA or RNA strand or 2 single DNA strands (that do not necessarily form a double strand), which represent an interaction between those chains. Two structures are considered to be interacting if the minimum distance between them is in a range of a few Angstroms (1-10 Å) . However, in shorter distances, the forces between atoms are strong enough to produce changes between the chains and promote a different conformation, and, thus, a realistic model must keep the residues at an adequate distance to consider it correct. -The input folder should contain, at least, 2 pdb files. This folder may contain other files or subfolders; however, the program will ignore other files and will not check subfolders to find more pdb files. If you want to include some pdb, include it in the folder before running the program. +#### Output -Beyond that last thing, no more things are necessary to know about the input folder. However, it's a mandatory argument, and should be indicated with -i or --input-folder tags +The output is mainly one PDB file with all the possible chains joined in the selected folder. However, if the optimized option was selected, several PDB files will be created in the same directory. These are different approaches made by MODELLER to optimize the energies of the model and the files it used. The model will be saved as `final_model.pdb`. -##### Output folder +If the model is required with minimum energy, and thus the `-optimize` flag is used, then several files will be saved. MODELLER will save some mid-step PDB files as `final-model.DXXXXX.pdb` and the fully optimized complex will be saved as `optimized.pdb`. -The output folder is the folder where the output files are written. This is mandatory, and no special folder is required. Just a warning, as, for the moment, the model file is written as 'final_model.pdb' and thus any other file with that filename will be overwritten. +#### Parameters -The output folder is indicated with the -o or --output-folder tags - -#### Distance - -You can indicate the minimum distance to consider that two proteins do not clash. The model will discard interactions with too many atoms at lesser distance than the value indicated in Argmstrongs. Being too strictive can discard some interactions, but being too flexible can force chains to overlap and not being energetically favorable. +##### Input folder -You can indicate the distance with the -d or --distance tags +This is a mandatory argument and should be indicated with the `-i` or `--input-folder` tags. The input folder should contain, at least, 2 PDB files. This folder may contain other files or subfolders. However, the program will ignore other files and will not check subfolders to find more PDB files. If you wanted to include some other PDB, you should copy it in the folder before running the program. -#### Threshold +##### Output folder -The threshold is the minimum score alignments must reach in order to insert the chain in the model and determinate its homologous protein. Usually a very high value (0.9 or higher) is recommended to ensure that the chains are correctly identified. However, for sequences with less homology might be needed to reduce this value. +This is a mandatory argument and should be indicated with the `-o` or `--output-folder` tags. The output folder is the directory where the output files are written. At the moment, the computed model file is saved as `final_model.pdb` and, thus, any other file with the same filename will be overwritten. ##### Fasta file -The fasta file contains the sequences of the chains and the id for the stoichiometry. And it's a critical file, as the sequences must be homologous for the chain in the pdb. We can hold up to 5% of differences (by default, using -t parameter can modify this threshold). However, if one of the chains (if no stoichiometry indicated) or one of the chains in the indicated stoichiometry differs too much, the program will exit. This is use to avoid guessing posible chains that could match as that would produce unexpected results. +This is a mandatory argument and should be indicated with the `-f` or `--fasta` tags. The fasta file contains the sequences of the chains and the identifiers for the stoichiometry. It is a critical file, as the sequences must be homologous to the chain in the PDB file. We can hold up to 5% of differences (default threshold value which can be modified using the `-t` parameter). However, if one of the chains (if stoichiometry is not provided) or one of the chains in the indicated stoichiometry differs largely, the program will exit. The reason for this behaviour is to avoid guessing possible chains that could randomly match, as that would produce unexpected results. -**Note:** That means the sequences in the fasta should be of a similar length than equivlent chain in the pdb. For example, the fasta file and the pdbs directly downloaded from [Protein Data Bank](https://www.rcsb.org/) may differ in the initial and ending residues, as they are quite difficult to model, and usually are removed from PDBs. +**Note:** The sequences in the fasta file should be of a similar length than their counterparts in the PDB file. For instance, the fasta files and the PDB files downloaded directly from [Protein Data Bank](https://www.rcsb.org/) may differ in the initial and ending residues, as they are difficult to model, so they are usually removed from PDB entries. -#### Stoichiometry file +##### Distance -An additional file with the stoichiometry can be given to the program to make sure that the final protein will contain no more than the indicated chains. This file must be properly formatted as follows: +This parameter can be indicated with the `-d` or `--distance` tags. You can indicate the minimum distance to consider that two proteins do not clash. The model will discard interactions with too many atoms at lower distance than the value indicated in Angstroms. Being too restrictive may discard some meaningful interactions, but being too flexible can force chains to overlap and produce interactions that are not energetically favorable. -chain_id_in_fasta,number_of_columns +##### Threshold + +This parameter can be indicated with the `-t` or `--threshold` tags. The threshold is the minimum score an alignment must reach in order to insert the chain in the model and determine its homologous protein. A very high value (0.9 or more) is usually recommended to ensure that the chains are correctly identified. However, for sequences with less identity, it might be needed to decrease this value. + +##### Stoichiometry file + +This parameter can be indicated with the `-s` or `--stoichiometry` tags. An additional file with the stoichiometry can be provided to make sure that the final protein will contain a specific number of chains. This file must be properly formatted, one value per line, as follows: -one per line +``` +chain_id_in_fasta,number_of_columns +``` -Let's see the stoichiometry file of the example 1 +For example, this is the stoichiometry file for the example number 1: ```{pseudocode} 3e0d_A,2 @@ -197,53 +228,54 @@ Let's see the stoichiometry file of the example 1 3e0d_C,2 ``` -#### Starting pdb +##### Starting PDB -Promod's result is highly dependant on the first pdb used to build the model. In fact, results can be very different by varing the starting pdb. Thus we provide an optional argument to select this staring pdb to select different starting pdbs. This way, the same pdb with the same parameters will provide the same model. By default, the pdb files are sortered alphabetically and the first one in this sorted list is chosen. - -The starting pdb can be selected with -start +This parameter can be indicated with the `-start` or `--start` tags. Promod's result is highly dependent on the first PDB used to build the model. In fact, results can vary a lot when different starting PDB files are used. Therefore, we provide an optional argument to select the starting PDB file to account for this variation. This allows to use the same PDB file with the same parameters, to obtain the same model. By default, the PDB files are sorted alphabetically and the first one in the list is selected. ### Uninstalling -You can uninstall everything by using if installed using pip3 +You can uninstall this software by using the following command, if it was installed using `pip3`: -```{sh, eval=F} -pip3 unistall promod +``` +pip3 uninstall promod ``` -However, if installed calling setup.py, files must be deleted manually. +However, if it was installed via `setup.py`, all files must be deleted manually. -## Examples -### Example 1 (3e0d) -3e0d is a small complex formed by 2 protein chains and 2 double DNA strands. It's a subunit of the eubacterial DNA polimerase but it's perfect for an initial testing of our project, so we can test how the program works, the time it takes and if the result is similar to the original one. In this cases, we have 6 different interactions, as each DNA strand is counted as a single one interacting with another one and binded to the protein chain by one of them. In fact, this complex can be thought as a dimer: two monomers formed by a protein and a double strand DNA, which interact by some residues in their protein chains. -So, the first test to our program consists on joining the different pdb files into one single structure, without further information about the stoichiometry. Therefore, the program will try to build the protein with only the information provided by the pdb files and the fasta file. The command to build it is the next one: -![3e0d structure obtained from: a) rcsb.com b) promod without any extra info c) promod with starting point and selected stoichiometry d) without stoichiometry control][3e0d_composed] +## Examples +### Example 1 (3e0d) -```{sh, eval=F} +3e0d is a small complex formed by 2 protein chains and 2 double DNA strands. It is a subunit of the eubacterial DNA polimerase, but it is perfect for an initial testing of our project. We can test how the program works, the time it takes to run and if the result is similar to the original one. In this case, we have 6 different interactions, as each DNA strand is counted as a single one interacting with another one and bound to the protein chain by one of them. In fact, this complex can be thought as a dimer: two monomers formed by a protein and a double strand DNA, which interacts with some residues in their protein chains. + +The first test to our program consists of joining the different PDB files into one single structure, without further information about the stoichiometry. Therefore, the program will try to build the protein with only the information provided by the PDB files and the fasta file. The command to build it is the next one: + +``` promod -i examples/example_1/chains/pairs -o examples/example_1/results \ -f examples/example_1/3e0d.fa ``` -We just need to give the folder with the input pdbs (-i), the desired output folder (-o) and the fasta file with the sequences of the proteins. The program will take the different pdbs on the folder (By alphabetical order, to make it reproducible) And the result it's incomplete: +We only need to provide the folder with the input PDB files (`-i`), the desired output folder (`-o`) and the fasta file with the sequences of the proteins (`-f`). The program will take the different PDB files on the folder (by alphabetical order, to make it reproducible) and the result is incomplete: + +![3e0d structure obtained from: a) rcsb.com b) promod without any extra info c) promod with starting point and selected stoichiometry d) without stoichiometry control][3e0d_composed] -There we can see that there is a missing double strand DNA. So, let's see how we can complete the model. We can see that there is a missing protein, let's force the program to include it. We have two different ways of indicating this: The easiest one is selecting the starting complex of our model (which contains those missing parts). This results that different starting points can result in different models, so, it's important to select a good one. By default, the pdbs filenames are sortered alphabetically and the first one is chosen as the starting point. In fact, this is an important choice: the differences between not chosing one (Figure 1.b) chosing an adequate one (Figure 1.c) or chosing another one (Figure 1.d) can result in models with extra chains or with less chains than expected +We can observe that there is a missing double strand DNA. So, we will take a more elaborate approach to complete the model. As there is a missing protein, we should force the program to include it. There are two different ways of indicating this. The easiest one is selecting the starting complex of our model (which contains those missing parts). Different starting PDB files can result in different models, so, it is important to select a suitable one. By default, PDB filenames are sorted alphabetically and the first one is chosen as the starting point. As we said before, this is an important choice that is evidenced when we compare not choosing a starting point (Figure 1.b), choosing an adequate file (Figure 1.c) or choosing another one (Figure 1.d), which in this case results in models with additional or less chains than expected. -The other not-so-difficult way is to indicate the stoichiometry of the complex. In this case we will focus in selecting the starting pdb, so the command would be. +The other available option is to indicate the stoichiometry of the complex. We will focus on selection the starting PDB file, so the running command would be: -```{sh, eval=F} +``` promod -i examples/example_1/chains/pairs -o examples/example_1/results \ - -f examples/example_1/3e0d.fa -start examples/example_1/chains/pairs/3e0d_ZG.pdb + -f examples/example_1/3e0d.fa -start examples/example_1/chains/pairs/3e0d_ZG.pdb ``` -Yay! This is fare more similar to the original structure, very close to the original model. However, this is a very simple protein, and bigger complexes may are harder to build. However, from this example, we learnt that: +This is more similar to the original structure, very close to the initial model. However, this is a very simple protein, and bigger complexes may be harder to build. However, from this example, we have learnt that: -* The program can build a model without any indication. However, it might be incomplete o have extra chains. -* The starting pdb is a important choice that can influence the final model. Thus it is important to remember which starting point gave which result. The same starting pdb will gave the same output. +* The program can build a model without any indication. However, it might be incomplete or it might have additional chains. +* The starting PDB file is an important choice that can influence the final model. Thus, it is important to remember which starting point gave which result. The same starting PDB file will give the same output. ### Example 2 (6gmh) @@ -252,92 +284,78 @@ Yay! This is fare more similar to the original structure, very close to the orig ![6gmh structure obtained from a) rcsb.com b) builded without assistance c) selecting starting point and stoichiometry][6gmh_composed] -Thus, lets try to build it: +Our first try is the following one: -```{sh, eval=F} +``` promod -i examples/example_2/chains/pairs -o examples/example_2/results \ -f examples/example_2/6gmh.fa ``` -In this case, there are several chains that hasn't been added to the model, and some are repeated. So, let's select a different starting point and limit the chains with the stoichiometry, so at maximun there is 1 copy of each chain. +In this case, there are several chains that has not been added to the model, and some others are repeated. We should select a different starting point and limit the chains with the stoichiometry, so there is one copy of each chain, at most. -```{sh, eval=F} +``` promod -i examples/example_2/chains/pairs -o examples/example_2/results \ -f examples/example_2/6gmh.fa -start examples/example_2/chains/pairs/6ghm_VA.pdb \ -s examples/example_2/6gmh.stoic ``` -The 6gmh.stoic file contains the stoichiometry of the protein. In essence, is a csv file, without header, in which each line follows the same structure: *,*. Each chain_name must be in the fasta as a sequence +The `6gmh.stoic` file contains the stoichiometry of the protein. Basically, it is a csv file, without header, in which each line follows the same structure: *,*. Each chain_name must be in the fasta as a sequence identifier. -This starting pdb was not in the model, and with this we force it to be included in the model, and start growing from there. And this way, the result contains far more chains in the model and it's far more similar to the original one. +This starting PDB file was not in the model in our first attempt, so now we force it to be included in the computation, and start building from there. This way, the result contains more chains in the model and it is more similar to the original one. -### Example 4 (5nss) +### Example 3 (5nss) ![5nss structure obtained from rcsb.org vs composed one][5nss_composed] -Here we want to show how to use two parameters we haven't shown yet: distance and threshold. Distance is a value that considers if two atoms collide, which is not allowed, as two atoms very close would have very high energy (if the separation between them is less than distance, it's considered invalid. However, to avoid some errors, up to 10 atoms colliding are allowed to consider a valid interaction). Threshold, however, makes reference to the homology percentage between the pdb sequences between themselves and between the sequences in the fasta. This allows for certain flexibility in sequences with some mutations that may alter the conformation of the chain. +Here we want to show how to use two parameters we have not shown yet: distance and threshold. The distance is a value that avoids the situation when two atoms collide, which is not allowed because two very close atoms would have high energy. So, if the separation between them is less than the indicated distance, it is considered invalid. However, to avoid some errors, up to 10 colliding atoms are allowed. The threshold makes reference to the identity percentage between the PDB structures and the fasta sequences. This allows certain flexibility in sequences with some mutations that may alter the conformation of the chain. -```{sh, eval=F} +``` promod -i examples/example_3/chains/pairs -o examples/example_3/results \ -f examples/example_3/5nss.fa -d 2 -t 0.9 -start examples/example_3/chains/pairs/5nss_NG.pdb ``` -In this case, the result is quite good, but let's focus on the speed of the program. As the number of interactions grew, the time it takes for the program also increases. However, there are a few things to comment about it. +In this case, the result is quite good, but an important issue is the speed of the program. As the number of interactions grow, the time it takes the program to complete the modelling also increases. However, there are a few considerations about that: -1.- Adding a new chain that passes all the checks is the most computationally expensive moment. The number of checks grows as the number of chains of the model increases. +1. Adding a new chain that satisfies all the restrictions is the most computationally expensive situation. The number of attempts increases with the number of chains in the model. -2.- It takes less time to discard an interaction that has no homologous already in the model than adding a new chain. During the first steps of the program, until the model has grown enough, this is the most common case. The number of alignments done is relatively low as there are a few chains in the model, but it can be an issue +2. It takes less time to discard an interaction that has not an homologous molecule already in the model than adding a new chain. During the first steps of the program, until the model has grown enough, this is the most common case. The number of calculated alignments is relatively low as there are a few chains in the model, but this may become an issue at certain point. -3.- Once the model has a considerable number of chains, and particularly if there are several copies of a monomer, it's also noticeable that discarding an homologous protein because doesn't fit with the current parameters is a very time-consuming processes. The program tries to introduce the chain using all the possible alignments whose score is higher than the threshold. This is an important issue for proteins with a lot of copies of several monomers. +3. Once the model has a considerable number of chains, and particularly if there are several copies of a monomer, discarding an homologous protein because it does not fit with the current parameters is a very time-consuming process. The program will try to introduce the chain using all the possible alignments whose scores are higher than the threshold. This is an important problem for proteins with a lot of copies of several monomers. -4.- The stoichiometry must be carefully selected to reach the desired structure. It's possible that the program cannot construct a model with the desire stoichiometry, but it's also possible that the stoichiometry wasn't what the user meant +4. The stoichiometry must be carefully selected to reach the desired structure. It may happen that the program cannot construct a model with the desire stoichiometry, but it is also possible that the stoichiometry was not appropriate for such case. -### Example 5 (6om3) +### Example 4 (6om3) -In this last example, let's talk about how the software behaves about the number of input pdbs and stoichiometry related stuff. +In this last example, we will cover the software behaviour in relation to the number of input PDB files and the influence of stoichiometric restrictions. ![6om3 builded from: a) rcsb.com b) promod with stoichiometry][6om3_composed] -```{sh, eval=F} +``` promod -i examples/example_4/chains/pairs -o examples/example_4/results \ -f examples/example_4/6om3.fa -d 0.3 -t 0.95 \ -s examples/example_4/stoic.csv ``` -In this last example, let's talk about how the software behaves about the number of input pdbs and stoichiometry related stuff. - -There are som chains left in the protein builded with promod, which can be added used custom values of distance or an apropiate starting point, but let's ignore that and focus on topics still not covered in the manual. +We can see that there are some absent chains in the protein complex built with Promod, which can be added using custom values of distance or an appropriate starting PDB file. Nevertheless, we will focus on another topic that has not been commented yet. -If you indicate a stoichiometry file but it doesn't exist, it exits. We could have changed it to a non-stoichiometry builder but we think it's better to do things explicitly as you may or may not notice this error. +If you provide a stoichiometry file but it does not exist, the program will abort. We could have changed it to a non-stoichiometry builder, but we considered a better approach to explicitly state this situation and let the user manage this type of error. -An important thing about the stoichiometry is that is critical to make sure that all the model is builded as you want. For example, without any assistance, the work heres is quite good, but with a selected stoichiometry, the result is missing some chains. Here we want to say that is important to check the stoichiometry file to make sure the chains are correctly selected and the number of them is adequate to our purpose. +An important thing about the stoichiometry is that it is critical to make sure that all the model is built following the user preferences. For example, without any assistance, the modelling in this example is good. However, with a selected stoichiometry, the result has some missing chains. Here we want to stress that the user should carefully check the stoichiometry file to make sure the chains are correctly selected and the number of them is adequate to our purpose. ## Limitations -1. The sequences in the pdb and the sequences in the fasta must be similar. The program admits a certain degree of tolerance (which can be selected by the user using the -t parameter) but using sequences in the fasta which are fairly different from the pdb will not allow the assembly of the protein. This is particularly difficult for pdb in which protein tails are not well modelled and they're not present in the pdb but they're in the fasta. The pairwise alignment with the tail will drop the score of the alignment under the threshold, forcing to decrease it or even mistake to which sequence in the fasta corresponds to jus by chance. Thus, our recommendation is to prepare the fasta file with the sequence as similar as possible to avoid mistakes. In case you needed, in this package there is a script included, pdbsplit.py, which can give you the sequences of the chains inside the pdb file, avoiding this kind of errors up to ceratin degree. So, if you're looking for the effect of certain mutations in the desired protein, might be worthy prepare the fasta file according to the pdb sequences. - -2. The running time of the program is proportional to the number of pdbs selected and is affected by the order of the structures and the starting point. The most critical steps: Reading all the pdbs sequences (However, the higher number of interactions, more chains might be added). This is a tradeoff between number of interactions and speed. - -3. Selecting different starting points can produce different models. It's difficult to say which is the most adequate starting point, as may vary according to the goal of the assembly. - -4. The fasta file require to be specially prepared for this application: There should not be repeated sequences (with at least an alignment score equal or higher than threshold selected). This also applies for sequences with unknown aminoacids, as they are marked as X in the fasta file and therefore will mismatch with the sequence in the pdb. Those sequences are usually missing in the final model +1. The sequences in the PDB files and the sequences in the fasta files must be similar. The program admits a certain degree of tolerance (which can be adjusted by the user using the `-t` parameter), but selecting sequences in the fasta file which are significantly different from the PDB structures will not allow the assembly of the complex. This is particularly difficult for PDB files in which protein tails are not properly modelled and, therefore, they are not present in the structure, but they are contained in the fasta. The pairwise alignment with the tail will yield a low score, resulting in an incorrectly failed homology identification or even returning a fake corresponding sequence in the fasta just by chance. -## Bibliography - -Chang, J., Zhou, Y., Qamar, M. T. U., *et al.* Prediction of protein–protein interactions by evidence combining methods. *International Journal of Molecular Sciences* **17**, 1946 (2016). doi: 10.3390/ijms17111946 - -Ding, Z., Kihara, D. Computational identification of protein-protein interactions in model plant proteomes. *Scientific Reports* **9**, 8740 (2019). doi: 10.1038/s41598-019-45072-8 + Our recommendation is to prepare a fasta file as similar as possible to the structural sequence to avoid this kind of errors. We provide an additional script, `pdbsplit.py`, which can extract the sequences of the chains inside the pdb file, preventing this kind of issues up to certain degree. So, if you are investigating the effect of certain mutations in a target protein, it might be adequate to prepare the fasta file according to the PDB sequences. -Hayes, S., Malacrida, B. , Kiely, M., Kiely, P. A. Studying protein–protein interactions: progress, pitfalls and solutions. *Biochemical Society Transactions* **44**, 994-1004. doi: 10.1042/BST20160092 +2. The running time of the program is proportional to the number of PDB files selected and it is also affected by the order of the structures and the starting point. The most critical step is reading all the PDB sequences. However, as the number of interactions increase, more chains might be added to the complex. This is a trade-off between the number of interactions and computational speed. -Liu, S., Liu, C., Deng, L. Machine learning approaches for protein–protein interaction hot spot prediction: progress and comparative assessment. *Molecules* **23**, 2535 (2018). doi: 10.3390/molecules23102535 +3. Selecting different starting points can produce different models. It is uncertain to state which is the most adequate starting point, as this may vary according to the goal of the assembly. -Nealon, J. O., Philomina, L. S., McGuffin, L. J. Predictive and experimental approaches for elucidating protein–protein interactions and quaternary structures. *International Journal of Molecular Sciences* **18**, 2623 (2017). doi: 10.3390/ijms18122623 +4. The fasta files require to be specially prepared for this application. There should not be repeated sequences (with at least an alignment score equal or higher than selected threshold. This also applies to sequences with unknown aminoacids, which are usually marked as X in the fasta file and will cause a mismatch with the sequence in the PDB file. Those sequences will be usually missing in the final model. -Sarkar, S., Gulati, K., Kairamkonda, M., *et al.* Elucidating protein-protein interactions through computational approaches and designing small molecule inhibitors against them for various diseases. *Current Topics in Medicinal Chemistry* **18**, 1-18 (2018). doi: 10.2174/1568026618666181025114903 -Keskin, O. , Tuncbag, N. , Gursoy, A. Predicting protein-protein interactions from the molecular to the proteome level. *Chemical Reviews* **116**, 4884-4909 (2016). doi: 10.1021/acs.chemrev.5b00683 [3e0d_pure]: examples/example_1/imgs/3e0d_original.png [3e0d_fail]: examples/example_1/imgs/3e0d_fail.png From 05ee277f1862eae7062849a9b9b0c97097b7b5cb Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:12:25 +0200 Subject: [PATCH 2/8] New renderization of manual.md --- manual.pdf | Bin 869006 -> 870747 bytes 1 file changed, 0 insertions(+), 0 deletions(-) diff --git a/manual.pdf b/manual.pdf index d95b1fc2577627b4a0777cca48d1ea1410ebcfb1..adc6697aeef68f27b10e4534e7b68a5b8e5af42f 100644 GIT binary patch delta 177779 zcmZUaQ*bT{kga3ewr$(Sj&0k{AKSKVbH}!AJ3BV!%&q%2Z|k+Ys;m0zwK^`8eJhDQ zfeeJ5ExDhT3gD-$mAo~I?EhN7uhXC=i}jkAo}$)dQYzCVvzelTerAM$m1J^IU;$K{ zVUT}&kvEVadlm|Gdi!}rK+%1~u<`xfeSG?UdeQw4Fqx#MJ?y1y)?GeSH|y?}H7u?F zFYjZgulUv2dwj_feP{@f_jCWPXw_F% z6JXOK3%4gY84{SQw#>YESDqvNYwouEJEF7%_1VvPjVgE@O%v~*mSv&WV4{CCy=C1*&cLAn)%+azsmSV`Xq`Y?bAl$ zJ;>aT05BEO=0yN{c0;j{7hcYP_FXm(X>H>8!?ZPDT>$B-gYd=B4X%>~g&?OV2G-)M z|FlV8jRETRSGl>onw3pLlR)UK8=6}8x=DVyzQJVd_$Z5&Ur;{{^{KRLU?)?;$ ze!ASU_%)x`S{W4?A4!*k8ICx}AeUh-vkY&98UV45y&HYbN+3T`nH+{QEE4>~U0E*` zvMfPgcVCRY?b&8Jk%|NGwtSG{vk}hkBR6}HI9EV;B-oUsM*Pu(t#W^YliBnxR5e{ zo_QzmUmSx{0V#ZUcc!H4I<7~jx zY|RzCb`bmZe%}q%ny0&=m&3!Z(czRe$t;oFJwTf?n^f|~&I_WE;}c5cfXy*K&=SZe zk&6y#1=u9?Tenj=|6GY`Nr=D?=FiXq#he2S;Ol9->{t1tSzdVtgQkmD-a;3a0N6J0 z6}|GfN>FXtsoctMvsDy1^H{26Of!&RLQxPPo@>hag0%-x|lbo)e)T38{blnV&`kN)XoSNH%lS;1QYh#DSpbVMxRf zltWrR@;ETna$4bW@3uI(7+ak}0d^Fzan}hHA86Y(WshBssFk1{swjo%k3_S4C~>UX zwU$KIQr68iVkqo>`H%~p@Cy~5Xbf%xuVj`dQt3QAMqzII&CIq0^J4Bmyj(9Ot8KYL z1>dP%XHdW_>K+y>1%lao?|lrdWa%Fet{`c@&dZP1?;r|l9hkuagx#1a0J#O_>Mw|6 zMYIa5U<%jHZa)ahYOL6RtW)d}F3%U6cKJ^g!=xLS2uQa?AIVq*(Dgs2j$B~a;-l8V z^!w7?Zpvg*)-YdG{zYX6XtXf48PCdclQD)CpE$GNLA^?dko2)MITECE#beP;JVJz~ zAKDY4rE#(6YEG#`JX#_PfKqfbk-F>^ZfX#oM~*t&RSQ;cwF@e5k_cl)2o|7pvktXp z8r~bGORmU)gv*V+gcHH#JtbE|avHOP%ld`-JWL@JE!ys)mQGca!=XDxU01?a&`KIv`&*B5X8E)QUQJeuow_}fW8M#_=h z!8#JQEq(|P=bXtA0A2hEmMd^(&7KgH$1~lrkw830g65&uKhW^%Wc2PAs4nPpg@=xr zbwuK^-YW4#mOuvp30I)fa*!Hk-oCi35!7-}hl1xFte@*6kgmx#C@B5WjSi*fSHKxE z|IluxH@F`PU)q%Mi2VwF5s^@3<*O{u@ZSl19MIyE{VfPrz$yLPeALNj&8d+0m<0Gy*Y4+|YIC;c=gBA>`6ZvPDRR7Pr5>6aXgof$mD!x85h-%12z zI9k}-63Bf3pp=>p+j_thrkHU)0=|Gydupf)3L%o7i?F2c$o|2#p@d9oMc-sEQ}@pi zls*Tcgp{ERcKp-(-v=I!?xb`dj$94-oD@12juyBh!pe11EF0ZOoOP(Df4p2M=B}iU zYcL$~0=9W-PlI9bj~(2;LN7fM+TrhjCXvkF3l*Z4tSGH?#E3 zhXa~|4ltzLh7xK`Pxv9Z43%{qux9dn_P5E2HD9U2=i9*0}q&hH{S0Av3m$K06a3@&E zQt@ARK!l-()UlD7k^ZhtR|CAu6Oy zWALz14nC#d#6~S8!l|rEl*VxwmQUpZf|Zi z4SsE@Vg-#%1%wy;Ao4P{0IR7=HAeJPFAWhFfSyRwSj_Fly{YKr31hGSxsiL2vYN5& zHRC97fdzRzL~P!afV5AmAnrZ?)mhfBe^P_-^XKpLtC z5R^-cgLS>J^cj!k(@N|6S0gu19+F{%pn{spXtbX}58Cu#oTE}u-HU!D$hixKX&!}Q zjNAt)x_L~pUKta|G}^D;n5hixg24uL+()xt*XPhGCe92J2#SOz-R+pL8m4f@O}W4k zc!6j^*GkypSR(SkSl4Py4T#sVlm}`J5U1(T9x%c^Bv^T5)t5um&678ag_wb;QRM}x zZsvx1MlbuIg~Xm=B=W5fr6#OI zi+G`10Hf&Z1dYb7VK9n~tnJnyK)xvaf^Aq{kU>e>A|VK$;oe*)TQr<$vdpIA7$#1M z&c8g;purY`$VKR9h^@l`Fwf1w zMOUfQs!dm*dZe1pXcc4brXe5F^f6EOs#%@?nLEQrKip_frN(iJX-(%2faBY}u-nk{ zZ0qn(gnlxjnJ;pR2bRTn+?^h-s+jP4FohX=HWLg(kDF=@FXdn)$SOcQg4dfsMtC&L zRZ}G4@&N1Ic#w>0x?Z!{2DUj1$b3VTYPOt@rZ_eJ!JPx|3>%zb%*RcxjL+_n#!T2H zLh`AsqlR0NfsskWOkNHVFw>NdG2uE}L>zr;1jdzBTj2@0mvFjokD$q(xP{54?Z(=w z^%cz8OU+>)=Emf<>YD?q4V*aDT=s*ZU@ncWkvf&=6CrinG>G=iAq)S|$=iTv1PSf} z6q%I4H2kV7waMmm!{940QG91+8Gj%H=fBmO8ao>xs$1JXnsAo~Aj$AdRo!s^7rzY) zCU}8_7mbNONfO`e-^%ns^!FOZc?8JCIjyBHy}Hy5eT|)ThbeyZvM^XNUH2yjuPxOU zYWXpO`#TiUQX!8_6FJB2pU7P`IALW*X8YQR9H8Z|%u!jFK%VDl5{i2NLeEZp*atXF zzpJ!+^I-KOe#S&SApXonr5sQAo4GcsLt2Zw0^$hIaKOhlT`oW{7|Ppih`ocXJ<1>9 zO;h4_(bMF|eeP6CTkI!qM$Z<%BFQ4#KBX^0YVXXb)!6-<5pGaH@F8HC1(ciVa~Z8{ zfmP210^QrE6olFQLCqgv)evix?L?a56W1$6+h(kN*}^3bD0O*OjqFTrdWNk-exrp< zxIC=G_1#2dMffw+GIq5EpN+$CF&(&G3O{C*{C^SRoy=m2SkCcT}B(4 zm9KLOSuef8D>;q)jrd~uF?`nnNMfIBG?uDza>ln|e4HaooPR7$o+=)@KyWQvGqdU| zF4)n8CJaxa&m=k^4K^7MW{O-{)bIpX72%7sE?veo>>QZ~ow_?>dsJct)eH z13HLZm5Z6x#*jXx3H~}md_XfAfgQp$DY!Aft^2sH9?;FSZUu+vJ^iuc3&*X9` z6jaUtj)r$I_&Yyf=WQ6x{3eWpg)^Y5GV2Zc&yE1 zH_-K*_e(sl<%mKKrbm3vuHf)^{z*fbS@kkW&*~A~U9h%`TyE(S5-{@fYk7B4K};3{ zRz>xfD+HO4=>6W+mQFa+?faRKu24A_jL>!BEcOeYq5z_%?*Q!2h8isu?LUy)%N5|K*Uu5t}gfbYjGAR$voCK(2i3YasQ znSdIQs;w1&^#6EvUWcn6ZsyZd*QB`-beQ=sDwj}BI6;$SQ96kk0Wa^*b7k|!vL*1D z^o&4t)zw>-_QHW@1OnibB5Cg*LErm}s4m`R14Gi`-J5U3U{_G#ITGRIvC^nAR_V#X zz3-c%q2S($nc>F@-453_KbabSj0d9A9bo$A#I%!n!|NtJQ!wu~iN^DU!&V8*yem4` zUf1bwTu?K6upGN+HSQNq>4xE)d;dywPAe~D8i#p6?PeW&^iz&x4oXW~I{jKFZk&h| z=gU!}joh3|_kow#pV?sr&)2=NKEgw$sOH%3qPoEfnT&759m9{G=Q^37&+c%vIKV|f z7mMb#QxLrXX(gg`it-Zs(YP=R6y)+c1@FXv+bTqiKgY%K(3% zZxPmnooV>$L*=*U0=}s!w%3Xw#j(?HBj&BP!n{$YUi?ZG)RCC|hsYnivreDO#8Ni# z2gaUnk(MFKWboJxBJg}CfWv850zfGe{~RO~Ypv3&*%Np)cONxAqX_1R$Hdql%h4Ia z)q}NN|E|U9BV;WB1K2h;mXB#Clk!|F3FIp&MWxfr)^fGgEwWaTx*;{Qk^|e;W$S_@CC&NT8lROTwmN@C_r*c+Esx| zIOO|?8!|QFc5|*=a8ce+cQ{?NvbI0xoZrR%VkLXfz1%ime2?Ag?(FLqe?2@yYNVV8 zT7r75d_2z*JIktTZQZ45-o;s>4m{FQt9w&z!jq>)X+a=o9rx5=Ns^7dlh^E8j1Oj` zF5MIGk)esjPXX$|>%?^134r1ry082`hT~__z=syh_@M)aAQ7++PB$^he5|<3An7p# zHsQDxs38-`+{Sim+u+Bt!^C@;pzK)ecIlKwLA*}jV%^W6ANb;<++Jz306aZw*kq_* z)~dBCjXrsC`nSohXikY%++Dj7$T?!WEjTnjay@WP-xvJpIr}nl5HNLn?3Nmq%eaGl zVDHImi=Ar6=#3_gR%UjZq6s}}cPJJYI*QuNRxEXt@mKN2i|{;a%-@U=9w!z2AH<4@ zCcnPeg3&H)6IY$zlqjX60+!n_tAnIgW}jw*P6PqsTyiQl?p6i6rzEqdU6Wq^>BLX3 zd%@9d8*Kc)3Yg{&V?ggOp-06l6XOd>u&*RF_}(CFuPbrao!#FOa)T^?yV|Y4$@KjH zcAs|!?>i?!79^3CO;$HSNLNS zXzlTqk)J0cXa%u1j=2~P;PI!XXr`Akpx+4G9*8HHR|iI9eE{`d>yJxpm+UT(1R{NJ zDhc><6>;m~>BPmD>yvC9&yhY%{o}}oh0NdH~AW(Yg%rySBqxq6;3 zT~;_}{^mM=I{_?$)@(i{4cXUa4HGW$Sdu+Sh9=Ryt;Gv|eG2<^v?0 zt}6%j-qT&r92XvzgAd!}Srux78K;xlm!HO*g7D{j9%u0M0<3+Y!Y?6?EAswDpuKM) zN}pKZK{{3#p#ZlyIGNaC7P!PTQ!!5yGPJkS?{BE^(LRxV_!6nh2LL_)6>XSiAWE5p!0s=R8 zMu|eI#4shffgx`d`0D>2yyO3!oaBpH3fg#WfEPJE%PqqQpoO|Z*b#X`mm_plQmXih z)L9&?b-GpcL;Hia+NvIF-1l!n7lQVoZ5uba{%0bgTo%>f-20l>YLkxr-3;Dm)GBa@7N z<^q6B_UMdT!$^Y-&y~_a`{h{}om!Z!x=-<>4dIV0h_Bx<+-HHrB^nJai1w#9L5NZq zKw@P7sv&oPWv(FKC4pfp^a#1;53UV3AvmA_11Or?!M3Z9yYqU^=_76Vx11fJRY_+2{5t z`UZ3I8zLq#1{DAG9ffU&)ZqKA(vt!qFFelN$Wl_cqOy9p1!oBse_fdjPa};-MM`mf z89m}27SS#2S{i}NNe|?Q|NIHacPjBhw{R5-yZuu&Oe+|dKHyLcK}FuFx&rYRQV0;R zW~(1mQ8Pg4K6yT^g0?V8AraOi*Yh{CIdKZj8r*;z#jtxx3Yrt3M^Off67~)O5;l)D z+9~9Cig_iMBBu2t^lL(K$VFYsA!djvD_+xm$1f^l5-zct;RT@*d{6n*>$NXNbcC|P z1azm>H(@qWi9R*X>(DggiPxpYfdn`g%7Tv0G51Othp1tfcB2KBo+68}nP`2w{F@_F z#gO@kw?8~u%zF03!(r&ot8BhCKBzUg9s3k)1X1#BRCITZY*zZ03|&jL798ZOY`LJz zC}&-IG!ou|{Y`i+;2`r)Qq^_}wxpuI=g2>CgkAZ~D3&ifQ&nNJ8)djXbaFw!laRy57jf@Oc_8hQR;+)Y875_{{rF&ba z;JIVhK{RmR1nST~CWcp*R9=R%Y<)3UT34~LgTJY+IlJnC+O;W>*hwX`uz$6Di(}Hu zlrv5{)j{?VPQ{_h#}WEsFR}rxoG5-M@Xzx|<@MM+8iAQ*wbm1*aJr{mVyd$dA$~#} zw{VrZQ`7u4PUVi)VJEhWf^|j}>#cNtZ0YYm5N~ovG}A4+;$}RDaFO&Rj^F{mOio5o z1!H$9*fn#`@dZ)x;5Bt|o^@NCqniYpm%Id{LF;YuODUn{ z32ku*8s`+`N>?H{WVnwOygJ^6Z{Bx;(um%e73y}f_VLedU*9q@n;s4VMx8ttsqQUqh(h%Bqo$JP)`Sns8Lp-n<&22!Y#ju`4x zSu+`A)~_a{g2di4gY>@b{(XH{DVPMfCyInpGBs~fzHMin?s<($q=iTMBh-E^$Cg8w zx~S}q)j}~gfF${FtWG?gG(mWS42{Kuh?tfg-9PT+f*<>P)J4GX%Ey1cig0&((J(qV z%30qAwG+>cDw&g)2#S@Lf!noR1Y)6fx9IBH34uih}S7?$dp zG$Hj}3g#t_=L|rsc4d z*GZYrrGTH7dW#dcx@l5R?3D7iLbvlYTT6OvVl4*uCLg`oIGd6Pv8$}H-3O%8HaT^7 z3lXuwWFQk(@ov|OuO(^NE5pjFvDJk$lajvH%1yBV4FrI!Ax5kT+6^cuXQ(vUVuV=o z_dDGX&Scxt>GjbYE}+8C+HhiDvz6AJ%|MHO%H28bvs_H|>K~@J8sn_vl^Nno@`21C z3&0NeRTJK=$1hdg$RZZD`nST&h#=etV+v7(;q+UP<4zHA-iOltCq^X#Uf-t+Tf-~k z9}E_@k12qT{!YWhypmEXWDHw(D+I%${@LwC`;#J>n5d8jHI{1!RpZYm4&5ywwkmm8 zFQ&tTA+@VnjMbk(Y}V~T9`zt4a*oX5ofUhrNpVRo_@4Wn@b02`7QVbvy;orEX`0oRzRJ(^c?SMgFlnT zoe_0`!F(jJx33$X?P56HK|kWzC1u3eEVdt~A5EqF_vyds99KJNJCvOmNr#eC{OwGX zKmia8)UYkxsgYC=xATH(fLTyRtAndeVr4)(`0gG1(+s7#EjFD=^j zZzq85fKjva{u7z!dDSMUmcXH&vXPi49Y?lA)_Mff12dc~nG>c)p?x2oE77`-kq@BF zb&B$sUxVDD)2Z~n=MSch^WhEQW^CQ21)}U7J zh&p#d3s+OyVF_MVM&13G5=jkfU8UZwcPB)Ok#PQYK%gpm$d+~!+7n{u9DiXfH~$Ql z?pwdG{{3!-AtG!+2(QDdLXJc7s_2uU@2V31?qj{a%1W^r2N*sd5Ldn*GBL`#yZn5-&HMr} zsbRpunAw}UxH_8|+5N9_Ft&zaO)iu`1!ZOZU+AR;OeAe_q4d0JoR3E(SO)R>HIGes z4x-*PR3oD9xCkXCgfK*@x;S;7UQwcE0tu&Wr=vgxXgYhnvuG{cwnV!V0wUjDZ@wAp z1%^-RV5<3MJzmH3kpICThbsTMzRlkGH;}E+HdjpW4?#K;;ycm5ytydsa@otFHyD1YVMK7r?-!O9=N2DH2`6H8`!<#c;5@^8B;;^*d^3b);`@-G5k zdJ{nGe_yHh+wEj?&Nlg??4IrL(1S zzGUQ+Xd-yiblZxzwo{e)kbQC7G;%m0GFC(o$B7Dp7Y^F=*1Cf57?@G^EOQ$j$;&S6 z%lm}XQ#mV;fLEWNjyh|vq2s1*foo6#>?}Y@!_Iw~`~l5m=hHi3Ql5fB)OSQ>Y$EuO z-K+Ap@04|iXO|dxZNJL+TzLLled2@#tL{WE$r1^ch0pG)fhd!7oHXLeHzvz>Ls?_! zK(R^O75FjG_UXDBy^ea9=1)N=9Rx|5WyUBVx`P;8C9FTI)}EJDW~5toT{$)Z*bf;u zg62M`Yb5FqWB}zV@-zzI+iF|(PLmKVVRjiX`>$e!0t(=g{id(B>?RSYe@J?Qb{O4H zq#hFz-{1%ZF}%kE?LS2ne!$-mhX0UgU)eH;jZiemf+b=#_!}Xxyzp2m{<)6VwMjXa z7|4&qQm8!o6(+kiab^ACYr5wLD77kAgHd1C-fpSKuZ9nBLEJi=c($FB){~i$L)s9j z-xFa-a6uXXUb;ZX8bPG^fQOYc&}_j*%TZ`G!A`6)Y=GpGEU>pprza&bWFSzf(N5H1 z_>au>ULR2sg%}Y=eFlMiMkx6yZWDxvJkFIN?xmt7Lw% z=7;e^aX_X9^E?%^eiiEhoI_O&kVMiw^9pKvM)mL$HzT>GlYBjp#5KvG?WhKy@meI* z#Z$x8RIgd?#<^sr@Et7d263(8Ul@FWBxA@Bsa6|#5#yM`AqYw$DcJchZh4y3IGmG~ z-z8JTH7diMdQi39THjvAuX*(3F69k zSLpMyzlz$1oygSc%x3ac{cSZDlmmSLzt^9~vsZ(57|konPOq5w;JkV*|B%Ul=aG7kL*oHp6WnU{*uhUDF^6hBuI4-5K=GDnrX17b%Bj?6=b(dj4H}2@&|L zj{_zcYsu(_rb>>^oZ8bsYm1>G|<^j(!X*!)J|K>4zLd z50a`N&b`}4)c|VFVY>7MU$56R94sm#x*MzQaYS$4pVZOhyn2X}_ID874yr(kro#(4 zR4r_PDfDsXpt!oFvuUi!|IL*D8-83+H^`GP&eKFIaW80`~Z+Buh1WJBI;R= zVQhWED`A7!U1OsY20ld6MwlWMPm0)FzFdEWti_%<0xk6*GdEcTFM_iBC?{tt#2u@EwrpSuALNHh$VUm&X&;>&r%`c9L#1bCbS^Cd z*7zdNRBgN=h#w$=SqPZ|lPqC*?FZB&+7^js)K;4^Ln-(sDui9e1rQ0lz=XUAGNB{W zOW65>f&`L%qN#l^_}YmkdW?<3%nB@!dQI%RtYLC_tot77s6k&fYA}Ao6PaK@^yI*z z(fRQ;fp!YJiKUT;%=luU!M&&=Dq+N&5@pt*R3M_2naRvfd`(1~lVJl_)4)!zbOc7A zQ2HgEk|a5#00t>vhO*=$OGq2{Jx$HM{4+VjR0GmRl&o1;f((riiOJ9d5uziuSEbJZYQPY`@ zh~}en?mabb%YdrPihJU`@Kd;$;N@R^!q{LsPrw0CAPDITYn`S^ z;Z|&=7TWhq;GGzmPfbH?Ev}F3t9CGCQ)OjDM(0=Tfs1*w?Sj_-*R{YdvTwRbcN*K; z0fqspr6Ma~l;^NliYm$Eh{UpUxyjs!wh|(!HFm-@GzS}E7r@bCk<8FGMuB`Hhehit za?-GNEGZE=KP%f{%soKFJ`1+K-_|}4c2#0l+`uzlAx9-@m<&2^mT6(1R*sn^_evcb zbWprwIA78x5EoS|7o6m$1c)3tGtgr%Ki8R*aRiMVEHtOpMivW# z8)n+voT_w!?B+>0vt4gdgb|50l8=hOTc?Q4Qer5&>(!mTVFnO);dlbHuEoD5YYb!T z)cQ?ep17dlh#V?D4D(|nA@mU7sFUWm)VAe)Ee_mX@ch{($p0zE!nw---jK>`UwQoI zoACi^5z|Sln-WHR-I*W*KJ`M$C7Y3a&!LYU{KItiZSju=VQOsAs87a> z74y$S0VS1~M0vUn6KwN{y@_0Hv^6J+OnLY8%DQ}!3RupPkYssYMPrCzTp!u5e_k5Y zh{O8687ul>E#=sE^>VVPt_YjVH6(vue;*x?aH3^sB?+Lr!v-FFaS)f??{kJj2i&=N zSlq*T!s(h}QJVmy2Qxr}&=~07%htSYXawPn?&?I%QBiu;-YOUvM!78GIGWS^oTZ|; zaWdv5J1-i>iqW#FDpCS_>H610qJ-TnoL)=>-i|vI55rizk-%#4|0OX!1yVOi@Y2+$C&{TH;iQ7QZX@)a9ba}_N! zIAE;g0qDL~KZ9(~%nQvtH9`vhOqTR%1>tu3(E+HHQ!q|m)64@lN1RcrDv@#DP2qVs zv@3fZ!sz7EcFZ1*#&BrPJ_)-~8Rt_xK5%%qa5CC`^o9lxXd-FE_4mhm`3mL8RBYqtE9&^%U zKt)`SH@;y(gQ2W0q&Lc(2$Y$wwC~8FH(|=0dPo4@(sYP~^hb%OC`u}}I#*{qiS|nh z_oB9hp+d;=qId+U`)&uFvje(SeDT_GN!n5XTTbxl!E00(AtJ>X%?@F$Kt3c*1GvV4 zh#kAEaPX1Qy__AorfQF~GR+vec(e&HJHilE|Cn=i;{0o?^w3ET6F25-BXx6eEL8Q^mMi zX!7`CIyFklSF_PAsq)d_)*t|*ZdHy646>%SBXKS;VVWawrM5dv>>8jqHr&qvrF1Fe z7&K{v&&d@L6D;6E*++0j;_Ybdi_@#8`Px$pcyxgpurqGM)568@Q_ISt0@BRLF^>9$ zFXdIio_latyE&3$un}-Fs0PE-Ia16199xr@QOGJMlV@!m(%Hc2Nz9?KmY9|AoaQ#J4{Pke*T zuNRiy~0&-EtFzZd>oxnwfw;!wF@Qik56|RR>8bnfpB0vSm!pIEM;}8Y)BDEqgC+H zww(~Y-cV_)&Mv40OZ=iivu%s-$%T(4Utsp)Z{1m^w*LqUL5U z&^th|LR)MKGm05Q% zvQZ}Jw(Wr@VSW{xrq~9o3<_BbQNge@4UvUGUIP=qM2)gH;Ho1D4^IQzr;sQkmVGuL zRDm64AG5yn7D%l%Xf-voA8AXVzQf>V0Y|-Eq}C}ZZ)@1OCawQ(_zad1&;e1{dXS%% zY7EwBDX#s~BNP@M3X{lWd;EiS0ho~)3oxR8*BI>4S_u$T{=*`)$CB0@lU@k#o{`nUuQ*6L6S>uAk_J<3L4Qd1{ zkiLlxTlTGswJumHJQUc2e*$I7bnF?(T7@n9UoH+MV7rpv(-nwu@)^P&Oac1q<<#NW zUC0ux-oZXgLw6kO5ZFxEZ7n@C-ahAUw`_XNC!6GeWe>MH*XNPPBFQhfJuF}t#WtsN zN)fU)wq-pE#)*P_0d}ECrZoZy@e?>RtVOal(WIu4YZknVAT4exXM-{1a*@|LQc-ps zXMd9jV1>|gHc^D{xi2BFX2*yoX4+sjMJ7>uiddG-BNJcj~Xjwez5IKWzX zC##RiZ-@=vXEIDnij&BLw=MtrxfQ!2j^pbXz_I`woPINM`7r6g4oM?^d6xZGCvD_~ zz{x?dssQ*{t=T%5%Tkc?Ilt3w&anpskr{;=0G$~C6Le)bH@t(p-&mcX&iRq5>fZ!T zs7_aoi*@0cHqKF(>JMykMJJ7O)M&&yi*pQJ{&hSmdP$Qtgus}5n?}3UD+cg=Z$M>SzBn82vp}8P7cV*QWbHyPJIqE zh8sl^mPb)AVRbfxg=z>55iKLZ86~lih%HEI5PBE`K^bu4y`+8D(E-b$FOk$&`Qc`DIk*DTU&eG2}5hTQ7vI6(@(+3%XmC} z|u;# z_t$fWQoVhWxT_QP`#=@HTqOoJJIs9qxF+kx@dq9FDAq!w2{MucNAOWduMchzjI`{` zRApRJu1B-nj3XCn?i50ys%`NCCZl2pU%lK%H=WSKwTd*mHKc(Px_WkYNfMGJeyy*Q zhTD%J2)))UH)|YM=E!Q0qs&u|!G{Ii&Bg>|m{u!dH z5i5Eh7Qr-&7_NRG+|%`?(Pd-Oa`D6eoIjT%61V&A6Jh;6HvmYIl@aX$!C!|A;L23L zviWjEd_HVwAK{ZMwSol=fO_@cIB2Lw9NBW4It2AUFK;1l{YPV7$d}PCxX3?~hVj2G zQcr79WN%NXfRSVkDbT8DZxb(3Q>BA7WIE*2swAC|2(VZblEO+h-z%y<*z&Y?<(!k? zxljJ%c33zKYUPRn&h%pdB4RWAl51m04u5MzGu^by!N+pUZBb$a44Z9FN1Y!b#t{H} zsrGz_s!8PTah{mYAw^oJf`g&w*6ex5EoH?(*p$301q)^b2Elkt(o;Yoir%jGhbQ4+ z3Odl0>`&VJDttArav?}UR1z+B`E6rl(5lR1I@jp0MdSkxnAb4CL+OM_)<=sd@4`Z~ zN$zEQpra7OpaY~ty(FHUa1nb@Iq3}GpYktSzU^f7nKCo3iSL7o8WEHz{ODy*s0COJ z#lh?OX`h^Jp4DwlM&`kzWp!o=D@*4F&qm824z&IVBKLazuqNr3+^2voJJ{;ie8yS-N6`u~O;-#A`j81N95T z&z{k*j;ho^|2C)7b;=WyRU*>nlo=oGo)$^CyVM>hn2p#iQJ(x{7@4 z<6h-?dA?Fi55OE4l3ckZryoo@E8!&d|M}Ya1r9^x8u-8aFFOYj6On^4k~It;9}J_M znZ1RpB@qkv|IgTI$^P%pSnsFC9h7?H`oQ%5Nflm=TgmzZPgEB7=neYSD)Uc}6n1DK$}9>C>MB1#@D)C+HF=KF{uHJc5k zQ0d~dck_$C!!Emtz-NnN3$Xp``A8wd;gvzjGBBY`1t7Bp!|g9(N?f3rIAjVemDK#u zS(lWpdM(Z z==s>%sKmJ)P*fI#_}M&8Ff#CnK8*!=U~7}`*?O?RzFk4GB`d~{p5h`gAZ537Lptfi9;H1Wi zSlFDm2b)<4@4-~sDM_A0HQ+id9pw9Rwhra8b_2P-6cozX1U3#DD{S;Fa0Q2&!F;V% zG!ER$m$=^CvZebTlJdf$6%AJ60c^tt3PqK-+D} zhL&zCM#B8*FFGSXkU?jo_wBn($aMHUTU+zJmtOitytTha;r z_bJ3+Ry86#crwKD>&c14ZS|&OM+hGru$T`yv>5kCP4O|#CotTRIdVeVOgI5mohY@* z5uvB{u1f>tQo=}|q(M0lkU!>S0BI2MIXt(cJjpi3Em+E(|4tJTbU?4n(_# zcyUu?vFh$nc z=w8kwICDmvdK-m>{_;AI~YlFJe|n3x#@5Tr;b1N%jc%ok=$noZiK8(!&R=Ao2QSJ*|hjH$nlY-l7h z`q3Q`ilz)Q7va}qebdecat_5MxxEe`_@s&@X`hxvVn>-B@o&msm-Ochi z+`{<(3Y47PpofV8D!57vzgADsHlES+(v)j6Nq`PHV_q=*j%Wk{2jeiTzO$7Scg)iq z;2QcX7OoSyjX*=gj>KGsJm}2LB`pRyy$V~JK}U!--m<{P73)`I1`FDFWh~u)w9~8` zm9&4jOT|UZID6s_`L+gHa^u8lJht@EObHN8S+`cP$$;YDe^fh`WAu%6x51|0h>r*G zEun|(Evkd0>qWf*y(Ds9v&KMkhyC?;vIn4adcHpKCi*{mQdfd-CT@9#Puty$o3jR{ z)a%#|gtQ`v30wv}{Sysv2g|7jSE?W+5$CslI`KKr>Nz|J>|j6#LylLz&xUcU+!e2< zi1aiBF~(e|2maY*!e?b{J!c*x>244b$nW2+KiUSfEMo(}4gqk&UFZ5ab zBosyI=5U?rljSXnGyqMBkO6~xDH=30AyZ#hMrb?;Gz5ygT$E$db%g4OJa3$Y+pc3d zeG*^_6%g@gv%t?yMxz2BCV>n!<{(TIe3I@_h+DC*XH@9v>x^!9l!WHr`Z!ifi<$ci$vs?hqdk3p1 z4TkFk0^W6TlSk?tv{2)Cf(hrBQNX!{cMSktWfR?c7NdCywfhLvFZjd17?FbnLDyI>V zJ0bxJ|1lR|Rd@HqQCcA0SjqtSu&A&BViW;BcX}MrEM1%wCU*$%dc4*<8`;Nl|Lf~2 z7T6=Md~i!^@1Ou1d$wg8cQB4(ic-K%Z-8-M2h#gO)>$i4rg)($YF>1asr(rddefWC z8PC)cLP{Dy4;)-3&OWLMk5Y*pq5Uj#a4FnaRWA6EPliouE+PIpJr z);(cNE$O4GIs8eQpE%Vh82odVC>}yq|B7|Y;ukmp3Dsnu>h&Uu=g3a-TM{ti=ShSg zO_@J?aKq2>R7ozrly`5k#m_AkS2)w*-#F@cfV-KX*f}#3bLfv+92YZVmgPMRuSb{( ztx^wvrq$J6iLpndaPRR54T8VbrEarIg$`s~=qG)5?w#7ZG?RJ#?Zwp++XVuxgqEZG z96~sRo}Z?S5|vUNOGq#gFbhx+BKwl%>LA#p%Otvo zZSGjeEIkRlDl7+HtG;CaWYF?bMQM&f{gt%54h5fU2_S*r1DwMV@Fy>~D{u;SrSlK# zq0FxHU)s?AR&MRxn-iX!j_^}IhEYlP^&u;Ew%QWxE`^z)jmYY{i7x>040EE#NjjAp zl7tg}gxqt}NsnXNkuv!28XM5`-xX8E65reS{+Aj?+*M^icPIRRLBuJrWNk*5p!ks| zE0HBc4O1#a!3q#&09JMFf8);NBcKFI4Q|u8X0m~STufY56X(sAaQKvo*M|$Nk?Dnz z{jPlh8_gVB9-q-e-RV^E!BVHNJAB07>mft6NYvV%S#|GLffdO+1+^88cCte?l@ zq&~IPg)IN9)&UZZ4fL?2D||$j2yjN!+Rv=9uxhFuQ8aRt8ozgroB<4`3nLtJ%#1Wu^S!K2)&+jiY@@5nK@Tyi3k+yH& zJVj9^5JkEAJfWl4AXGX$?W>Oc*x%}X{naC@)H#ZfQgi=MrFf9(*Jw+qREZSX*QXH5 zuDQ8b<`3-MHF2E#!;Xy`(%)iHrpN~PQcywXIpcX){Za=bi~T3GgD+_ic^C86GLM(x;PlO4mTsdXq%EAY*tH!+; zktvHN+P__oS=M=VPV1)|G}KmU`59H!W<;UAbFq|v(_mojP+QLoLi%~`i!xi?v@xb~ zq*Z(4MsNc7ID9ult>j?Xjk-X*{SikjeJCnFz79g{+8el13N~zF@lK$~C&~PhWlgIA zX6!8lhSOkP?8hQiGmt%NpaTXV9-+)pRP+p|hzKkEs@Ol*%(_XI%9E#mxt?ao(Th_b zjtKv|9(5Y2D%g81kbi7xP#=}^-S39ZY4r0R>A^J_obu+BdW!bLfwLLvY9C7=7rQdg zrbXtK9+N~AYuu2_9J{S+DlKeH*}zf|f!}YaXrth_*%kU!4EN_6KI;Gg?A^bpg@x-a z438j|R^OQ?7XHkW?!!Mrj2w7=y{?044*YRZZ{SXMHVYnyTQw0X3}yte1MI&3OWW-u z%(#7chIpERUrrU+yrlo5hDBc(mw#a|-ljibaRG-5fo>+=)gGMrqlMW{wh_0wp7hNW ztTQ82%UE8ZFaEPDUJ(+Ij$)V)hrl)}-BT)R#ki0GCQ6UMR>D)&@zbcw_IX0&CSP`f z+deUTjTFDAX(8r>E*_ykxOskKApvcv&=y+4&mzB{$LI)gxyk~gR_?(EssaPFjA(A% zOs-VcBMm*FnR9bWZ6u;dx4^=STG&k|2G&5Gt^4b+ zJO#ivn_3UTGj$Gdl`B0pIuR=6(_{OPCPa^ z2(*x%Fd%WpK}_o_6f%-N8Rs}N7ODFe62S{(jD`w}Ph16%CK#TfB{KZdqV0_3l_9s` zztdCJ?n>2n3}0nn%r6<}ZLOMS(>l-v)v5Uh!WVI@@yfgT{Ntc6DlUQM;|=)!^>315 zjR;!B_%8wG$w30HBjE^ON|HkY4IEv?J@FuB zA33i#rC!iEs!+Vh8zYV|urBbgTj1}!I7e=Q;>AVgdq|&V+dGNlKZY8NQE4qXN0W+t zXfqw4Za3K!Je3go<&Fm$%A+-4EAZ(9<%IG8a$XCA0n3RNb9#5Ws~Jn---XR5U31Qe*ODK;9>cjwugsX9p?>hms4Cp?0ZoEHBZai~< z<@Qd;$>4-fwB})J-comYqAzY884|iY<-pMZbM&RK^dd0K`8|0Y;1>K(xWkEopN-l^ zzERKYP$KCx%6nOnUcHSCH1skop+6G9`11_l=|QCkJ=+>Hz-*kn?PO<@+ZJhBZJWV? z*)?m(lavsTLmiV1(4b`|3@bfu8!fs3^OhhFPRCE7zqdwj*AG2lNzscQ;OQthBGXLb zV9mY6acd*5{+}Q!=VrtaqJ&lJmhiJbP9G^)2-o$pnqb<`2%xK&TM$DiT0;>%UGDZs z>_n}7r+`+C9H}akY7+Rb4^@KxATc;ZXNW`>Wg}#!jP?Mj(#tweoB`NJg(+x2;3qf+ zOI=dM+e}m?J#^B8sY*V?j7s zBq@}_TeUE4krW2Wxn&){vhLqu635C(MQ2*`E)ILgC#e)Yjhq)y(484RCm7dusy~phZ zto1{%bjUBXtsbfly* zp6kfogQiDIp90Q%DLYH>C5iy?%n6#9c|N@#GiP$k+;wtBh{~Um793HTapD6T2Bq`@ z{5JV{BDAh_aSQ=K2#0)jc0Qm-TDL6=%6_Io@4}*AZrh0Q#+~WUOOpoRTuDj@jGQjU zx2RU@Pl8`7ytK)|nr}eR!SB$|FH4qvLXs$9;SiptY>oM>Y?1H9txwjjn$}^urgV(8 zUi+nCNCAx-4`gX@57Sn?>sGR3IqX6w;o)|=QF&ELzlsh(+!Cq3VuI7LHMDNO>8+h) z#iW^bOeb^j%J&?@BkY*>N6KZ@`YIzYXH4-vx%uXAPQb&kV~1RNk5}HpRv6;@6f|#G zBT*oI2mOLi{^2Llaw~qkf%L&d*|RxckLNqSG$An%XmS+GEJoU~eM=jk(xg3v*8mOO zu*+WU44EvT+!P_rI7K$sW$^DpYPN*1&7}$D=`i=h+xeZDZsoD`svv9^#h-#kTQ(I4 z-dL@%N^{cCr1CK$UhyjUBi#{_E@9`DK&3ISphsjfj){5iBj=p?%~8m2oukq|Za+@e zSPdG9t)+i~0f@yq9DAP1Tv{b#ioI!$Kh@>2H{xZ07-&D7H1IvsfvYYsLcuA1CR2p`#bfwuM734IMrAhQFT3~$%r2b>^K!g|5DMDc;?x@)o z57Gm^h<2$de5|GOP8@PA9W(O&Pvel*g-_!n5#W5A2ts_2$+(0@orbjco15hyV6~V= zK^Rkt|37KK!SUbE3QU|F|3#0>*3^#O>_GP8?;W5QZSK<&N1;@_&~CBZSPUZ)Xr(~d zAez%^qQ%kG(QO#-z1^Zvjy!~aH;O`|AKaf}-;FNVw)}Pd{i#gyBR|Xk^+`DHE~;T7 z+2YN2Bh$d_h$c31l~cLh8;FP~ijj1v!gr-w69f>KVt?M}p9Orq31f z-5!v@MzM6(EDmtZo_Yn7CPBaQBTCt|n5ri$^ZO+4?6b7CcuPbYB;?c_#2>cjW=3-B z<2_5_X=v70UI=q){(jMtH^*bm7G0uw5{VI&!#&PlF{3^G_V{t&k{l4RC3_m|woO7% z#jtN#$CoUBd0}E;mRbQYmp-tBB6??!{>a1)G8={dj;npxwKQK(7=1{t~s^@JA zBT_+tC%dvc&AOzh7k=H^bOt4Vf1mB}4+Ug?#ii}}WKXKQ=#n5_z}(J(DaUkf2twcq zD*yfN*j!re(%Eix17RLh^*>b>y_*~|wrY*J;#)%9av!ASTP_4d-ys|>8S=OuX=0KW zj4gBaBPfP@W@+#>(W8zT&iHP!f0gH&o-NfP#7FT7@aLg$0$P7R+<&3Un|YoGlf9k_ z=d|e}n4Y!}Q;R#&HALE^egFA+lao=%cT2`iF+QsE!BTzJn3)#A0&2ge4Mu^w<(i|;+&Ld$af?OL$4S^a`?$~Ob7M;wb-wHHReB*yz*B=InU#JP<)sB zGml?;uz=0SiOta?P~D+vR6o3IWPip9niM`fH|FDt=E4gg)I!#QK)Gh-kPHENw045` zkXBM0jsIB>MQhjXX~S(AA4>tgs{pPM{a8sJ<% z*F(c5JJ5V<^RM%7-61IW73ti?uo&RAiKB!}So{L`h(c^l1@WE{Vk<>-I+NB1<3U$| zJvlz00FuLE=Lm#DSWPxGT3IpC*|9dWIdBf~z#QjbJC{M-2V@;68bCen|MJ#j@@Ffs z>155YbHklE&X|AuLtOj7&=^p4g615@dt+eh&ty(69OwvfKU|K*GJJgv@*(8d!PV}O zg319_|C%1_Aaz~>({&-AawGUoF7hKdWc$nzdFXQy=1knM-&l{62LIwLed76%a`2&+ ziK1CbM>#tuP*ExZn;wr)F#skTvmOHy9H_79OU7$2#YyFUr$U8n2SXv^Hn7H5tmZ^z6M;z`#cFobtlD-QJBa zcR&JvsB^s_6J}X|P@(7(L%kMFA|hTd&Y!g%en72<$& zT*>J^P9B%fAe;C^_HhxW0_wxhGOWe=$5#mLH5td|M3_FqaiY!2zeBrj2*cjq&VC30 z8^hR;Sbfb!EyjLJs`^G`{6%*oxJzDoH=8r&CHf%A7?Ir4%4~ zM)Cp9Wr}SlT{igkeeA!D-!u+LH#HzeZ#Z@}AwiTS3Mgwfh2+LN4b}CoRP6tg z>X`uf+z4LHLx@gVNa=u#@G@sc91NNF8*=rO19bq$tr*tIGmV+$Sahy7aqJ2J?(hp_ zQB?2jJ{kzzIA{nw)Si%YMkp}?5*k*7rBR@ZBGIy^V~v_902%%`D!BcF;iaqYHFKy> zQ@1KXVUPfD;h;_Xn2eFTvk~2TB6md$r_c&x0F-=emw(NRSux4b5`!im#GE!{SEWGeckv z^>xn>h~Rm_MQ}-jF0g;IKb2dPx}PYn4xvE;Wv199>a8%<;zMEIBTr-h0E3RpnGz9# z^U)No!5=24yzpT*SfKN(Sl~NAQEP*$C!*3{p~hwzLLr0KFqzLK_%j1Ukgai37_y(1C zHl#qv@&4I-BLBmR6k-A>)-os&^=)9~Jjmfb*wNHpPv#>ySO+z;S*dT;LPD@3F5G+t zmU-tU&2dSWK1d&Nj|zjuXI8`n_Fe9CIdIYCiB3c(jKaP4|Guox+lS=0h&(u8cZ6wJ z(XemXiM5X_k_kzlSxvk-Op+8N7qaEi{Q)Lx-F}&LH_`Z zT$Z}E>qof>yK2Tup(xlLN2adXrdp@ z;t{g;qvrw~4nU?dDRBQ2RVkPnYHQdJ2_xvnRjBt)yR;rm3yi}oOpi18h?TCC~5s}S|W>y+}M}Fh}PWy5-2h_#ICBG!~W?QsV zleZYR#8D9u6Puu9si3#o)~a6=3-6>1CFxvqf@cAgxlaCkh!{R=zGa@SfWkJ<-uEz#GkPCkL?F9EQ%f+?cw3G1jYOEFnkM#{twy zPUGpq1vbF{u8MB+=@i%E&hH-|T5L%2Q`z*rLN6R)#|NSGrsR#NM+W7X)qZ& zU`hcti@g#SNuHSwsbwrv&7FFha=Ox9F>#UP&{}^Tp)Y+F4Q}x&lb-N#m%}q5u8*~=){R-a z*RD}NQ>b%C0L2>Yu!0W7*Xy)Rq;q3uABTV@hM(e!_l6`-R$uNQ$I=+9+I(FpLJcJA z`^9!0Azz(Fza2xD38H|khqsTdR)#7kk3aP;!lRu_zg&e;q36a(+=BclI&NHXLa@Rw zsKyrFKUgGvlMNpO4RpJwS=sPkr&yyp^5GljielLKR*$RyZOAmS5bhI;(l-=6yL(y$&tS0s{;Zho+BboU*wmkE@@Pv?+=bh;RMhZwn|2T z*Y`Of%C*KYwK5Qx0QUcQdzd&G{}Tuuf;n!tCH7oXyB7~Hfo8p*k_kTwdWuAX3CtM{ z8D!Ot!&+$CvQ2E+kpE@R|kmY?Qccs##&P(MJIlIL3}7k{|^HY646t86QD0D_Wq0%j*h+tdED+6vwC( zQrIszq-?&C;&uTX?^if%iOD@F{|8cYq?RR>r7L>SMt2qwu_1hRwy#o&gb%I;&?(z`S|St7?$H8syYaI7EN*Bgki{isB=_ARelJf3smm0BM|^5% zeeYn>j9kBnnC&25vj82SEq=&XDz!fU3`qi;bB?wY&R`Z*0kY{RdJ4C)(h#GV^!oIV94K z$5VI{2)DSsrhluZNFF;864gAZcKB)tOo#vY!D1T$pR7hGlD{)xATkk^P!x*jrt5;F z96?ULug_ONkce^>^Y9q7RsfAiFOH`U;VX0h!RPs)5PpyV4j#VaRzBECEIvMd(UAQw zD9$77A5%U9o8b6Jl0+LsYCr~3Sv7|WQeA;?3?Y0tf5J1%Z=hYE%C^$=)kf`UQTmgR zjOTYJ%$T`?sUyYSJRq zVsU&OgMC>)O*j^W)Oa#$iup7WufM|Vo)|14COAJh1TYERQZ1CUZhmS*(fh54x9UV% zfBz9z&eRC<%(Zxh+|YGq4vTZ>3uD2#9))lSyc+w}mn#kdQqpP4K#w@OVa0@uJKn*YIv1;+2q{g97QaAyqElf`Q;GUz+ z)7OuxgOxFGz8_w;!AuiB~ zjq0ZHYOeuEuH~?tmfck!eM6GOt|UJuiH;EP z2@YuIUO9ch3av4Jo#BK|RqrZYcSP1E{e56b$P1nxxtl9b}0nwYgz)39~3B6jTP+ADRXfyD5OykyaTh)fy%Qj_^&j=S<|iQ8j&i zy1cw(gU}N2qe!U&Pe-e0CKQ{hxVy*Zav}R--uRW`{q5`*7E>~q`9%^Gd-FSi^f^zJ zCeLVsAWVrroLic$_}C$fki?6edhr#^aH-c{D5gDlyM83vN5&s^gX2L{QE~5S!1NVB z$o&yjg4p2?KJ+?2M>?h6r1=B1Oi0`p7Cdz!mGjV3uToV&WcFC;98)o8> zdHNM(rg}3mu7!c-vi;IvW28Xn8GJbR4=zE4q7JNwCaN@cMb$TGeF0Q2$Yo^t??yD7 z+QrsPomgWA3Im?4wv!=dKQvIAT|8j{~IpC;+)jp+b!N)QZWuFP~Bi{b}htUp}ujQt-p%uzVfWwpq0OAd$+~FY0TSXA^(uZw$lGVXcIh z65^haRtXT;A;mS+G^E+9t|WMEh^crc9}LE+dB7&^(A*Fsh|_cb&*EPIoT=C3X}!|w zvg)i8b{8znPRI}OVmjIAs02>euiW3G+hET1e~s54ZtF_)7VVc+z*x}*7pDR%g};i0 znuZ~A0yOLj^C+x^O6@Y)J7eu(0-D<#h>L2iuD^e>>jAH5120u&upqGe!(YyVQbN zg!a-}R_ftZ`*W_?oxG#KMH^;6lqAQ1MgIcjS0;Q8>+7le{Xm9d-YQl7cf$lD19K{J zJ2*q?r}IC~sI3V+FmsT%NvpRBumU3Re=W8CTM7YBMQjI$|38|Fk%O7(KQ-01hGy(0 z8&RPz?Tkz(G!oaHV~P!AGn&#H$6%3Ys;BN%R9}XtyISMPbRGM`!w!XuwRjcr3M?SOg~yOEl5cclcaGz z)D$gO>VnXF&~3d?QguL90H6zhf2`3oAK=08H#nGyY%RdH3mcY83h*0fwZ)aZrFc%s zlIQ~r7-V;OgDK);wh&07K@13CnziZ6b?A_b^rW4}5fKaP@jjeUE@WU}uxPo4v6S*g z8=lT3Gul%rEbB7xUP(}(j+^wQPX*oZOu27*;I|OiGxViWUP*(+;=-1aV7)@4Mp?aL zP|N3|#y7x9NSXXv%nKs^Ku};l%&OxVchoKaz_WpJmv+}`{4m_>gj)nx0m-`LYH)-a zX}`UzPkG0J8w2eC+79%$6Rl+Dwpk218h=0tJw_ib&F6u?1#DLt*sG*+25~iJ9#9W8 zelM;w2c0bzvls*!BUH-*Tib-tL=l!6&Sef}S{42f?Qm~>m^FDi+$rr}8j*>r~}_jw7KVMV=X$`TnuUL!d&?z&m$8bg^? z!U`r03Mg0tmyR`GyOxc_ZE40DS~IknWZIgepz@bA&7Z(J5Xm z@WUZ`%66R3_?X7tr-iKV83BU_$M_n6aE5MtyKZ~~bT%|%kj3##5W%Iv@TJN!z`tcW zzYY3{0Ke3?+R=vy`+`q1PhbFQ?SZpudJ6!LT7z2PMC_RVA;7Z@c{1M+tznrX5wZMb zy(@FnWeHi~Z+aFRfbY3mP6u;DYrI)jiGO%-0om5}j!aiWMn?ynN5*smn)S9TJ+eGw z7D0N*9v~a43}k|G&ya-akJ!Rxn-*8wPa&->zOp4g5R+zw3PJY@hFkNpPD`u?{{f&W zLyBu>lr=CM^fwrYb2GqFyNYjFE%d^X=}v|lY^q;bBCW1(RhbmoOzw7us;9z%YeUJh z!{jR}j0VMnzrwmn@1HBty6kg^Rdy*SF$&8)dCi8)eloo+PQX-GxS>|-%MKy=!^_wR zD?X)4{63p9B{iZIt_tc#UsK@(q5%+fCA7UsOnJ9@;z4!iskT-X%Mb_7i2&X30NP+Y z%F942+4k?Z9sNlye!BL)S>Rdopv3?2xV~wh%xobzCZx3Qt-&$&&1@8ztZYaEN)n!X ze$5@d*b~nwZhJU9{&TBF{uMrYFSVQXV#>e!dVCE4kT9nHW-ijE+Xf`<6w&Hr zM;@$y!!~K|IAgq6&TaeUpw3TMh1)PHgX_t7W!Y&jhr@Zf76<_clARGg7*_;&D568F zFsZ5WTyoIL#mnE1WtZChiLrReOGKB%yix%6c}aD%vHRyGzJ^A>nxB*%I@~RG{w3!1 zq7hS)jDhi<($;r1lD_8bv?)18NPtVJG9-ZIK!+PXP))m?E|kGl$*Q9!iHR+?l4_AlBeho#~_-7L>h{t$;6Oa zIg)z56yw3I1S2J*03ix>go?j}6COwWK>~9*G^M5KS{(W(MKKLp%)EN_XJ49rN z6d>50CTW5oz-o*(xFP~cnAR5gzyJzyK))Pxd>h|p4S3jmE{0Vf=o4FYGeltD>oEbIjBM>B|cLmrp4 zM;@d{gDy>xWT=m1Q`m&I1z978Z4mf3*-6N=j!r1m-_^OFeq~xBxnCXFG;<jZv*_`#NSEE_7}rQZvbc& zIwAAmMh0u^LJON{qb1C=KD8FLwe>=lHhzCs6Kj81V3+Kmwd^%WGF*RXHts{%kv?r~ z%8~xUKhGjg+%C6n?p&j}2hVFEJ;2=gp|{tQAOG{_-{S)P%BcxF`)knK+k+8nX09>4 zjol{~!2SJX_<&_4WP;0gY#1 zauInwU*shyA)R`S7&i$mOxJ^djD$-xgo{G0wCv`K!p1ta{TBw+tq@jq8zAgmvGPsN z=K865ZL8_zlC=n?hQ2m)Woepq>g%@h_{yQD#F%AplJ`OUan`!y!N)pPtk=&`nC-wqJnAE7CsS91ixN zO$$+NW8FxZ?XY>k4-$noMzRSqd*b`es6IINknO6OLu}F(5Al203Gigsv*7_l{%ziS z>wXV8GMd=(*JeATq9a_8hUqNOZ9EidDhf#q_~P#18Vm1>zBCH(`Iwy@1NiZ2;nSu2 z%>Oww=>7WHUo4fsxte?*blOj;eBmcbu^yQwOZJ3aDFQ`CyX;r;&#=1buM!EjyXYq& z>y!G05h=YPY>44@4PecKjiTEj@*~D14<_4&!+an|!RL~)VSQnr2*@W{SH|KEb1W=< z;Aoa`Rl>II-_XI-$pT8;yOWy_GW2psX@gasf@OU&NPsDcT?ioz@sf^;HLzg)Yj-`c zgfj}w;FeGY3%k7US`0b3nLSyci-|G4QVk-!p}ay_;Je+s3oy)}VwdY-x;{EY60gv` zvaHx<0Tq^m3GFZMmSJ-gsL9GZG3~GeXw=Tm03LeIU##4|TgqE^Q*N2AljuAbj@Sb= zRm~Y)4b#G9s$mDYHRJ7UJgVJ>+h7;3g!sXSN!E9+xEcC^wHuxd>N74IvBATP$8<5V z!Ad6WHr+6;0r`nXiL($0o?ZDHsqv84W$Km3!|NyiFsPP}>ldkra~I;3>`{-6W^5lq zelK&qSwHSvnfP5^f_}Lxl~Hf3jMtqrNoKnv+q_jo^yZhyhxZ`ajRlw`9!IK0o?)UfzK803lO%@8gq9)#H=W)Q<$r91RXRVnb+zg7r~N z7s)F>oQ8gyjJmMl=G%Goy&vEn2!Yi9p%*ayPqq{m4$l8PZ^isaK$Q)pXSQ}OUfI-8 z0%VW4wFqZ)#?W1vyLDPEl86u*Ju>NEdg$}RHm~z7!_iY35`|z6=>7g_Rl%nX=U~tt zk&zeR>F4ntXA-8Aq?v)6GaDu{MClWWI8G=pQu0ZRQ9p$^+d2p+IIYCRi&Y|wWmZuESy@v`{_LKIj zuz%5LQ?wo)EPA|!k`^*2vkdKc&BqKP9`)w@k@l^H&ULR&D!oU8rb7GTx7A}x(O_s7 zOv3Vglne=_Cm20xVgHVeHm6Nq0(vcR;m&&0bw!F+HnQH;Vpx-gr@jxzgb2y|-Io8E zey?gPz}uljNNj;0?KSX^D?8ksOqc(9GQihQ!&dQJL&CMp9x)eRgw8cS#J+O%jxupg z!gV5BKG*Kyt_Sc6>PM3ajxYG&i2Nn?*F>sF(^0{aP=)Fq6aUK82zNZUZ$?Vo0gU0@ zp9@Qjeff~5gx^!TfO=wG)#T`6pTx_(f8|#O;A)$m!-@Y6P2on2^>Yi1{fuHpq=qHq zz7Try^caQrQ!n4$P`-|uC+rf(ac+afV~iy(37J0->8CinkUOi}X?Z$&1in_iTdn8b zbUA3Re^SW7Wz7fLSWKZp&2dQss;#Dk^K8uTjT&sS*&aKbQTH8lFb~3Z$VoAbCV5O6 zurwl-ARYwVUovo#DqdY!kqu`o!usI8rnTl~zaI=Qn;BqNcZlWy^3kjJZ@w6Za@rWd z>aK;`c5Aim6n9`tW%XCvrJFe}&FUlFMoaE7V_vpUTxN$*VOR6XuU*5fNjMMcWwdgN zzEkAe4Ls8`)6B>z?q+zl+NvlJU<|J-z;R@wjQJM0fGqJD>3r{pLkCTRIgTJ!&=`aJ z9jaRTiSgaeg&na5C>&cJleRjsmBcI=Kpa%;IbL;Wj0Xe! z1FH6Les!}Y0Ln^(i_75-bu-vFWnLTx#U4Y+O@oHKZV4NbNKqI-tc!W-#Xc7eIF}PI zOqZOp*y|2?m>LmCHZBDJ$@6D1bk@WP3``Q8XZjHAQSztz7n3F4s^|?02aY+N`pqIB zIA}hhaqI~ObI~;}hCTHX*zMj#L1H;!_$%nqt%PoEi8?}r>tW28()i3@H)%wza26dC z^8@11^MM;T(DZ=w(oFiUOUg?LU<&kT^0C1YI}g<=Am8+6Cf}l)Eu&PT+Jt#k$A&}g z)}=H-ayfB!VvZ7{vE^12yDVk53}SVYK>bR9l)Z^V83#I(6t-T&jErP|5?YFwzC(6E z@~gNA%`J>%DM-k*qOfqdwEv2SXMR%up&UWv@LW$5@zT)YXS;vQz(5BCK*{^g98XoI zLt3DKk|Hls9sz|jyPJH_?$4p#gk{TEd=f72VB~U=tx{nAGio@%1o%2+ceIz2&}Os5 z$DmMGOu%TDkcy~zC^y!g_E&@tXR_)q>;q1pw;aGg_szMIAf_8ze2=@t7Ajw4QF8eQ zC?^Bm%LNV=gBUhs()Q5-2&5RzA-TPHqd{Ei6KH=JC~E$F^z(z-qU2xzug(PTo*}iY z5v771P=NC~tk?)+_~?nILl|_ZBp)kI0Uc55z+RI>8yu^w1ejI z#lrPMINJDKm)?qGS{3^RBCRik=QJsoKI%EN9Uze*8IN)F8y%Sk@R(fhEk|OFSym@^ zzP{v-n?vx~1brx+5J@uIs+bwRjY+_BRYIEgn?f627_cr&4}vY?|ta6n=3c4!wH^s zEcR`9En_5X_IV*wl@dRr7tPTqtwxebnILQSJQ3I94Ay}%*D zN)ns;?-39JldomDT3Rhug2U63THIFiw-s{R$jp4UG$DV=0aTHwyvHcRsmai#C}uXlHC zQ5OymU1%L7*7m~EH`(UsT6yOwaT!jfSj<|%0<6>=+g zJJm=YI(16s&_*`{6q$C$!K9~%i$GnAQjajsdc^s8Fya8|8!MwTK{;R)#Z*IQ%3VHz z-Y^g^z%g8(AjIIsIA>#A&6+jkmSbNJj;%laqF$!@3DGLGS)6nSup|B2!>i9_37dX+cR z_Gd+{ni@4)*g{H)3|O*hk5%xmJV6c+-6bL4;qLv0VWf?uGZ=e86(V8EMlx?&^Wk)K z2c)XBfxcBx5ABMz-7p9FPhlEG{-Fe=C?-v0qU3=B<5_iS11=3Zj>sPrB%=rDq1F>^ zfb*_@8}lh8QWL?TfD~eSe`{CJn|YZnRZk0yzQ_Cs+{(R255$uKFxzB4&4)ctGtoO{t2-wr~hkWqKoTm0)m1q+ZvZO2O2d#oSxDg> zfVfIp%_LszZ!CCmw#zs7z)W=d&(z+B)K-6vR-^gGXX?<*x)Tv_lS@5KOfc(bW?iva zT<1+*WM|{D#CUWTjtyfzQiTkN;u?o47Ii$y`F>hT0R;>d$H7mJIyWQ<$5Hkd%fnKjD33wOAO zHq=kg&iK>o-PYbIADI_DK{Fm0W*kQ@NOwSy;(6x)^$o-C^Y7h=v9p`Tr0h;UME`(g z=3Q-|ckS!z?sB_(ObtLJf;NgWG#Vh_s2MNzTf$m!qZEN;j09)4xmdCAT7p=ZefvNr zbhfhy3ci|WRaUsLVsuu%lLT>kPxPK{6JK1%cSM!<3Jf=O|IjY>V;1XIj+Mq)X+VnA ziaA3J-sO(2eEBg=j4!?kKfN92z{e+tU-X;#f^UsNGy%txM@;U z63(D{zL*1Dwv2lE6FOB>-pkr5I+bJ1$I70Mf4eqr^zisu)ul~v>a^lNE@5Mb`iW># zM4pv>&$dW=lI6vUns95<_<*cQ{9CMht`>UNt^K!uZt6Aot>A+n^0j}nXvjoi+Z1TJ znW7q5aLva%3p~K4?A@|eoeCaV6Q`}*4&g)~@o=uyE)1p_6mMO3U7Ufl;ys0r=biPp z-X53;iqI%g&gH!fv2+g2SmNPb!~LrEkNb;eM}tLf=}dudf+!V+Nqi^TjdUW**6QEc zu@-lpSMhCCb6xl6n}W;!hm-H=X{4?4--fyVL$JM2PV+gN)k z!Co=(>DKy4N0PU}GyWvXm=o`VO6sdjb&giCWZ!pSYX7}~AO#T^q&PIuA{9aCyVDv8 zO2fvw1zh;5efmSqSU(ep@3OBxBeuCx@hH#zhh#*BwE0bz+D=U*csdF6FN;qQ0qCkX zIn-bR*ayJb_`ob|9gg<#{$u*Si)mls=5aDNL0YW_!8kBaF$uX$%Z*&{oE3;UQ%SBH zWtJTo5$bRf!Jx3McsJ6pK&4A7+!J`6$8RfMEVDR8#6ut8$0Vbi0g{2MiGf|mDKuC) zC+U@)g_Dq!Wtqvi%p>@@bk*}ND&CRHf7#^^SVur=R#0W1FVT_>36Aa=_6>^-MsO)B znt?n#y`1LJV0ev|AUsc%zk|`qy8Iaz>74|ysXtjTkh6QA3BYBpsYsWh55Q*BQv0sPFB~D6Fc*l++%!x*36eM)`nNs$wGDq*Cj-Z zCe)$o3@LO{-qL9n{YmxB-*FF_m1yBr#jWFXF8p78-YO`<>&HN_lK zm+{#$JqTYR@BO?2!?xb|U9&i<0u$vuptKZlsAyyBpom8y6GBrk7v_;_{(e zG1djL5zL&GhqXMm*4IC|F-?$K_zsmL z8(_w#6fkWDpIV;-YpurMp->ng1tyDY=5wVkN!~z2HQ}>(Q^7f*kUb+~$xxHmgkKtp zqR*K7`{gCM@nK5lu7Vbiu{o{E?9T`)?-zH~ENgUqkAzB;@EnC1dT;qfSU*4oma>?R zD=k(h8chry8wnYYN45I^ADGsm9IV~QPaCag0e*7?AFSzr8<6zD7m|P0-JLt+IVWDw@C%n6z`j%hSpas%!8|amoGz z0jT86exkSc$#8|daQt4Z7-AZNX&1%yWYcKN29}nIVh$BuA>lYBNsXip)l|;%e zM{NyFxvyer3fRCf2{A`Fj;3N&PS6a74?$VEe);EU^ax?#*lpCup^RL*A(S4%z zLIaK~f!3tFQt(~`$tw)cF5pBFHzgS@Cz~P%x{nWv_sTt;9c0|P8dxvx$EhepL&yPh&~V34tH@|<%Sg`^ z02OVtYS6+v<<{q+#1^Fqm!Ti$*}Leoc$=X#k}>VWhtPZikh48HdP5%j!qaVB7z&Dw zHKqryXapmAh;|(Xsvlo_p}czEFb6MqHpjv~#Jq`u~9b6(Q69Y6Iai&*Vbez2~1E@I!lUGei*C zP07*OWDqevOo{xAc-BvqJLaFML(N0=^}mqhRLd?r!F&Q4(gewr!*du!@JAC#0aX1hA>Pvi4uI${h#V zY#TzT|JW$0k+mKZGQ@x7ZJiTAyGBmLTq6$ljd);?kTQGx<;XUgyScEHfZv#2z|0rv zn2E_nm1eFhEP^iVnm9KG&Gp#1k4Cw6X-N!nsF6z9L+maR6v8L%jRu*&3m%OB_0%O+ zl9*cyAm@0d9sM}JyLxY5Pg#oKFN$g`TaxQy{1j_NaFAh|`8bP?>SU8`($Q_?(loJ9 zm$fkvYeG^#D~NmA8On+~w?LnGTq@k-=n2X#M=4E+&x9E+mexj5|J{pSYrNB&gF*}E zmi&O+Y37$--98BSjH)w8P{>XKs>m%HXgt&pzz6=+uG*}|{*xy`d{NhZYvQQ1t&3o$ zM%80tKv|}-AG*L6m8(^}Z3AQ;y$wT4q?Su|B5dx_Pruf-CZkqG!{~K_kf%@@6^2R) zcF{p+p|(}R|LgN__c9O7x~1;8(ZyTV+vYLhZ(qP&gP8+=)lp{_fLLazC;#p}GM_L2 zRGwXjm=~;J8A)tCmyR=6cPo)>w0ksuC=MDV?mNU#^7TjC%yBvpT)71H1m{uNV`6L* z{lw&A&kL>tTTN@XSZaL-+5O~I;1){YtTo!C;W~V!JJ+CPG`b(lN_-Bjx4Ae%q3M3=Xt!or6 z$R4o7A7D=5M3qhYE)K04qNf?ewreXgK*6Z6?&PBb0J7s2L7q6~WrjJI_{KMzna%_H7&=V{$bYw`Dw<}f@n@&LLtr6(~`SQ6R z+-CZ@G;BOV0*LT!joJ1elW45_$=r0f>U(Joes$U`lwxO%g&F-2^w}e8@N(82jPqB? z=*=sK;{HB%Zv^f7_0+yUh%3tFZ8H=+%v^Q$R|$T~!giFw-kg(uUR#7lOos87GV;ss z%`0-EBAhFGdxvmo$|IoDC#ka9_PKC*e1r$EKvrz-s9flDET@e16`I~;}` zabHh)(~V*}tA{!8kKgu(5o7J$Zy=$ObxO-8p~_m<^7cX0M{nN5hrdby>efQF%gEg% zL!w46Uv9ORW?{{Zi|LT@{Qu16S&*ywg^qbN`N9YiH_3iMgN)aJlT<7e+dmTu8;vM! z4Koha$gKK0CESZjonIjCcEVoLZ1B)X%=l6H*JD4ipI>nLAjVBX9 z_RxsQxQ@@rmUIi+ie$2Hd48q;u%k3z4v#bn{jE1EAjvgV?CPHntd*ur+L&34)q)%S z(&&$-wflgkAd^dwh4>^rqnR|%t>E(ObIA(6VWaaCreV^iqq2=?slnaq%}7d*V~nv_ zW1rrzvLpId4H~*yN}d4l-fbNsUXkAQJaY%&u^y@sRFKXP=tIrrY9}A3m7KuWA*VlQ z#w48ePZQnVcMdq2!w-@v2gce3eoJJYsE*$4L;m%AcKiajMBm{1f0HjO8%x@C9|9&S zC-?s?P_vS-a&d5Tq%q^dV}bKZxhv-WkYG@L9ZXNcGDcgLL_roSh}%gfL7P>E|mllk45H zpXMfKchpeO=nHw$K<7{f=-qHJ=hM|RrzVJ50GWQ5g6Wd3YdpAYA#$4NG~A-vhC4Ik>xy zHrLh1iccN@CCIF9dn@-%L+!u!%BI)K+rhlSr^C#*`Zc69Fh7pF>u~)O9Pk4E(aFHR zZDQz0FS~LWQ+ol(mj((DNP0DrQQVgine{It*^Xz;ZoaMwgDuRqbp30BU>5=6 zOAAlV?4-vJQ(YYc#A8^K1e;_O=jWTlsp;kZIRf)Py)!KAAr*X_};-}gKFdT{Ve4AHfSyr7+o7*8+tWD4FYkc*V`q2ng*69 z;U3cDRpez9b!81gWcK`Tgyo~2uD3v0I=osx%s;7f)Bgh#3QSAE?;Re2HTEY~rA^2P z?aJBWv$YHr`kUP7bDbRBl}tVHE#HJVx=nca*`w-8pOEpISDn=c@YY%dR}vTmxg>pC zzB7ux=vjcef!2Zr#)0IAmr+jSsRIh*#4H=cx4!U#zI>~EAd|87IYD_6Ezmcgav#%&JLTVc0qZ@Sv|H~; z91YNGK46o9$1n_l>=WblfbjeG4AFa@{(N9;QKhR&MLSIgFSx)u1h za?vNit;Zv$=M6HYUa2qt<9lfd7(3FB*$Fp9_-O*~XO!3@o(Vi9{(bN5Cg#-yrWruc z@u{D`Pq!`S7NH>AJfNJuClFt@L%#Rkw_k+pKpAYJ+M|GaRfKOgMgUhPNdE6sr7#wV z2ctt^U2<|AHQ2>FLanPaXy@9t)EdI+?MecOcDl&F$7KY8)=HZJarLDR6@nj|f;2{b zCx64=Hv=4`qBKuI(IT>3#m~y-lkjgs0-OyWad}47=z;%zxH-3;*iEdH)uk z_OyKeR_$?hebM_Qn(TY|jA??*{5D{s-uRZu`RYKlyKeke!PrPcA%e#R^+}^6f~N$Y zo^Nu2w1=y6NVhkw%kOP7I)IT8$jPY$9NWJ;&cDKz46-ln1J^YzuCKYjObV(DY5T_ux_~m_T6Y^)tF{{TJ$XoWOyYZM%(xq=SBTP-m<&Zx z{;^I%mwagz;R=AtJ+Uj8p{LtX7Nqo$T!Xp!?}de)%i}|AM)_%VT}=TKY1Hpk@oKCF z3*+MbdO(7UXG`gI6D&t{x^40))B`Q>c(4w_$a2OE;k||Hp@Qkq2lMa6EAY>k?B4E& zwqi35(S#L*LFfdV!7ig?yteb)@TV#L!%i%ms58}}3BjAstQ)Q>JUaruEmr)bae?i7 z1kB$>YZ2*8BG4w!)qgCeB^Qe9NX3x-(aA$bF;un!wM$a-3^?S6w`Nia>F5N4HM>!JAPb-?y^BYQA)=N8p}L9O9zng9HXG?M*nxgwa_!W4}TR6%#6 z`()D}xL1e^vbnLRiJSIZqY)^a66u7WQA{_A0tG$ZO!IGFv>%kT?_p{T%^29SOHCt5 zU7c@p$wWz8oJdt0^8{1Mxwf*^%;0D=_?P*<6r~9&6o<*?w>MtO=(V znVzn77#R?P4`{|@gnLI(n&cQcT^gCIaJ`iiH?I~_H($(J1pud3#LDyv9wBC_=q)KnoS@ebp6?XcZMtJE=P-O!g4Xwk<_bOUu zbOhPUdpz#XMVZQ&2(mRBWnnzEKS$z<@^oz-)xtz;TU`Hp5vvZfji7B3n~%`4dUQl0 zPyEH&f*P(OZVebbR#b$6#H9U=X(n4G`0oWpKx15A(q-;gO%z;{LTKAKz02gE!%+c*F-Pg@Lf2`*8M-=R(k6s_xT6=_tscC z$QBtX9o&V3>U!CqNO%P~-Bypu!+n2O+4d?VYtcYO<8DA#Wv|wBu}Zw}VZjBHNW8r1 zGP8f}T;U1*Z-F5+Je$(Buc|BD$^JC`t&Nrw%Vl`{CCb+FUn2qCW1h@^+mYUXP_Ne? z?cKD#phBT`HX;ut?+o!w(g!kH+CM~D#2GFiJAjX;S6sI__E#j-3q!?Yfz!Px<($jz z9a=s4%K;EZ(8QRF{`T>*q7*{!xyLlbY-j`_%g&_>2|+g?#6IA(Fz&W+AenU90@{b# zOd-(@r74fx&g`U(IVZnJfe;|P6@cYikmZfu7APMU@na4pt=240%J*tU=N)$TK!gk8 z+Q*i+uB$&uTqV0(_A4mfAf0ccsZkA*%B5;vzY|z=N%}25$6&)~529K2QB)Gl!pduE zSVq-zs&;S9f8K4;7sx;LXTe>>N|a~AhH@l#QcI-Mc%|?{ zT=g76Z0nV`W(8geyINMw45QA-f4uAnc7c}5l`}KvsbuPjxE}8GGPflre@z-{DSyzF&9U z{qQ@WIuF@&jOsUJIsv`hWq#o(oaP(RQz0URQGL642jjk2r-D;;yFH`T5plbD5{d(k zo>uy^U;Qli^J}CF`|OA7Y(Qf}L_@mn7y?j#z#M_kA`#2$)m1?`O7b|!&!TwFEsQ6n z8ndny0@40uyQXefm}#Zim=#nndY$sCpyr(h2RfDevOQD4+`uZ{c2iBKVFatXi9m%H zyy6xO)~t;rpwvKXoCsANS+dF5K{MkCO-W6UPwZEo`$@;pz8{PrnVsLsPic>yyZ%uw)W9C| zpL_}-=o32&sM7gEqR9ev6&+2)75BqC{ez}*K2sgtjt&Ks8U1V))gc<4+n_pO_nga` zhDg?s8URcw8O)i!sVE+EXBoh3G6JH)96w5JFf&(xCkKhC=H2!uJ5)81bWuAEe z845U}>0ebvYj%1wJ}7L|(uEcv6+a|6`k0IMU-YEY7&VX5#FrWlYfa?p6DwX1ScM)g z6HdwUD1|(f=c{?^9SYAJ2QSu6a*{}+*!*)VZ~_u6v(|2}hq+;t5L6d%{wRBsT9gUN z+Mb#;jJnnQPRtj7QP!HHi2$s)As=X9yN4M7eNO6oPNVY~luTxU(@SNK`y?t@H3Cp! z?Z|TfB?Qa1nb+3{s}$l;oW{khJ>t(okTR2mjgoAl=s*NSySpg+XW_<{xNH%)G}rg_ zwPZ!oa=9Z1X!rr^T#{$(>bJsRgLOQDIY*o&H-es%f!7r+?)5$M-N4^N+l^BAf#gxJ zH|;0CHFka%62msgHMCG##3zo2G`w(f3I3vYAzgiJrC+wQ^l!8Gf6Io6L}6lN;%x4w zKB4ZgTk=lVgB`14v#WigN_S%3(|(*<>?8-ehUG9mv2~TfU%bL0A(BCHEQt`lwdp-B zgu0GW$0UmSHOyWoUjr#?8U>HCOB)fIZlY&4xIrq1e8~|zZ1vs0-9i$MPAF!_9NB1K zaMc)x<$@POijFvQbKV5o>%|Geo-9wU{P<3;>S;AHKZ)#bMqZBu^wZyrVk;g% zqqHy5hm&dXY$xa4hkZNfrT!6y|HI~SUSV;uEk}zS+4j){ z=8elMc;O{WE|aHz2_1nsca!vD|5{sxtpVLAC(|D5=Yed@?MmoNs+9JDZdEA#JMMN) zOZv{(BN>=Sfagsbill-Ni1dLApam3oD{L7`^B}xim%HXFu+;xm4kTful&0!3Nw@(%b$iFfejbJ0ISTO*H0Vy*NF3Kg%|cStgO1RxXlWXZTbv*iEI=4rzj| z_+sH^o0gmNHf)(CG+P)*-(WdnSKd~YdHxD{k!!R3 z!>4#1;ve!X7OE0CQ$E)B=52Z}*-w%ox`O0bD`pulmgkz!sGcFlmxM#&woa~%C3>T^M#xfnSsqgl+2 zGA$#xdTraSY8*MOSs{p05X=H}Az-)i=}SBG2<-^c0#!s=0#>!OK@}luy8%Zb!+ci` zuW`kWS{LF;T~K8p0BKCe^sMdZ-<{q}sk8uq<`h{pPYXIjWUSM-rIy$3SVdV=C3wqu z5hQXLg;@PkV59{t_Ecx!nX@SR zECY(~>dWTx5KQ@^NL=i`SQE?BtGqMj6Xvn)%-owLiAZ4+yS@4=iCJm7Xv$QjC>&;W zCeFS^6P&a)#Hw&Qtv3(CANl~l&ZRP#QSNj^_vcd8aEQm9@m;X&Vvs1F^i)!4g5IYk zFPY?D7T<Op#Qb@wx6)qqbv64(gsn#Rf?F({wO`H4&e+h$ZWUb2R~p zo-n89tE1Z7-R_e!Iz5buGO>8T+HBqJOOy0~zQ_ah7A zO*oKO(mv5To^LeANf|+OWjJ2HK%6w;AKX|BU_EEsb&-YG{H+rJ*j0Ex^jz*L0Lcda zlOa%oJUb;LD>H^~IG6T-8X7QIHirKhev?OKgu z6_0mW>^~mDm;hzZid3j4pa;h{JxBf@4S%?((7RusTylU^9_q936hQ-`>~B1(b_nDG z3cgfpHs6fl?4M4eP+QMIB)1_=V@s;d#&%hb5$U-8I@G0raR?gZK*7G{h2#TAzIO;y zztFm(a1WwH0>MeY4W9A*&Gcf4n;LG3~7%brIkXX%cFnLsPbTM8WhQ`(`C;w)@ zzGj0w5sU?(8f6>WxD*I~L#DpGy`y{|yOJ4T@53skW zF*J}HysR)ktOeNm_Iwc@NtipPT@k_%xYJU#QDEi|G@Q9`xGK5YUJq^Y{{{s-cZgN5DQd$+P%nT zz(52Z<0)Hm)b@a>-?pW+QN0o3y+lxO%>nHy6_+d#Q*pT#JYU>39NWeaP5F^@f{II& zW)nF=4wbtwrg4o2$n7(9tNaHt5gmEZX*U%KWdG3dS%LX;t%2(U)}hVYTI1eTAO;@mD!|dY{?Nk)Tt^B-apZnt?(dV+knc>mp~_CkwD=@+V;-3eG6zHrz6do*j;*OQ$@<;n z(F-2)hVQZ+Zql5Qd45IX3T6{$>8kx82n5KeZNB9UInHQ#LM_6gUm>vMbOPEB$Z!@K zkM^Duk?F%qHnA@I<72XqOiQBajlW8?qWVK$ z)I_H+S@L2+V{4?!g|zpC9t zm*1w!bG8q=C|MRR8Pa0hU<1zF@By^E+oUX65eN4!_`0bfpSQ(SrlwItz3kR=Nf^-C z&7=<1sAseAr?qBPbHeFdmaxl6V=>lRkmbq(CY%^fu()xdnb_Cu8FHTEw-IvqQ1h{2 zwnd{|tv35Q!|ZE2S(%-OD0BAMva2g=%ny()Gq+`nQtuP5sz2g<`=^Kb_Gm5FYxTsZCDX3hs;24r#?s+OTws-oxAcd+6=jxJpLD!B<;>b- zJiG?FKqQh1Q?e}-oBu0TkD3Pa_ZWjOEw_Q^SJZmq#!c9SCD7y;Yf^Rt?^>${#x%^9 zs3fmI{dd0DFd6#-Q>aRjpYr2(aL!QNcRT1j;n$x=8uTUcH~Q|y>&MAk(#dj+3P-!T zpZ*Y1#Pq1x+3(MppG~pBv?yC58QuaEu6XUr^GXJ3f*vPr=6$oEK6&h_M!S9E;_x`N z$eL18P(O5XVt~JANT74ZL1Me4s?waEX8U-qmxC@owR!YUe)jv_reonrrCfH1)I(qg z89IFpA{J1PKY|w`x&$WKy<3SP;1lRr8Wj~2*dFO41%V(K=?&eaSeYpC85(A_b3{D~ z{r0euga-N9hUqGdhx7{7)D)S9=0p9NxkH?{$oC=+2w-J$g_=xg(JS@GMe$LUsqkO* z>KsBBdKQPqvn4t=HxntW;TNjP)DVmnPkDST!LB?7TIcf(9rj<~2nQI2?O|KR5u&Nf zeXCvm#{Q|?bzsUXa8v~nap1~JFD5C@&*zOs1`mxqYvu67D>j?NhcA!vUS7LXL#Vq(D(ijYRpDbJSA)G=822I~>L@ReaUv zeWwRs#;*18OfU~Rc`@zu%!Y`FfR~-25uh+oPQ6F{aQNERq&~2eMWj_RFh8E~dQwGa z5-<1kZBMY+;UU&N3$FS7?|UuSHLc=j!2^BS_E%r-&7WV{u{ONUu|vbZZ^nF}USRx} zY=9pJg`HdU49B__Zd%us56qYGC~6xwWoTO6Cswv}_9q=7t*h8Z%((1h;F3^+s$Rz? zo%#iWT`eeh;qz+)KR=$VOIx_MiE-2Udun1m5o4jj`rZw4-j1_`Cm+EC6iwfZ522cP zrqu;e@?_2p_+T@MgN5F&-Nz#I1YBm0FacW08pZb`r+&d-?fn_9YsPMtQAgUaRNV!>kUl#>lnt~iH;@k!MhhC3BNk$v*hP#$V zy2h8BwXrP`pdrfHng6~cV*g+?E6Xqi5V_f(SB5C5_83KC z(*$wIbgg;as5-**!>BaoJ?1-Mt9UWth)GAvM$>m}csA4#z?ZVSOHmPD2wZP0h)tR# zOfEC2YzPNEMg_4vQcUX~YqEX8^~Qwajd~>mkhSXYGwuacP~nW*s3^qS2r|*aQyX~L zr(>Gi9Z{#XiU}QetBOhIR4R3Ka6|TZX){d-3rI| zw3ozlW)e!~nK!(@8K-bikrXrY*rDE1J*hbgZa*xRcb#pj{ZoX?WQ%bqsJ2FOOTDk| zhVd2V!8saf`_Sz$oiRH4flIMgnK~-Qaix9(!gl0lP?~)YEPIVfUvNsvMcRT&MPjnR zfp!>hZ#DbWAy1F;9&xgI%`8~i5!VNSyuz65}!YSUCpkUqqC8T+Et!YGk>-<5r ziZu~~H7}8m)}C*2~k7 z^rNVHMz;$auvN(q&vAICs;lZF+8IF9;q*<$gWc~8YHFDlwb2xeZ=Zz(x*U*^NG#Woz98(TKrJ?yA?R}q~ zjvU5rwBntt{IC(>i{$4~8q|QulC8EfCWTl{g%jsVq`7ls#RR6<@%Up6_y8mDI227} zf_Xm+fIz_DeqzR-_g2MQ{LZ!+Vo;|o#Z`f zp0ECbTVeSpm!U3?;7X;YN%MYR83~8-7h82;UhCY|i+%a6RAPj#veq$FY$~;>**;`R z+!XSyFtXIwu}Fx6%AP-DshB{v(|B11uNv>yemhK|_wFaW;kYxG>@_EcD>Jrvc9d^@ zMw*FBR^+cNGKRkrrI0?Pk^?E;HGIN14lH4>pS8L|$=RJ$FOt}j%BX2?bj?K+Nc0Q2^u4je9#*c=!jPM%~A4@=tC)5{K$J>X@xO# zt`Cgzl9{jAqKQ#Rk^y zf_00;BKSm@jH%H8eCLn3p9bPoPlVr2d`h1$TEYAtsnD5OmWuUo#crty96FMKd)_qj zhyY!0!@+@IYl1XGnx}}DEctaEYnAd-! z=oQ(jrt=-O4%iZ&1Q?nFX9`tv^H=CuHCznCHjW)5vY)Y5R^JDnrzo)a$Nm^syDj_4 zs2hhv9N&%W_f3tV`w5l2qhgNVE^Lf-xu)S&M_D4O*nOP^=N;gfO$b#0_r2-~8~xNx z5rp@PAzI-OiCO}Tzm3BWlui8(bYMUWP`PZ7x@awH__N|EGk>T-{D!V}w_7pkp?gRO^ z%ZgAYYqzK^N`lfi3Ud1yL#|3Y3kwBeasR?5Zy}m~Zq;zw}{qlFTUS*6;W$V22QRg_YeMFIBQ#nC`iqseABj z;dT9%!uw=RS}>Y$B1!AEyD74__6wq~tCOM18V12k5k3&c1Uc=FcdwKi91{w09&72c zV1so>Plr^1>8j$%c@{DhVV>~W0njpW9m-!>N8JK5yRk-|1N~D@+;Op%KjBE8EG@#O zqQG%Ix9rEDRaq0XqI5pUe{8a?vxvc^jaA>$N)o_Apshi?exQKVsm*K&wOO7*^!-s6 zOGpeYv_P5rdL5!L4pXW@#vZSl_AQs5K{zXfnWpOng0S|iX%p3=YNyI%N@R~VGZubs zc~qENpdvKk_?3dkJgjHMfqf*Y)=8z!HB{~Oo*(;NUZBe0pc-W*`X61dfQ3d|S9$%C zzo#A{PXDwE%?LWar+`}w<*mv)t)ZPf8LOtJN!XAYVPg^BU&O36aH(uBG%oF7AT1zP z^<1_F(2%iQhL2|3Aa3NRpT@5HQyho;KN4_R6u9MdVTExf^nRY>aarBL!kC&8?58@? zK?`h+Aw=yKC(mf8(`hr|z!#=$2wfsw4TjV8l+|Xk%CN*>!I-F{%;IitlVx5K%2T+9 z6m|5qTb;+jJ6FT(?nZoUABi9duF1Q2DTDa|L>ELR6`W}PV-H%>Y11m*^+7h%ReUHv z&K0Cv>Clhn&?B7USo^Y8E||EO39&x{D%H^l7nihR;yuVnM>}@sb4B(9LUtLlOB^Rb zyR{pcqpi~KdsWm|#0ww`H(Y~;{S9&9-H#rWsa~v$l2tnHSD6M_-7ZzV5bkIufoKOn zx6ML5HFg)KB(m$O)t6Rcj4!1?dHXcr7+UicM%63uRMIRzvxilAVj#URgK)nv+8v>KF!OF!x|guw zU)L;sA@a~-l*OPLo3VwhM`pQX_r%!(do`HY?}~AnF4iX<@uCp;8QS?8m42F0zVlq> zs4!Om>Z^>dzJH|Cs)q(z0$(-6Y$;1@1Nz>eGIiK;#Scs zA8yKU0w`b~P;(PubfeH(Xjj=PG6V^#nrQS6q2X(ui=gBOJZRvapN&9$8C{HBxJEkE zugB2Tlu?cLeDeWi^1%G>Kcy7BjBEQ|+n5$t)sd1p;(oR-mevb8LC&QfBts~&g;FUv z8*L-gT&UET~j3> z*_a;X6;#pRBp5HwY^A={>*L9=UKNMf99v}gj`XuK$lx}DG-eHux2TX+?JHy(`*mRy z50q3EZHNB+0dNK;DpS6yVM!WWf2RE*QD0%3{F%a0RJ{Di>kfd0NSxV<@9v$-`k3)B z`b|1kDd%(hc_UDE#F4i{&p_{JI-7lF!eOFsHZfz+VtKyNzk>C%CWw|r9f?t6rVuqf z(o#cBOeE8p>9=8ksTuIG(l8h$+sXtrFLd|`?|obU8IP^=)%FJ+FLie!{9m!My((gVX2AyD8Jj>brm)yi)93Ei*R$4e z9e4*W7g%nzwIX9~6D_mL)QZubR&>_g7v%yvJtwOOFy0~4c^NdAqpCQ(s(u8O8=Y=4 z3-;Sm=~+UU;mmv=tXLYTPoNR`h&VIiL1x(3yML8J!<+$GW1|!DwFP~h7^*KDWvm0S z!cgeG;zJWY1^S#F{@l^>ZY1(H{B?VcCXd3K_om-};>)COuh6k{u8CN z5pl_6W$*&HzFa8m!NQ8+v#61!_sXK;1XcZ}<(F_4Yv|?!cj7jNezG-%>~ZZ&*K8Hq zc8K6i>pvkt;78(?!nSe+qGAm;wj98$sP{Bvq}EbdumvHm@|n)OGoO)0St-!4;G@B1 zF^^8!X4}&-0!`#;Bc;{zT}Kz`tK8ooQL(~{ZhJuL_(v=JMUhO6U-`fF#b?4v{5)A3&=@aH#YS^U@@b^lG` z-N?$0^t;>WGZYvhx>-3U%#&@D0j=xhp}Z9Nj4Sezzql#rk6>FUM-;6n;`ZR4tT5KW zKT*K1K53H1X2mn+Z`k3V7#TQ@m1k3`d*ncI@bs$!X1m~O9nVP&-&-Uk%kBq(xZlkw zY{+pnaX!*-ss{xu4 zcU+x|$9#z^?ed-Qvz%-}0qfL*oraV%m1oF9{vF2(uKYp&QE4LNQ14Ibm3Q zRoLIN8A?$;dlM=p{SB4pcm|cU|IO^U1c&~Vg#7j32+8rl8JTg9TrE^OMVtsk7|`uQ z9Gfw=m_b(CbrP8NrkD+-OlM`RkSL50=Xe4FFQKTJb}a$oIvm02WVlTIae0>|S(g1H zE!LS@>;ePyOYy&obO~l8KlgViM-&eFq`iM3R>iE=RD1IYEVQWSld+S;eJjFnA_MH0QO+XK6l%)N^bXdxFhz!|QWNZ*^v= z_Y+dgnvkjYs>>^3w;s`_ahnRoLj@*fgQ@x&Nq<<8#qmyf-mAqaErV#^vR|2^)|wfY6Y z8;CyL^?sV0m#zG0owJo>jW(opRYX&L>PSoZ@ppNm>MF?Afubo0n`>_2&5+1u4ZMe$fEqqO$eKScI>Y7DTKNF4P|yedu*1kYTxt`I>BdK z^)iKaResH-|F#)WFLh=BLiq?klJ5voyPQOo&;HHuBR(0rs|pdLx^W4liQ^Ug>Ue)X z>EbGGu+B;|RK0?DOB{p^Hc^>>{_1J5H`ec7sW3}1_tUR5e?^fuP;5s>exoo>&;78K z#npC$Nb~6+j7GGl$!tMA*VwhOX%&wH!y(qWJS|`OfG<7Qnl;P=>aQlQMXQ?+D?MM1 z*IGSVVnr)6|0)5llO>{(Nl^oH7rEAV)A;^IEcXT+hmGgB6nj70^3Zy0Dft#_0({xu z3Fhize}=qNh(Fjgjr$ity_h)1hx8Dg6sB(rK$Or0Xotr_~)3$jxR;2_{A(#Wh zJ$HP6_ZL(_R_`N|v=v)2pQs*}1nRYoUrfJ7R9DGLP@BI3h$0F4KGgh?%{^katdMC} z0EHk$LSB4u&x+rU+v7RVjWmNbk2jKQY1kq+GAg+rwF*@`gjX|w(OF+MKDQR0pv?Ie zGALEYr)VwX#rk=lHTgRx1Qs6GO(mj~<2jr-i{3)w$M{d%WM`ww&S5FRL~VD=_Ji5q zO8y2HOKxo~00?|p(8XDB!TJ^ba=yHYH}8P$khk)mxMg+FnH-T&*^D*yX1_#6xaqj- z4hjr#ua{9%G3rIyy?i-@0+jhby>R0Fe# z7e)=H;}aZ~PLO@sJ-_NAHw0{rFLz$Zpl{R>SiX9#0kyRggASs-|5!*Z8yzei_#+;1d;$ zgmp2`ASwTJz>5Akt;g#lQBW^-7t%RfN>q=%V>&o7v~?HBwyJcxh%)?KvdL^?<|>c@ z=%2>oft(>9m+amq%|>=s_EW#QIQbon(1Dgc0Mrx`3dta6HM@U{M5S96bvpHhOJO8x zhS7+(r+PKUg{+H>Owc(jyFYCH0>8;e3Hw@hAWC=PDp_I)5ui5e%@28i-ae$T=+rmQ zW46Ck@cJDwkV@gW!(FB;2+04V$6+3kNp|iOk)$aRTM1$9Ib&d03VtDTtYea5jKGSs!Hk% zialohDN>Q9v#=*sMxM>8cp#UF=KXMsDO2kjllb~H%#rLK z1U+|PoNcIi>&4hWt}ah(80sp2?$Z4e*!>8;l}NpyX*}MJ+jUq!)JfN;I!!Dnom#o; zLu%+#Q#LNQq-6JmyIIf_6MO9=LW+ZOifcd@M3T}#WBfCl@xRzQ$JkteZf)0gcWv9Y zcD-xcw%w<;ZQHhOx4X7&sV{qaD)Q ztmf?Snt#e!q8FI|Zrb@sI&tZw6Z@NC60KzL#B+SfUgb(sKyY+a5;0+NAmV7GB4fU3EwRt-5mnu17VhkVU$+rA-5rc(3{E>jB!?}&!AIhF zf5oq&`yTYF1cR@)*y0&*v zw;-Uj##E{FUGk&#_`Rgo@pfWq?}i*@vWgb~uAr;yE=B&;p+gx(UKG}uxC*#6{0D5I zDt7?~>1>s7;vPV+C#%Z-LPfbcvzy2=tB@nKJ80 z-??ScQlX+pl#x+{dMeA7&zu0Pl|?>yAGD8<06!Pl44HB5-}et6kB@{WV0=5pt+Y1- z|K8BP$%B19tbD4q`ZT+#3xRIVvA-x@ETA@*5qlIU2$4)MO3^7I*D4VYsmjDcBwSvn z8zRzan=y5A84hr|GBR)7RePg&rw3P`G`_hCh%lD_knZ(}37rBTTGRmoMA8g+^==q& zB#aw3l9E`pR${#9JwncOL&Xf1QA*F=V4feNAxk#JhV-nY9qU(KT3Ty;;n0>r@52pA|@*C+|B#(L|) z5ba;Lp!q(g!17Ob&B*`)2pspQ2VI|C;BqyO)w-U>rZZ5o>cSR;n&+&O!D%#pLo%J1 zfl58>#kdnL2*o6OJgaa`ktzD(-l|xw@Y=#NbvjJra(u-#C zceqHW+L}TKp~Af}ntvU466Ng7sZS+zs=2-VyHs**ch>s1)KdYrA5=5GLtb&>^eq1> z9^TYP=XI`mFVHWOrZCNlEt3pSOZ%oh<g8Ms6jzV$5Qkxfv1t{YLTkw$WM0A?0xL{;WiclbPvhUOrtkrINuF$mlTi4 zt2oI1yb!2va!~N!FLU+!I?;D%{`T6ZJES$|MLc=7qr!lUq8z(OBrC`}S~+TN*`|V% z8M_O^cv|t681XA!AoL~gv02>}jMJu(9U5Z|Pm&Y=H4);DM+sy+-Q>2l{K)Q#JW$y& z(S;T6F@hC?Xd<;X6p(wGmeK(8U58&vtenZ-%xV-9-JvYe=;wKg>@gBojj)dY2w?kGHlv=~ctCWFIn+^(;<1xwUhv@i9WUM( zaBifNI=1aTfja8X48evog7E)ny40t^wC`pLy}Def#Yrt8(bI!BZD?f7n9s=s9u(k* zi+=#d3*Vn)qN=2CHy@a)=DaH7{aE%+DZiLEOZ6iRwU)Nlfk0PQK2y^uSNH1rS8URj z=O_)7_1obG;KjPghw!+(#wW_LcWQ$Qg0IP5ef!K*J?D39wCoOsxIFskW~nP;?5yq= zF8hHeAIZT)H59Fh`*xgGqsBdyx$Ctt#K!^N_0f=zx8JUUk7DVGA*oAzqjvMVtOM5ElidzHQa{U*t4RfnOHWC3>Vb_4= zT4WuX$f5?5j;1=s-eHJJvf^5(72ctU45$*>H3}mBFg2^IwK=gJ!-J00SI^QN0c_Hs zhg5yw)(Ice-jz`fH3OeH-m~~(vdT`;J8M(rmW>i=8Yfc0qbufD(o=-Jt7O#AcTIGUlWx(NNkz?Du9*BB z#ZFe>Y~K6NAvD(4XyV5s_$}Sg>~q%)X}=YcIY^%5>P$j6)wsUnYw!Nhk+#(gi*0Tk z(p5U?lx#y|*nSj^Ct^S$a?KtfS7>FDd=~I!bS=Tsk|_-X*vvh;}}8}kF{j5v&LYdectDGECTjBaiu&rie9d9&2}y)!4- z{S|wrPh{Tvxoh4L=?(@+71o7T`lAeW^Juik+G>31Txxdx^H=Z4YBNDgikz2D^~}D^ zkfib)4AUzvv=tlp!3bw?o%zte&SzodPHj_CyoY5G6M_Evi(eb*cfE;9FDp8Ah z*34T$38d@}Fy=3e_kL84n6zqJ_u6eLbWE!`Q~ZnaY1f3uekcU+s+9vX=lkHs2d!|= zx8qV1a(d{8lJEHg|3O~33;NLb(MuV^GdF~;=u!5H1^717*@YuxeYZo`9sQoRDi=IF^SSU7@ElK`~Kc6D*sd zMdKG1SHzbC&_Du6RnL0ZXL*xx%{lRmL|4v}-c)KLE)%ly=aA)@*(QEO5VhJ}>-;(a z85rl4^4lV=H^FF}+;sZ;%{cl+ryL1NWzjL?#V_1bpItJ&OzBvs>r!BONe3fsOPPDG z5rF<}0GUZB@IW#cl&jIx^_P%eU6{e`dq?(bbU&v~D?$X2EkiH_D*dz=ZNsTP19bWP z?+w18$M6Tn7Jmk!qeRkrHHA0PU#@RFKQa~X<5c;ES_rf>;r47B*CmoWsiBhAMYubuK(B2|N1xwH|x zOgvLw+|s77cM18u*GcG*mNn1tqy7!tsf7pu49H7T!B%h=~(9{2k;6zFN zWr|OSE7)GPBVl$PsmD9wL8{RwFW&195mJ8*XM`lRbAI7Fngb(OU0Yo>LT;9 zIr~x1pQP4G2Z8+3%j7DK<0RKoudIqviw8wieVSCuTywo}&o*(3m5AlYmZ( z6*TnIf8<}lU`_R2el;s&Jxom@YiUS%KHH(95g$6kva#fTRD63GC^WpqYeq|^fWrR! z5&?Q*G$-vj^2?Du zLPohIQH)E7{K_u(Lj&zqy0uY_4)_P~!&6%#9H_z(l@B%*LNo!ePj~E6FcG!C zbqy>}EIq8rzjo%r7PG0kWx3$fEpv)(3OlqRH_$)n0%N9?c_0bU|2D%#t2YdAyyEOo zua=;yvEIL6A#5E;_fOe5~>EYeYV#8^gr}iNg znWgQ=RC$Fho2TGP&@r8j>w#`Fc-o2v7c>=mLcf*-BCax^inbyYHVs4DTO~HyRB_gH z^zdr-2Vc88oqce`1qjHn7g$@T6Lae8ABH(!PSXIdmY;whM>5%H*SY}kK-C#a$!8jI zGW)G@7psZs%#7qL0DOxDeBh*w!uesY2&3A$y{vrX0vw$gB>z)mOi{ztohU~?rUe&DWqriMOQfgBD%l1YkI()2c`CgYD+ zaLet|BH8NwtdC_3;9JGm@npM)|JJ=Xo&@$=fW0m6Edp)~s2Erst_J3L$tx}1FjdLE z=$f>jlxnO18{tegGt57aEDicPH{r!pUz}E}_#~o;)Fi(#rhh-PlUf2x*7TgSJ63=kGAgvP#G# zyXJ{rD8J8KLqx#BKVsBd9UM`ZOzCc}In%WBy_yKy1)B@tQE_Bpc%Sj{C)x;i2aQn{0B zra$eD`=c`e3?LgK`R7IV_dl4w`=TKZ;k1>@XM?uftUXPRQ3$3HOaDQ6!e5gy$-_)EtS`{ zCGrd;g^mgyok~BTv-uQmf_q@Ko6N&n`e!?)BXo&%sFsu}=Q{JD3(HU5XxjGva0WK< zk?>|U=KMhhp1tUXe$h22tk+NLEP=sE8gT(Jrzmi1rPpARUfu%gH$i(VF%p4s!Yz!$T@1`^g@w^gKWE#*h=`IkFTV1a7ZvFVho|nBG8H{o z;>CV^gV7|j(`tFZeBFG3)k!sqW#tD4jG!(FPpQkToX4-(ij`Gn2LYX`R0N6EIQxQ( zpz-qEEkeoHidT{Z1YXcQxKnaZCdc!8h$tZgCib=$+Saa3huUU}q42Qj1hXXKFmSWg z-c8VKr!m|ZY1fenEypBJy}@P?bBu!{fCtfT2-&G$AR@xZ6G-_#2>&L0T9tXOMZnhl zGqdzOo49+d=qP9F+YQ;}vlPGTfgjrlhezHCGk9IIamzFjk_W@%NCx}sfEJC%?DuH| z=mJ%u_$7_TU&_S9vLl(tL`n00IETmLZ>k~j3%B4W`I1pA>ANBTZLF5ROHXUX8w-Z< zFM~|Ki{GMe!T=)RfLQ=yThz%IOFO;NW%a9fYbcB5G-&sM(8o2|_k2!S&=dUYQ*j3z zJt!nvz#Q26>YLKzf)5pXLEM19hUtJCkZMi^6IcAHi#e&TME!z=<2G)u7()AbZ4zA}!m&U-}zODY}k=X_h$0jC?&l!t;aUwf9QIoTX!{uo+~p6ga1wM<)bdqJJ%~&T}AD;!#<5?QxCd4Lg}m$>`YymgNVw zkCXN@j6Lpam^*isEg9NBc$ZlZ*l~FazQD`B-s*6S@NJA4k8gA!+|Xd$(C#cb7T!vl z^9zGkqlRHd(JT5Jsi;L}7(e3_oe3WCV9=go+Dj#k8})j!2aB;><;E%wKB8iQgr0c2 zK+c|rC_;{@6lyT>_A=&If3CtRe-s+mmZ9c3B=my6a32XLU1LOHIC2sMumm}~>i+ty zr~nt7isqY>hdPy#zLpqABy(Y`MHob-& zys4L*G!uejv|lz-5wI9-UeplLvLVM`qAsUcG$JJHMyf=?q6v>C^b|+xoAIkCD>UL_ z>W###nAsB{s5tpkgGxaP;N=5ar_A+xL#Z-`1%B1y!7U#orSLa^@yTeL70ml7!bh?X zMU>w+(F>>OIJ;kOx6J-kZ=T>QNv1>|R7CY0r)1^voYYTGbTWX9PJo~C{s4~TVV6K} z@pK;8>sQM11MHFbIe_Rw?Xof4e;3)6J4X{($EtAApp%J%@kd1x@Yv`yHzu4rijypk zlR&%8zS;K(YH&HEEyLLpjy#MgF0<4+FInC26~jd5kR|W4j&nPm7HO2l)7Uc|bR}MG zZm<~2dn@u! z@jv4~Rv;-CwPi$OPDmu2a}{fQAo;EFs~bbqHsbp4>>s8tKq>$8CI@Rh*-OPmJJ&Wd zctZ8$UruKm{#@MI5AF;7z+A(a1hq-_%TT5EFrx$AJF_rTi526nk3ezs^FfYN#T8KT znB?xF_Zm`^vgny;)n6E2n!5KhIzhga07G*Q<}=kwppj8EHQT(yvD-VbUnJNR>Z~Ja zTljKNXC)gR0A8GbULc8{Cx>$uDeCqU-vSh~ldco>QHp} zSfS=Uvk*YUOG#_9OltxrL!{cp57bGZm~tQOzKeXA0H%M!6&S-!pL*W^PRY}Qx=f$<*7nmmHXt1e3~vEZk-6R}^Xmb|J*$6AlOCokFvTahPobmNom zzyn9!0W$GZT|`8>Kj1FKaUm9$PjMw3+Tlrm?149#C5r5tDmaOQiA3xuWpQ7GvOw$7 zr(wcR2e{CK- z`NP3R>?eP4hRSB_#S4HY_2AC-^}inKYiBk#0BZwJ1nr+O(*7>?0+&k48Gi4uk-w@T z%d`Db0^&VvQmeD(nZu=8v##E1dv2LI3p)~El&#s^PCd61EiI9Hu`-e=mZ8bu9db%C z_BQI^omAh`(IW2Ok%(b?K=X$?Z^rQ3bw0{uM$<6-h3QrSB?>WNT47#JyY>pZ<5v6njjc<@5;yPbLDW%{X6p~dvVz1mx83C%BmsiWgvLH6HsrGwBc znw8^KXx6sN_*a09`gdJy{laG`YF)R06R`dFqb1bHSDSZ#>F* zvW60utR%Is^cSsUJjc^?f~8web*BG!eer!-WK8+BQnq?R-lPx`XpLifg`!iyQuF!H z;7}~X>cu41M81jg^u#V3H5nz?BF^+OX-)Pw^*Q-=8VYzcjvCF}`c29K#vNAejEHu0 zgu3@&)tLUcOJ9R-JB#4O3-0^FSvJXhWtlp$0n>*yLTd)B!eMQ}u5Mh(@Wk?j2TqBR z&yCny{QZ&)ZG{r*tOLdiXS%N6rFO|avKFA6oGBZ+m!@e&)bHA)n~@!p7L8Q%Y~6|=q* zH7GRD{cePgfZbkJ{iH=Z`g&Qag{X^r$+pa{aF9CMvHf_Z9H`?rm^n(A9K%iLVk+8W zk4kC~)%K_kH9{m&3(bvgi;c$^?PYjb!scCK-X;Q!7c952Z{ks7ZTC`T2I7z$Y~=`| z;|y(U((0>u5u>7~ngfE=oULhU z%|10StHb2>+nB{m|8H?0ShXUj^Wkl@KeQC;hKpJo)O-s+_0O|2r?JX01a#C24vkjW z#b=(@%$oz?7u~b>D8Zz>e=VQFSz=O}OMaOV)#kph%pLl^EG#Ww)qSUo9P;qDu*iX= zMN40{vlrdQ{d93fii;@&Uewu}SSe`EtP5J+Js)PHY0%G+rOI9mY&1_5#O4A4@k5p8 z(Q>LLufRUM)f^Kjx%Fc;7`swG6?f%*-tSG~40vFG>aTw2*lP>RPWc}CU$&E5=n6_3 z_;IyT>p;~WT=hN^_S3@&aB5TFB7(DA7xy_r7v<}e@RvJo&Vyk=I-b=Wab4rO!QnC# z{4C?-Ptk5DR0D-Y2c-d%3GN=WTq0%C^G3E8qIBQe|4=MgDcLpglz7`mmQPea;-+cz-?VgOmd*-m9&jJ;c;46FT8`&Y!y?5g&_tjRb|QBJk*W zp|b0|TY5F~{DO?+{%#j#Va3`0<_?p z3A>1yF3|3X9ut{>4pISN2eAgI^j*CVT!AHAuRy8*X)igxzJ&ZroX^dIu?eDPM&xab z!m-)Kn<1*3(q>k1o0RL{*rfX1WF7Id;`^GIms!MhR&%ozFV@8WkZ>fbt}a@a9jwg_ z{DRuw3x(g{OS_*e*Im?jW1%&Sa_02tWUf%t$;~aNKPKw?hom4KKN$^BI-Dua<;(~F zTrEs{&#A{HIM@L!-cgj~$1ov>%L0dxCxl8T&ap8z8KWFZ!hqj&V%wjlU~UOH*}y>5 z!ni-T{Q8ozJMd3~o4iQqKJmA;q=`x*!9|V0$`!4i|C0&$@ASnU`N{?^ov{4kyAawr zp)y1tlc{Khb^M_HvxX($77kI{$B2D*FRPGqYE^N6N({J63rnq`immd+_Z$NZ1Q$y_ zzi})*~u^X-jo=ZPfFf2x2PD6-aZ9xxq_?jW^a5U>0!k}4N1FEd)MDIsoGvI^*Jq6>MT?@qm z0$1GL;8F`IS~Cg=Qds6K49;3LT(!d-T=<)xl#Hi3Jw5-?s2jf+_`N@Qrm8blIf2b(#^P`a3>FrU z)}6tfHVlPJw3}#xRh~lF-~zwOn6doaUN9I*d^gV*(_as0WODim)Z-{CHod$sn5B6J z)k-^$d(tq}7V(j)t3GS3C9#oinc7S1e@1Wcmyo^odO6m+V0zK;uVFrK*Cx`0Ao?ED zuva_0x|s1y910!cU5*ioGqLkxqbCgq*Z(2})j~G?KJ}hX7vmqSh_BF}!6t!N zHX!B+F`@(ro6}6oUQw8ql5fbv= z#?C#`tzY-0RF%WecdRM316SH3rJ1C3s8+GY-f{vi%KUq->H~@ekD^={5aa#nel86(JpKFa ztD*p2Q>U2rD6m@b55M;@aZ{KcpkGz+J5L+-tfzFpFa*^`{dX4ZuLX|*{txb}My!38 zeKs{!(7$4g2rXZ+^^j(vA|+hXaMa3Lx_ zrkapnTjJuwyOuqMA~)`OmEVaI6fjle^6cGN%G>KU(I3!<@CBnm3=>!M9B!PbH6w3Y zs0yYxPFP8FJ@7cARdTbqkQhS5ht+_mreht|omyYckYNNA9Zt@zv1%)5&mZES7i4jW zu?@ZvpZQt+tV?YrlR2xisvtMW3clz8L~E<$z6qxaTWg%!Vo;#gl^yXvmf-kom*rW+ zkDc!bG_X~nT=~^W991y@yL7yCx=2i}@tk&j zG|2Zf*#w4KV5>2pTQB0^z}iGGB$MYo%)x_1WFoX9gjGas&es&b)1^3bCL)2FGjIQX z5qJ8*q%=5E88gKTF$6erYoP<)&tXW^Q;&RdOI|U0RXTdNH*>JECNW@JH#K>9k8_=^ z5QJy2bb(S%un>r}-KP)!?=%|2_&NSj>qN!=&iW?-rZ(aAxJbOmD zR9X5W2FqDsE1YY<$WrLx*eLQowu5taprK3dGzD|W zF&TdD4~+c!?#QNOw_XR_^gHTjIyrA_2VrM;ZL14&N4U;H|M^7Bh+|}@n@gY?17^NI z;m*o0`l8Kx9!tA6K`OY^;1So0GpK6YJR6b?je6Q2C}fF1@pihtRpjtUf&h#?{>x3hvh%xj?iRDC6aRezcL)dd02I|Mp_`jc&x-2@&E znU@qmM>VOzIody4TO-$*P#uQf zu4#*DNy#7+@B;wRx&x7TY*ff)M_`f??{xtWyd^lpp~a}QMW+mGZ#y!XN$;=GQl|y? zJae{_k>KbhZ+e0$c2j=iL#j-@E+lt z^*MQb5>ZvZ*=c~z93#D7o=#Pr4%>zvNtcrzKn;fz<&*%?s#t+4r}wRi`{z7lO_f(= zK-)>?O|c78t?d#ZQPjlU#?aI%>XVamuKyf!W1Ss7O|9#bW@XSRsQ$`)z@N*3W=S?? z!g_MvOEJyT^h8LxEwVt7P|_RCi0(KUgj%}7$Y}bg7g=a9dmN>a9x=uSt8%Ap{oaTg zH>EijZdd>eDI=lddj?@|?urdS<`j3qWSqgn0g_0{1(>>4$a5!UFCQ>)7*T0kp?3+` zS}s;u?cBy|mzsEzgkWX%=fnuPcaWWJiY=fe0=|x@7we^SI6cC~ zE?a;{`B$+q3?F!ILwZ!`IdR}7SS;b|r=|!kZ)2~9>_>s&1+9>*$ZFO2B$f`!Lqm~I z-&={Vm=ES&WCY7pfTza;EvU}8(G)sK8jP&`m$CBR^3q_QF8HI+lK#zl;M6(OM- zhkXU+*#RNi;Vs03QaArRB>lm)9$Qa`R7$x1eI8K9>PRi;gAxRF3rC}2;R)>hCPD(p z>I=*l!6tE#?n>-o5Lz5#pULNKiD!zIVJ_io3YmY>fo9wr+2rK>=AL<5<7Z!m!ZVf+ z1Qlc~b$9F~(qk7{5f*k9hN6^;m5&B{1AlxXH6dKN0Xq-UH0+%gooqGNBd^51p}ro24p z4kuLcwiVsa@>lt4qrFqf{mp6{Fen$6|9ko2kx3~>r1Bec+YH!qz&GH4O*G=`6OIm* z|6Tv(u_&4JRFF0D?Zlx&xk4RKQgjqg9;-oo{+TrUHEvMGIQ+nrb65D%8*YD`+*B}g zXCz)vk>L3;c<$%2B=q{AOh!=+FCqsg+#{&ECMbRll>Ty)=^(0Ky3sb5q~l07U`2?x z3s829cMs1L#_gSF8?%rJ!|=eE$Y29`$9$ge416Sty&X|sn!>ja4Iz} zB1KJaBP`ggzu&>9ci^A?T*{&D?NkD6M+7qje|Gr_#<3D+RZ^#;C~@*IXzq!j8m8S} z&tCF37%x#jD-)~_QBIsN8Oh&XQIpeb!W?_jvV%=RZzF(FzL;n$()Rrq2l8x5zpY5Y zgDw=kSZk=0lbPWAJ*l2vchj~|1U z@0l>;JX2vwx7jcZIx;?RU?_9^An#6D?6p6yjC2$ljX8R3g&Qb=O!p2t(0ucI*9wX8 zOZZS}2ZtUhhJj`P9V|h_WVe8Yp6#d$gX}}fCEk5PJLU$51nKz$3AT6wYVadaAwhaG zO}jw{)2S`yBTHEqhlp~fqjyw~BSjT(_~0~BI@Bxt-R|RiGGL=L(b4R%q19`n7r?VU z{4d|o2&WvxKOu~*3ot6J*;(tDViy6ZXYc6m2JWQFz9F*!zE-1!O`L3}sp-F>{^1fr ziV|Kg8@Y(tK<9YiDTN-&uio#&pOzLB6urxnoOt8cEMO~R>O74E{$`AClqh(Wn^>SM zRYB9PRzH5=yVXtV5XH@|kT;XwF^NK@(5v33=)YYjB)>mo+H-U*7E~D%hAQNr$Hle- zrKdT_^2J>MnVpi_I;Y#U3ItMte|sTBR6~r!=I$;}w!b5?6zRiwEYtKgu8}Z~tb{FfL8;&04W}8l`MjlRU7P^p$M$f_U5!;{+nW ziN$J@VD}*T5{9v1*fA-eju0?v`du~lKMJJWE%PA*KBg4fjqU>Iz9=R$bH$hgDnXeQ zTV(=dyHS=65Wq4K*81Y%*x^G;SoCD!b|y;^-J+v@D7HB%z<%@Vw*C#%z&dMzh;@z| zNeL8%q_hf{>$FZNWU2O(?qe4)FN#n2(sF|^#&>~oz-wQ1 zsfiZ{ICNnaA7alJ4qP`ZX1KK~v+~&Moe>N|;eiD%JOll*u<|9pzUTSAjN<JBadaIt;C8-$$=IcXLDT;z$LzbQW<%HbC|wJ97|E~F z)kv+fyF7;J+dPYkE5{RlaT+y(AFe0s5<}<)cw1*PUe9sb<%&b4e5R`bk#Vwoakzs> z@pjN5zkuLu62xQ$sghR2Q*#y60@d2E`!vaO;2@g726*nVDZGle+U(|4g}gG?1tPy( z&5Yfg1B~nRv_TVtiY9beLGh2~inB~bQAngP(|qbHVnR`NesGn6h_0{A1+q(bf#f1f77F=S%4 z7X;T&RpMAQ&>_y#XZStt57(V9}=CZ$35{qUkL zN0U?^^+7w}C#oJqEOXPRzoaI7*C4iaBM5nO@t1g4{DWg_x?W^jCq+BWt{|Y6*vnvN z?KCit#jv17~8_G;V``NY*yX?A{swpbFjQe zS?_pr9*}6gh$g!c#Za5nG>$Hd-zec9O=CI)S0+LO1(r7DnSprZEbj%Q*%-fsu0WR8 z1M4wFV+x~>?y2BWx7C#1Zdw3N8hY-6w6EUhIv^w(Rl&Rw1$X2Awk-+)1W1;L=3?0vmlb}FIWVp4NOD+lXrG?@$_OaLU*zz>! z;5Ng8eoHPwaCkq23UU6xW^@-fymXCTG-hv*$rrvwM$3OA+2 zO7!ekPwpEShw2*epFE->A%70Mh0uy$2y6G5i1e}>rex(oZ@w{b zO>O19M|~gxYSfXuo9bbA1Zc9GQ2xEH^}x@-0a#H^5B_br3Gl?en?pBHL5r^ocUexq z?FUf7LCT2U*Aw2d_7`_<)5s3QAh^Z)YxQk^8yu{R=knXnDsXbp9%aH>7agENSMNia z!>3|Y>ALh%d#Fxfcmg355f3lo5^cTuc>d7Sfs$ndSReX^BZQu7_A2hke>N~z;MyX( zdc=O^(EI~3#pARQN5h3Xq45WtS6CDpTqu` zwiHk|N(^ik@F>-qLNvXe-M9FHw+;V+x}aJ5Mo;2XgiAi^P@gD^88jP2^HX+v?`fnX z*Z(pHc=~-#JhRq#;w3;})!)X%`e&E`PwC7}eMi$=a?>E=S+O2ytVY)4wl|RU`FHg^ zWDwLd1l52H$`nRt78z_p@=A9je^LrIniIvR*((`^jM@=7crT5_d_TkB|vQa5pHCa{OWlG5l)SZ&TS!$A>^^Xz<&wF$UcsU#oXBD zU17j#5k7fEP>#)P_3uZvFXm!bJw5`czsAuGfFy9z)%~^cs%e45T1tQ{Ah;m>c*mZ$ z%z^Vj)7z$`AY?9o5 zX?0x{oIqu=ka5aN3SST9Fik!uKfwruweQzIizj@ya{q)F6?GLU1)MAgxokjb!1RKU z;*kP2KcP6oJS=@XOWR7*xPt5eHX!+Q9YE^mXn#_!17iuPcV zKD0x9aDe-dE08AW4|Y^z_1On!_LTn*^{tsSPy;%-pqBVIY%g_fpfh8fPCe)MqXGu~-b?Xo8x00>nCs);im-MXZJV#$Q%Ub|waBu-E#=E|2P{ z7YYQpUB^;;ENj(_Gpb54iZbPK&Q@EF2Xueb|{~2VUdWHc3QHNlidfQz%aL!N{TO#y$1nZh@1vl?90YlvB`>PlE0m z?gKG8H3Fw?i&EQKQNuLUdzf88{g`S{$Kfm(&epGZOR^Y&Pzf<3` z$9r{tTECP5c#7vQsFpAd|1$D4km+44z>$#bjbYyV&q>sm90RnOm6Znx4&gEY@exoM zWldYP^}}-w5N7oQlqr4zNVPbn)Z2h%(CD)MO#QGe1J;VOqP9gDtUrdaUwcRZ=}X^P zBo}V|J$l^efNNSHJ{F3PpV#eXb3+qYXJ;^b`_Chgp4nZ%YY*Z#q83nLM@s=9r3m-x zXY8n!&_N3|t~~e(0|=G}vTv-ve+)0@Ob3+q@)Ahy*o`U^Wa}FR+AqEz{qQjdq|V9p z9wN;jclXDd6q3hJ;WhS5FATJg@}6TKj4|Rnx(z7G@JA@?50F@RFEWjv!nxg}uITfB z!nHuy1)toV-4KMn@)u6GIKc7~(xI26uy@|e?VuN#!*BjjAT0f?yX;JWki`%A6>$ti zpY;Q$q3a|=<8$z9=4(=5mxt^D(D(B8$?}?=wES;KbMbGuzzz^OcGGP6@w;gf{7<<2 z&bj7^d?~zoP1XL6@t`=kMSPVVxg)ssnUVcBznS#|ywlze_%vF(G2Z+J-WhNE0sql* ziU81W^j@Qs^oqU$(-oL@V#ly1V69a=Y5f(p0`#x;}fUl3#Z2%kL z&&R(zuO$p>Y&tRLN{922R(li3^x18mk)_em^GB^_ycZt|pd)-!dbIO8YXb)avgJp9 zEB!!-x?KxMk%aFb=HK^7B$GefgrhK71<6>bkCQG>w0UhaT+<>I?Dzg5?`4hfj=MC& z=gS2HYzWKT*Q&47T21dvQq;&K2em|dA9~-$4+o1kEMYByEF7YLNpVY*Ar?0}Hv87e zlr|5wAb~82#r$sH;3go}7}=StL6LNp)>L)O(}o3@rsFW-Rn7lvau8@aV32JIM_4R#~ghrMS3XjW4^kkH{`W@ zHTF}{EtDt$CgJFb&&z0TS}i%docwsL*s{*$gT+ zfWZmqM3#lX%R`zd-0Zz{>?vYu)pn`UiVh)AcdzJa6C$$>{n>OcdoSSu{d@D>zq_A& znSkqrcf|V&CXXez=W@3^mM;D1*%@`y7=7f+`umUAA5ij7>Jwerx%GiY&;v#}gIu8hR4 z#$<>@1%Ig}7nS{HgYkj66_&(Pw%kR-ohVE-D{i1`(gz(<=J(RE@S!)f+<<@I?gGd_ zwWFe%w2xodj2DsoJG%n+%v|0kx9{{0lD?6O7ERvuX;DgB5r~@osp~j0;0xw2@w|At zSgx2Tk!0vR8mC@U>_qZQNUl3r&#U0Bup>Tmvw!ky8B9-?nXAO8B*?+atN_v#%n+Vw zvT-7_44d`R`CO~zty-+LVnT8HdGw)YAt<=l#0hGPYodx+pchkEr|6MrQaSP@Es!M4F46_amyC$QYrkhL@18GOVJ zGd5cTQcJ5>ILyb`PN_i(62t!TXW#N=I*~hDI=#?BM1tfRe$jR zG=)a0_up3*NlP{$JbhgA%N=BB0TMGR95P|*8qJ= z@9Wwn=B8mpG6>EJ&o{0O<@v&sWJ0xhv7AfmH{!~1At0b11;K_sUFWxXJ06d{AkM2aZCe7Ldjjo()0C4UhH(;El& zzow1DPhc%8hc0A?*n~EVs>-@o${XDR9SvDi$-AO>5%P>>@I-{Lc6KAHf{q)RPa z@IBH!6WXb((&7=20CLA z!GTIISsK&Km`@`5U7&W~$A9`~o(w3uj&r>2^JuST(dsQS@wMS0x zM{~a$@{mYNemuW{DJV7RQ9YQlBSIufGX;AE%s%eM{EM~*4e#t!m_ zd$fGBamb!=ZCCkis()}Y4U5K8rsqdgQuz6y12{JPn_Yd)-&D~K$@74?I7XNcj>>Ib ziRcp}cDS|N(syTeF5LQGeomy>cR;7^_4lIRTT0n_x#I93E+dRenDdb)flZL7-e1B+pbigUITU?qX7b_6I z4Mjd><4z4!!yIjP=lZ>YUMXkZE!00wii)+wciu-bp?+HU8`G1^C`O_4+nG)6^ggEh z7|(|eh(5I{zeu5GmFW}p5JQ(7>@y$#KrC|Ab~OSLf-c&uHk1WT^(xs2V>?fn)JzA(*+wY-EJI6muk|3Okt>!8HG!{QnJ9&Y_?+Uiy9d}9Msz6rNd$+uAo%^7&H-{`Z$7v zqdv=Dnoqbr9>U{_e3cA7g}08U25gtz(+GyCSPW|ux#kXSd=Y*$SU5}lFL{#Uv6mSd z0#!;kN-0PX%r$GBdH9N`f+j4~$jWqB{qf4^@_(pGKd`vd+CSQGLbkPTzTIc4PoxXn zLWKpw4g8!&dU`=!H^@T~N_DC5BZk!LN>3Q7kdeP*FA$bXHUPP!BHM{KGg0^=AEg}t zHd$gZXb#bEcRatcw0~SE9*bq9eGAO(wbQ0}ZuN~4A7NZ7VJME``Uu`4E6?m?PkIFp z?|*<}_OMjwl68+Dgh2S~xy)o7i%P)ZM9i~%u3fWAB?o`Ojyse2iI+bJ)OJg%Q%;1$ zwPnqY`)Z1IRq*l|t}CdF-tdu`U?^Q-@}-L1vz}CtAH^8`0t1G=eXQmRsXyOuEaj0r z%imuRA$p8qOr&#EOTp$*M!aE%f%(MaHGgDXyq^_016}q_Ir`(3`6Ld9GaQzWTr6BZ z?ja|PsCZetiIx4*YSS(cN-_p{6d2UHBYOx)W142m-RBi9rm;0dj&E%I*!>Oi zvbNjJ^l7k})^&*sW#%$QqirSR^q!z~ic%s}zpI%~v;SJ9wA|6HXxK3J=E#e%P=5+r zJ{fJ^dpPAt*54cGQWclsE%VK?fNC@8<%17Cg7y%MHSshZZcZP+^pTHp*qT!Z2QxY3 z;@xc-)ThnUQ2JVVyupI2i(J78)Gmg__Kz2-y$9Ltz=_^?(Ge^qoKN|ir-OR@3MUUj z(M0(IT&M*r+T7BJ zY_7@BIspNg0CFAEn$UDNq+UI@RSXdge=T`?Wf+tB7eFfMm5jY$%m>+C<$qZn9A4ZJ zugHA8$y+<`0neS#JWJK3=_@%V)0VX{64+P+RGl>5SMx|d;xQg(1)|$C<-SX7W~hEx zq9ouhcV$hR@_kaehi<_fV@_Q}%%xB>+p(rHhz4@nJcT>!fk>e_>UN8&e3~APDeE=K zp~2VPH?zJDBX&7tuRMmn7k|z^(x>n~x;J~phDP$E+e1`)PjN8*0OwakM{0$6NR7pz9p~=$*J^#j+s?5FrV_#8W9O$ zu>27;683#hFvhqEC0*n}X5`o_HoE75?dFY`d@bZ>>x`E`w{aY@wX7A}SrJEy{G$|B zJTrX#1Rnzpr}@CjuzzlrRd2XdmTcT)fGFsgdXqkjbUN~pdPCi+S4wX?_u8SQ=DK3c z1~00Gf=Ru;;`HT|eD1<}lyq!X2 z;fv>7FnfeEv2vdfD;a4b1Oh&Xeh{euSH@ePaT(Lnj@UTl7w_~n zOithfFF!yive)-y|L%dTaRlX&;+Q0z?bI24)O+#6G=FkM|Mg4m3es=qL6$0JRh8vx zzj&FM`=MG=(?~S5<2V06b>?zwP*Y@-*Oq(-D%nj$)IDrfyjHnw5LGxbFR|`vFhynV z2Z`xUJ=3s-O8Ih1h#RjiM@?AGQmTMPSb`ZWwqT4V2Oz4j5ETwgbm;Wh1XatS!cKZ= zX!D8WIe$l|+P+x~SKAg1*sDqmA|iVwHSdv_)7<&^^~8b#hZgZAe?IRJx{uuyA`j%* zC%c@}N0qNrnW%lB1fpJOxo>77ROxH0(kY{i2$Sqlj&op31rvCXe71eYOrqW`s3-<8 zW<}dZ1aptwkaNk6L;%g1;U~GiMEM?3)5qR!B7er9jb>9cn%Wf}7?dih$meTzJno9M zVFAfAXcUI8{0Z9Zjc{N`b3Sk+QE@!6HFrLm^T4_^m+|Q%jM2^+V zk_fV!in2NLr;tir1fy|H`QY#d%@VAGM=`AC->{HfvsN=2dCqdMvG&*QC`!gT&TrNu z`54L(=E{?Hb<0rUcTLn4x(|&zxNmfb{o%w8b4&;stHDLi6I=32aq^QaO&rwqA)q?9)7A2T=YkrQi;H8RgYg@B`lp8y43Eo za69wg8rFxw*;lCwWJvfFgu1y)!#)ZR%(bX+b77BGPA(Y*Z*mvSW_UV#D9>Vc8yHF%Jn=T$( z_LLUIDBy!sZN~1L9+EwSk=utPP&5f&`08+&3fx58U3D%p#5&BDH||$g`+wg#`=Ji< z8`5EoT~!^^;Zd(UuxyBei6MH8*mzJGv(DI@a|z1^^RlUy4QrN)q(`kUzIevr-pn$c zaWj|0?Y&X4>gdb?MMLa*Dk^9n0+YxVa|;Ru`Y@Kaw>tB&Nl@Nf})(9)QQ1b?zDD-f4)uX;=%H4-XE8K7qf9k-eJJv^!$d3^&SR5ysZ z9BHmjXNfhb@RE7UabZ#;KCwMNnvC0vtlEWWD+`bu&k6@YwrfEqq$G*1#H)T&48Noa zDllLSDe%^LWC~QBURr=IzF)2~QTC}W!C8Imuz<9!sv7L|55L7Mg~avHH036#+ds2{_<`({mmT2 zfy~T~{Jw8nd$8;WyMK0P07EvrX!YIp=?qSSOZ3$rZq>WrvOW*3!LiTOp5TbEa>{NW zTEOPCG@Xd!)^~K(+oz6ZOv#vi8d7$yvrA}83V=+=8Pw&rH1yt($_RjN3Hut8FhP?0 zEllb2u3-4M*D{yu!t(ixovDDJ5@Ck{q|aVJXea+BiC*$2Du3VNwpQF8SQ`r}291t$ zJJ;xZEPEqJASE`cp93;gK9pkYsWpltFNE&!Q0{Q`#q%(<8fRO0702C55)N(c7O&QeGN7k!3fh{W^q-};O ztldh%{zQ9H=MLJSTBdbLj7&=P4T!!eLt{eQbvpsG_HUZKd zC|7Z9=Mfd}HmxxDJvm|CVJ1R}F~G_fW5iE@mOvnoH!HZ5Gh}G#leJ7*{ixB zxOi;mCZ5pwt%60IRzn(T=d^i=#r?G>Y-({uMit3HO~%P*an~{{hV3^J{$`rbxqX{cdM^gq^< znP?jfihPN*hjF%tB68ch%o@t|Qz3s~4?@G2)abOs0XL_ZPfQJX!x=xA-9W@Lh@j%VVi{@n`uqq2$(|#-YH zAi*}x1><;EVQPx<)AfRVK}R{=ksuj_Pe>a#*RfdlOAx!E#!a*E7e2@k{OnwvWRA)q z2%tMdzyE&7y?bS7m;o8u0`ppw44K zL|*h$j~0Y;AzYXkiGef}2%F=a+<#d-74$%-MokS6uL~Q1_$+z7nj?N0$Cb!$ky%<3 zKN3UF_4iG8*y#w=>M2>^d)*_Xp+12niR2LF4|g$B+w>FRe!CpbIsCL*xan++7)d7C zKl8-<0qbTJ0mJ1>Bo{s!&`2#irZqt(Pt~enRpJ0(4cXX7jq>$POQ2QKw|}-4)G@kJ zYFx=*l+1j|k&nRkW){^N;`Y>|3bF2X^IU0+?I6(`swx1|9qDs};|NH_TH}>atpS#H z9qSw5cAU>)s+kdf>F9TEmC;b0FZtd0Qh!gVCz9*UJAlxH6=u!n;B9!2ja!#IL-##o zTruqKR4xGL^5Pfg_M>mgeSZ+qwrv@iqp&o6Y%lFC@wJiLPXyOSrjW6+Z3!h+H_~Ni zh&g00hVM!vSExvK5zu*Z3>bfA+m?ULs^#lTw);c9T7#L-XI z!N)ga4f>IffGrp&$R;j`5*`WBy76FD{=nMf>dO-kwTQ(Jb+cETsf#vz`uds0$bmu`4%drg& ziSD}<5_Vlx4TMr#tbZwT+aCx%qPix~jKgR25G09#74;mk5W#A$BE^LQ8^HNgizXeO zOV)jUKnSu725w%{Usy-*etN9MP z@VJ1$BtUP7dN1`Vk5icSVZW|Fy@IQO@~6>Q6#u3~MV9~rDSxUx3x)@?bkQ}?4kR~y zjQf1j^D5SzqP3oZ37JD~?pS-J&9V$H#>A0y3t&m@k_bqoof{g^sCt1%EkmQM+n)+w zJvVqYDQ+^zH+Z#opN6yCiRSVlC;geZd^6??>q=4X2)?Yvt)=q`OVOTNbRqIjDL6wztL1_K80V&f_XoAxXp_fdCD zu@FPrS3hJGImGjYENwTR~>>`6I)TCQ*2iM_W zb4(^q=zlhaAU0RDkMRfMv|t);EC*=spmY}f=tNjd>K9ShqBV>)@fK}74%ikj^qDSM zMUOr>h+Sx!>g6ry4$O3$mT2_tC;{zP5a?4VAF~Nb7iF^A%9ugljF2jy?{>$Zt8PF4VaE|Ip|Y& z*cE2zw!BNjyDRWdEX8ddqmR@yyDcd77>Pn!8Hz`T=aG&&Pm)-TTjFvkTG0r%_}N3> z6Fw}nm=^~kWu+ArH_Tean!MH*txomS!Xd2h-Z+Sss!t-Nfx!?!2eIx&Rvt;2d>uyn zLVxDmXqb0@$GU0tqZYhGl#~b~+dvxb+6^;oX}zV51YSZzGFW*Hy!cQdCTqZI7)<_- z8d|av>5~@9S1?<*^f3vfZdfp(n}yMQsmUTeSJUHKG_%}zm|HgfVeP|^FE1(9XB}CD zElBKm_ioUVFopNOVvli(P&dMo+^l{u#DBIBP--~~B<4{z^cFYiL8OzdIuGPuaCPh3 z!Ne!}#nSFw8*XEeRb71Uxk!0{HR$2ijn)ohi(c#@^!E%*@0Bj!Tkmf3nrTbeSF7sw zP+W>>Z9qlqb%U+P?wHh#gz8>iJ}hO8CM|(dRJS|+S)i{kwjQm|ZLe+?gg_yV{(mTY z$u{?-Pf6rrgs@inqsaAp8{m4SmG3duj4RZTV;4Qw9zl7 znZJHPhVdCZV@>}IiHlMy3FD_Rr=J~=K8pwiyS_*G6nNAQFYg8Lz(k=bV@LgzhTT4$0Pv{yW$TV;ROeLU`C(jC6v5!W zn|aT`2!Fu8VD9mN_$>QTLR`j4i>qlhNFc}y)YH8^LR({N2Uf<4pMPK4huP>0Mq^d> zWm){OtC}^_BzGNFwRCy?kVWU9FiI#~k*5UiSz!-aV!zZ8*?`x+8o)htI`8eGN^X;# zxp%TdXxZc7<~-4t3_+cfr1G)UAgY%^Ta(;$D9(~JR7D2aln2UGAUqJlkNSY+0{h1s zEyR%Fnm?rg-Jr+wo`2Z97E8fznI-#;dL1VJ*V!C8$|l|Zyj}4;^dVgkS?z4S8jxmo zQ%)GtEDM8H{Odj+@zu~?EO8)&^_d?M5M^tO!KsVOOn=xM>31YTU?B|>J&$Ju8q%Dti0YI{%qBj}xqkFhr9w{x!J2;PQwvpyTJGU4Nb}h{2zZ6fMI!-=}Sx z*r4OWQ3Z;Z?BYpkj`|6EUF~}CN6x&&X(!^j@Yc?kQ!Ly?T38Hd@@0Y}j5-Hu3@v&F#6t=+kEz&Al=3`gF$}xz3u9 zMsQ!tjcZ)k1bl~<7*6%No9b|ZBQcTh*OK#x_b z;oh7>p?`3#b)v+_ZljV9yRYodEls1$c3ozk&% zXjSKa_?o?dl6Z@;c`_RDRByHMF+nXQ+pYcwQQQn-~93`0V0-CaxL|f1b;**(J(R7VHQ%fuLhc0Rq%Ek z{Vqn$oU=;&PaXQN$%-+<*m4&oyXz_5#uf~0>&d+u--%XqRuEBV$7rH9RuiNDulw0?1Ap966C+s_IlocyO{rBCn_65QHN^x?&)`>a zCu34qzHb8z2~vTtYn{}+=J$4qV2YQkF52)xalKjEN#_h*86q#vRGXhWvnT`%Fa+S8nB4cQ#-yHX?0{R)<_n@Jl2O7S*79=bW zsZ1{wNrWm}`au^(J>S#!RW)((On;p5T5#<|@+S};;Fn27zAUIzjNz-`SutOwYu}PL zhL=1kH>V?tG`xwCsDG_T=`)LMu|(+^IwirUeIG(9S$>VFMBzZdt~ zfIe>G$kYP}TXdRIIZ8>Z;W24s1Gk!+1N=HTi|qSFu*23Q=W{dulfz?tc$lHEz&5p- z;D(1+++bBp!M$E#c;%qBZ%%h608#}x+CCaRK>A1UAQ_s>eA@&maS!pYrbINhObYc> z<-h`s;>5`#tinNmZL>sA=YQBNf(;4%j}NFn@U?BT<9}|&_Sa=;OB~Vk zxe?e(s{F6N)xU&;lO=JK9ivinS!kwIUyUMK4B+W4o4j>7Fc+3A<#T;@Ob_f>t_>cl z792_e#-4Nlk);8J<}g*M@v=zGw?He@(b~K6kPOTCQK^J= z4R?=LxDZq(uP=N!Ef-aaW=y$|se2kb+ggQ|-4-+ac0RzE*9KR&vF4x1E=gG|GR!;4 ztC^|w!mgiP(F>Q*JwX!}3fSbZ&MahrA)avX0@L?hL$H|LYjFf4BysoCz7#d`L2 zyz7*tqmiTi7=?Gx?OE+7?=jTq0O2MJHpJF4rB?48E()bpXM@gOBU&EOQ#f;}302A9 zh^ZxOI0^Ta`+w68BTR_$lH!)Bl|vu@O?}{2ZJ!;B1OprD9-HB27Kq@5>La3b>&6V^ z<0+A#JGog3Fn(yrTBI22RCGdvDiD(i&Qsxfxv?ORW}T`#x0YZiywEF8>qpw90ny?9 zOXq=Mr4}*>7ePa-o%6Uqe36Y(xCL9_(wx21g?}O>Wq(Uid*B+2iDjPH<-jX_OC=}1 zvz*+yv4;1$VY-Xvt_zF4ygX1^bXB(=^?J*n%luZsE44zUA8GfDhY)U@hDD;e{s`4p z@OQ$`kD}pIU&^hyOBi=Y3qM5E{y4HeYV6t)A7QopET(UYiNcgv2B-hPZYwo+t$|~& zLdiMY`+rK2p=?IGY_|ZOsR7jewu925wI|hhB!~kqm3i@)9nF_#>j+)-VaIwaIR@@N zK2^Ve0&;34M9KouycYcTwNPW(-5p!-@@I;X>F)7$aV9*5_4OXtXo4u5SJBUAd@gmn zpRz$k9hFbhb_P19wQb5z5_Q16Xc*Q0K;T-@34Z}+tV!}y#G%?IeRx>;F?Q+364t_Q zq^q|ImUaCSI24eGZ+0Z!Wf&qZk3sZN>Lvs_A%@W0#?O0{D=gH(;7iBJb&B3kJ&D*K zU%?l|*`-2KA&YzHn2wmDEOn!fwTQ3d2X7iJKFtl)D2wUi3)hbIVpuXEttyB&X)vUp z<$wHSLE%3?^?~r6xzbwFX6aq0=?Zlva$0+V0e4)eWPDq9_~uhBLbbX&am(ec(T+CI z4sPT~P)=s+9i*VdKUyI8(pg1pnYK&caT>rI#TFde7k)i8bZV?pIw|Uwh zA0v7K_IB=pWzp-AH5aGGVxp7iIWxaqJJeD z&p;Ty{gBwc{9nhPj+BM%yqZDvcBc8i*2`GI6}3! zj!gCzaB0Ivxpftfz&W#J&D3RaZ;X(6_(s~IAKrykH=Js?#X2&}WE7m>v6OqtOhSly zUjJAP;aipB$>9_@J=irXZ-h3!bs}hQerF?s?B-xa9sms{lNa(>L`-Iv`+u6{4WBjW z68uauc!_pnqf*{4F4@{EJZag#~bizQWpLpJ9 zWVsd-NY>A$*C#Q17RrdUvCGLaHz0~hJq{h~GVnU)-W=GKs**`)K)7D4t(?sMP+Pe3 zh$hS|jHKbntCklA6)hkQ(q_ddW6QRw(Erw&E)ZFhqDGuLlV#u)lYcI^%n$AEUh9(d z8db483~6A-!{-_aXi%Q*LGCxca_aYi8i@H8$kwiKATs9(DIz%ei%0ZG>9jTWo26eL z(v!vwlZ6}{UP)LnBc!no@x-PNeWCr2$53KkEkmbQIaePDQ<^EjCzB;i#r^tg>2nrX z4+x}qp12OnPrfB=q<_s@@)rs6%Y&cV)H)S4T{xL`36OT1!hE!I>!wl?FG5(DQRth1xZ!hfl zg+XF-KX}5&t@ye*unV}J(*OkL*~j7Y{Lfj$WAe7Q2dVS)D);#<-2o{Xq#qd(1ni2j z6|6}8H5R1KhpmFyU^i&rkJZ!`de`)u&AF&qcUKewT#!EG32%EuL!$xy`}aF z@w0d(_|L>fiV#WVY}|+Wyni*0&O9*}X-rO^KX3koVtWwY z8%#=9Xv9D|t-#L2z}{4u{tJf#Jd#n}s%|eNV)5kdSDZ`m@I|IpJRiR$|A$}+xINAg zAmVXM^*B3|MKvars!m`|$-OfT($S?wINmR@xa#2OM2vFE5;GIQ8#68u+FPllZ{BYa zB0ei}>VJ&Y7A<%ln(_V(nVERQM9+_GZ=Nu!jOyelbtt*YOiZ--qN zw#Y;%Re4o=zm_Ybd}K5g^!8%iD#6N%6UZGZb-m}PAxLNsP~`7c@<+y1oL*ZSClslMsv ztXidd5v=PuOW9^wni3c=+`yNIU9yEU7=vUYk>L4tB3I7z0%@R$m>+uGwe1-SQuizB z!*%a+g9>bJSFU^?@u;=Q8TMq)s{W)Vqt9-EayXObVhU2!0e(f~aRASv<7?vUoqzaU zlJjQ236Ay(1&brzr}U`Bbu+zRVWDht&)}Vebd_-=(4Fe$dSX_Hl>5WtAEd=jomudB z7?StyEs$~TiKqfpm|`?m(hJ>D;Gp#JlZMtj57Mj%eoDn5nN39RFp)!sX18M7!F{q8^Py_*I^2WhD-?mOpgB z8V5clg3-ZZwQ$C66pIC=qiQl*@eWQTPc<=4mJk|gvKz9zN-v*&cb{M=l`*(lw5Vyg zlwJO%l5xD(VFA~IgPY<0cA`8}D|C*<<-f!efPEbFsE9er9gsHO#8{FDyMObx*iyuF zRqqqGgZa@Xq`@X&vH&p$$_q-y7p*^^d`VCPZ7(+Q+dHBWA;tvRyNo(cn{ zi8|9CEHTm@BPL?rvtGx4X?jgyhx`0<1uU!{qNqyz`RpX)G4UYwOeV&_!Jm5%4@Z;I z2P0k7{@1ll^zx@MNL`-1t$&R-^W`-_4;c8baKZB_rIPho8-&UVPOz4nFSX%$hZIY! zBgwOkTLkRmW?`1b-p$HUGaBV3L3>K$bWq*y9ze_0qpFKm)z`o>kJ5q6 z?l)Qs0->iT8o}z0Dt#EW~1%Gsg89hm|bGnZ7?_OKRMMGy`|hiu|mksH*mXqqATf@ z&J+aheb>WN%pc2igr=3h2c5*lrbdEdnBYNZu zI3SMKL1tVcivzzrToWmK`qSm;bIV{y0SZN>EVzI30k}DM6n|vVupDQG{uz!ogdbp# z)y0MCVX5zDt-N1p{miGA(3HKzATDi3xQzK^3o*u0k~}Y3q1?zsE{pY#ZD!_~-u5f{8zka{^SPX9|?7*pfQ~iUZ7^7T7pp+aw4T zxHdB=euy;}*njK0)NmC`t5Em5vj}N5i#+A_FX)Fi`&r{P9_}h$B+_7bZ5!j!uz8w= zi1TSU(D;@qx8Q$SrnSdhfsRpt{<^zy4wX39t})HQUA@Q%SxReq!sHd+=DFTN+F(Qm zVCNWtd@d2?O}CK<>|Q#w(Ha>-u1ssa(I3X%8zjXv8GlTkYu0g#tj5UOmaKPo`}I@o z?UTQU8%JrTO_uzE+w;0Caeb!Cb)Ilp1J&>Thak!g648vQfvj+|TLyhfa{XsA(j*(h zq4gJN;OYS50jF;4(-MTHu#1Cc=U2RVIE37b$bl*tR!{Q?hXiyy)@mI#vIFZ4wO~lBnSBfPFVB5f!cJEtZL3M?hTn@hzX`cTEKVZg^c??bGogX!fI ziU^%+Xv@>R!NedREV#8DP$?i3IGSmC4X=m-sees5_IO@sLv4w8oGrQWNN$Non=3Tc zW{_~hKq}gd`Yq8yiBU17>+SP*u8tQ|O^bH=S`cYHQN|KVKsEL%J_VDw)fMONUv z-l&sg)U+69WMVu%!<@))SpMC6G42#u@G%05`9d?OW}YHuO%trlNoR6s`ISH+=rd~C zj_v(fF1n4K9XA9jEEqyO)`X8?jMgfb$s1@({cEA6K9` zDj#`bn19UC%mkL7AHhh29a(zxL4W15B^&*I{WexpCt1KI7|uEV!aemdL!~={jr|m& zfLN*xUuRrXV;p>Mp_343eWR1L99*9-azQ*>1#_7RrGjI*P_r<>uP$MVCvAh;K~>kS zi4GAVVA}cXwtOsn+!l3#Jo!OkYCh(z+Y=sDZq_tY*No3>L0&@pF^hKPBY(_xi{Pwl zCi0Ch^Wi0LLd$QSBu0DU1(b~S9}5C(j+hD9iXrhGL?N6TC&YJSS~em@Up1SOf_rY4 ztm_uf+h?T31Sxoh^J-v#3F9>0CB@RRx~!!I1J3$s1P#Vq1O)^QP1SF0|t zv|0euoACi6>flvI;iz`BxPOG$#x*(CvRm2#sES12+%e8GhjgckYC_0kPr)Qo1fVdj zDsHW^4u9&75`)539d5;`Q!R7tQKb1u_rvY#HS8aAb3M)3=FRdrre>2&YgpD^!Ft#& z-NWL7T?;Zo=wm9u3m;zZEb4mK6Luxp?`(`}8Q)cM@aDO=?%O7dZ&S!hX zwMBg3Fl=xAoK@e~IZgIuyRq>Asdpf_vtmj4i}i||`lXz=R;NMXIaJV5NK3_%@lfot@$oX^1|H-<{|bI&g= zu_;gD4Zklf{M7uekAL4nnrV=QgdT1R@ztUq!zgmYAdl&_g6rEu^bh?uxY?j?sZt6H z@2HZIC#nxqt~a)+s||K3F8UelO1&pE_gu=tZ)S?uLZwWsKlBLm^^fE3!?=?!X-gG- zuE=@47+_y22#WYAtXrWi>R88mdVh9KbWM z+Z2~kdT0AV5I!JahGyE6ComdwJo4K?uW%!r%<%N|Fr<~59azO?rPtiT(asutAk&%( zi%;-X-XLmIfc3@7ZY%!pElAvf5;-XwWVA&cb2Bn)q8sN>20_F|M3*R$dhRWgPttqt3%hD|t-q=7k05oBzT-l<8EU&FPTk`d6#pJj0(Drx ze+g`2lqT5JNe*c5mIi(ihPpSjUGMO=F{urzcCtl-TWlS>9MfwAeHShDLej*vUL*Af zG^kF$@+ROzbw|HM!(erjKa4WXbB^e1A`1xqn#2knQHMMGiETw!K->MAio? z8R?UsxYIsIqbYXg6a|i2Q&7m zACh*OTsLbG-9B0Am2UF63M9vOYqo=V;i(tijhy~iz|%sr9y5o2$*Ganf_{mmk6 z3h56O{3u=f{!c{eAD>J<5gJvAxFdtxPPV{5mJ{fnkF`3mcJ_CRmzBE5H2*-7#4a7r zUEy?IfD6i>-G3J3JQjj5?$I34BWaOC!@E3mb>bSYBCQ7dRc@wy-V8-N=KobKd+@uv zQGd|{6K(WHmrP|liQty()4105^$6@AS?(f!Pii$P?;eds)iFmRNyz=qT`cpXLde75 z?07(mqIuRyQl`_K|J$RC5yvBpbAxcii=ng$E|6|?B+O!}o?)bNSfI(csgUHdKy%+l zTqmwNRgE^N+@&aWwdI;#L69SeP)2GW=6_>iPr`19b&9|o9GJCf)u=@pH!AuxYxzXW zyAA{Hrv=4msZ1-cQnZZ{NL4!XW~*thqz5TdH2IekpHH zYq!@mQSLL;{0My-Xd`44D$&aG=zr5Dauv{$HI=M!F+Vl>jN)%PBEJE zs0sgCon%{B3FkSxoaDAQNu z-TNX9KrcM6$K7*lu*8R!By$4p6;_F{ZPpp71VQ4wLF5(7>_tY5vLaZsnN@UQPrOBh zhO;l=DJDI~n{S@*bKO1-<66be6|-akHVx6!r|k5dowXChxwtum}eoGRkdSA11(}mgZup0tTc8gnxsX93MJ1N=BZIRN0 z=TO?JJGa|pFkh81{>hVnFU46?WIC|Bi9(7Ns}N>W3!%uBg}z9+DTJ*M9rYSfv5+ri z4Pl^>Mm2>_mINDA)cP}RF58=JNUzeb$S##R1N@;2({pQ4RbGX6(0?PAnEIAIj!YB6 z$Y(?=o49o#DV~KDG(e9zL^`QPr(+EmaK$e3Z1ox&N>xudVz7P5*|4SzQWkhzFzC@* zTUlWEB)u1?MPE~TlzW#Z%z%_UMH~gX>&Oo%wyJvFV$}VItv-fl0wO&JlF8?f9}8(u z9=Y3k6Kt)czbjB}m4CXKq6|MDUeO1wp`?@wlhU(WvzkHvYzBKxK2OdUG$64sH9oAJ z%8&=28T1*MGkRC1a{|qIJv`)`*xn~&R`Ta)Q;Qw&VY@q0klV%JgQ4Ig9IIN?|45!h z#W@S3nhoW05(01|lugn6aBRzUG>VAKHXrbAm`sMR4x}mCrGJFMENTI1EzsPypz}uc zki4Dlkm(q0cQIS+x5xTSan5>I4+&Q18?^BJjWFq<-F8fv31ACC^_}Z7V>u zyNhJb3h$|U$Ge$;B6s>okgq>#)Zxwf`J`%wv0sp3sEFz|$fU01; zsnvxX^B_KpKz{)|$l^22^S*&Po+VWhD@LF*l<1WSSLrX-TyxzP1kl~+g8VB$F^C!F0WP?)y)ABhGG_P?Jxm!Jcm^q;}i zz$cJh4krh(*?JEwOxjg++~5PdWB5q&&&x}jM3~$$5FDz%Uz9Xavj@{S-kQz7+HkWh z2?#$w)3dH4E>fI{X{87Bmjg-}Clnmwv^v}+tFkY>-@B5Fz0DhhDu^HtyW#*~&_C{K zG>dz~r`wjmr&`lXYGgw;&kK$^l1rW6sga@hZzk2$?L*GqxLiX%=&8Gr%d;m*?LK{0 zNDKS4XDqd;n+G?o!Flh>IW<^*a$| zVJpzrrns7}4+<4p{F^g_LmFa3BXd*fLM-lnqAPasm^n!x>mdAa8ivEVfi*IEP9L_B z^U(z{nIyl;c>QCGA=GTPUy^BoebPQsPvN8c{N5R`f>z|M)kuGP zw|T39)Vxl40e+Wl=$kZIjE1s5_b;5{K=yu^=xJgv%?Mb_3z@)rlW1)r|E482;hiT} z)ojv8+kRwD(9`2{flYY{!vEV`+PeVnZTEYJ_`iQQS-jE;*Yliass_+mY0^yV5*sPR zlMJ8(&zLPnapr9k{51qP`v?;C_}C!77n{k%93sXfr4fn8eng)|4q90QRFM#9=`LxA z{5t$cH}vX=7ci{Q|bL`-p) zu6L+c!;dy;(f!`4LOTRT31R=ruspBDK+~ZN%NVV#Ywr#|V3Y`@;ej&04WBoN!VtPl z3a3qLGy;kDB)U_Lhi{&5Hl>KtNw58dw&6j^GFx7flDM9UQS7fZu2IYvE*P`52cXxp z6uv|`rIG4XS%*j{4EY*j-hvyljRO?GzMYQYp>z)HbOM@1Y1wZQ9#xUQg`>3!3YhFV za$C#hO9=PqY;jA`>+*;x4IYf6&b<7@N(-CE@Nd8d| z6+X16WYsi*BHBM8ycFZjoz)_j!5r=+DV|h)R`le(#aH_V)olMRMvdf)eLty8%u-?P z<|JzvQazeMHK6TD5)DNq5sjN~;7O6m_sK5q263lbX+W^CN$^g9=~qs9329^1W@k># z229Vw1oDEzA`z!Pq-lbT)3{{I%nYgN4Vg*K4YkQxJlC!D@d2kZ0Rr`}*=UYyppqjk3w`fmq)-f6KU+y-y+gtM z1Jkjd!kC?_u^oKBHt~za{Q^_Rn^q0;1W-nOsVc?r-3R-wK-MZr_uqF+x$r~5?I?;9 z=BS?thn0FZ4yly92V3>Ryb_4_vQC3112BA`hy;rQ^Jq+ouwWc5un{=sC-Wl>^5Z4y za|ToA&HNjZ*)DNoybRM1wtjKb-yV^^KYd^wMHB#9e@j@6yF)7vbU%CZuu(_!fiPv) zs1~cYb079*zg17?WS{f|_bmzpZZ+R8jw1fXA~yqiizC-jHUpd5(}~7i$}ddxG>hmj#8<+O_6Co zJ_lwQg+~7rmVVps3bgR}a;7T$_Guzd`NJOM z1%;}AtTdwt-^hqa6NwMU`nl#AK*dH8S(qL*M9@#^2xpInc8TZFl@~3CA92M%+IEPA}0|l&CEDa%z`zHjYqej zgu%%s2v+$r)rBlWFgW;e;3X|rruXlcAGu!P_{xi@E$r543>Z&|`9HHU6xF0_`nM$q zd&Nm%xp#Sk0R`IL4>P7z-ehjx{<2h52gP4ABU*D8<+5ZhHKjMvhYe;Tqg;XHq_iv^ zKw|4at((mcGq}5rdS)w_a-|rdD5S!#_wY8cpASf4;fv9xb`sS!K;vT~*5B101Ksrg zbS!U*LGfWB##b-mU%o|Ppp+9V)Q=Em^q-P?BAwNg{cnyDBG@sQkqT@}GuYC8loV7p zVDM31ZV~8eSJ@YBg;LHW&=RF}zmE{?9UWK%1|dPkY9O~#-wIONzR-Z8y}gonkpVb4 z;Ty$;uldoU^F0_{z^(E@Qj<)*fysTAQ|~)fe?W{EHI9L^y>d>M5Twa2NfnNg-$+3f zme@CgtLqwnl>ha5GH1o|Poaa3?BSM_={Q0wjkFOTG*wZTE0Iv9kC@IP{+etOcJAa7 z;?l(a)suctQ5B0a0qN$yZ!PD@8c}#DLfzT&&cS*V$JvD|Nk3XnpUaSq9M<J9&-IA?Ey-h@4 zAUZKp;ns{TPw6Pv{`5OzSJO|ia@UyxSAh1vv2^+%64lx~1mW+7pOg;UzaXu38Kos7 z0)9Od>%~$Z0S1CtOJ*$}#aeZ&y;+@%4yJwM*cfC8Xs4nmqSDRVJF&fG4V&a<5&iwb z*XZmUoJNTc)0c5Hd@%#Syb|`nxt?-sN!{RK)<|v)A}%HFEc8sqhRzCg*oc1 z;#u=EDPconm5!QtYak;09%mG6{V1}U91q9x_+_7KDdCe_oRmxSUU*`iC+#D}ccWqg z*N{$~Cwk02`kDHIIvf!hMUaB^n)T5VQnT36!T^zqA>x zr)-WA4s!7TeA7gFMRzr=?=eC@<`1E*L1-oH$dq9mRcz}?{`BXp{|L-5gOS6ltH|1C z&$3vAYa~Haa#BW<16%A=PzXslC+e7Xp!b{2 zKoi}!U}@^(`x}>(aYoV1G7Uo$AWK8yQFDe?mmZ0$at5`qvFw+I? zEuCp1ekG<4*=ts$oJ-RjjvjyInZ{gM{T5q^gvJmXy(nK5Y>>T(5au=|fbf1&6IN-A zSg8qty~XPc=-00CTKdqn67Dc2iZ_Ac^}!_P+q^n(ojVe>19Mu!VLwtPU}yXRAST>n zr?FovGJLFUsGl=Z@hk|N`s!npsZa&T!LzM+RmGiZvs|*lm5zkSg6OZ!8_dy)Z#F6d~z*a#0RWktwN#a5)OT6pWu1O;leJK+p%@(Tt(sy`p z5h$4$aA*aF>l*#KW6(pOJlYbz=nf8T=0BZv5L=nqI>ng;FXDp=|5s=-Nq~DyyQ>^r ztD#CrxIAA9@^VJW5hUS|Zw^`M#7@DSAvz8F`>|5e;nF?{BFt{fSwz(fi1Xdl>lGwK z%0q-~JyD^m-90qCiZl(8U)hV3!8UMct`=M(`61N|y0ouE;)wdgiJh-)@m~BOoW-E& zLnU@10-#jPovR9^O+_h;(tpU|&b_YN6*kpm_>+@Qzm8Fs*w~~Nr z_TT;x`dmRz?#qE^x2W~&w4y&o!cGkcu++y*BnSw0nK2wIlP}?7Nr_KM-g(xRKcxzn zo1bg{SU_01_mdI4otFuBzvMQmiOg=?N#7}{oVBWE+RTQn5S#=7qBK5-XSf&K9o-R( zq?ruYSU+jt9L8gCOeWyFkiU|1%jx<)d-lM*C52&YE!*b`**WN~C+S)x`>38?h#v#Q z7Tk&~pb5hkh?!q%^*t{U#urzurm7c(uL6&OpPz55gu!>L@MTN9QcH;Z2hvWa^!pNS zah39kPFQ1`#MF(zV*7#)yx8!vGJM7?EyYOH&k}&pW%rO#H3f zT0s;LL15Rh(oc`F-*(KQWQQMlc>?>V>K|8bq3OQy77xp4-)RJ!+MZ3;AlqXSuQD>N zB$ha|(`k67L~$)2w77I`&uEtE zfy#FI5kO)aC>SQooc>orV;aTv}K$;U-8buOwlP>ALyEY>}x9WH8ypc zD;$x^lp6WCHiz3%{|$at1O8i?`Dc2gG*|_cUuK~q1U54qO1yFyR(&C;MeHtr`|5^D z9Ay>je`s<5eS!?8b$BYe&jOW~&)+@@r4>hyr=!F9Q%k1@z6DYh_mvkftyc04F}Woy|yO^~HC zHw{|?U*0r$RWzqH7qnSRdW$z0&PhR`KZ$AGp5JP~Eyh%6b=d1lvuQ7*gh>rjlUEq5 z69SHM-mHmh$rCvG>p@*!qmwExS<`_g4XGMNE&7|TlvLNL+YOJ$`%>?fygQ{+A8~DA zb-Sr{rQwdNGg93aleDl8O=4L%N<2hYCwOtd5|o(COQU@2><2W&N6ma^)xz!#Ze&w! z=>RKmcdO(Vo9}47_#!8ECHH+ARV4|P*wAtLf%)@NSW0VgG$ap&rCq|>{o*+ zgIA!veu&1^iIjE7$1rG`-M90%;jh?0KW=U^gREdl4*UV0v z=7T@M`#C`cYsN=(5PJwn-!hb;?}vM_SL*Zt>~sg*W)0oD-ygAU9%B>gtn%riwl2?E zaf&k(&Ig1iygMT?1(@?-jA;qU6$VlVH89TwyHTf`6v>bSWHJT2r9TZ-$4LJA_o79g zbAB}^uYND94+pkd+X8{~QIZritB*UcD2v6?b<~2Y5o97xFX)2BS=2dHtirKQ3N@8L zq54blfdDnTj8Fd9{WKANPztxV$Q2}vg#cIkH>A4VHqZd9A$X^ns9_B6Bbs(v_T}dP zsBusmF9bbN6wEx5J9yUG9}8Uxfd~2XPVB~4`r5k$s)CnT;acS%r9sqKc~%?fqaq~& z;c?3(SoRtehVc*4hw0;IFnpeC+uIBS80~pKkD85K ztwk4&pwib>$bHMMMj!cfu?5(iPXQNz*-X#O2xIjRv-Uv;V&qLL;#(X{P$Qck?ud4? z5ly4j2EXe;k2tr){-LAit>f$*gB3l3+OKuT; zKoOS&{w~__*?zFV3)A+F8$x0Bch}mY+)skQA)>|nV93rV2eDC{EZAH(T%?N8{Fz#_ z0yw@9n<*_EvuSZBg8dp8@HkFPWNB`Z=6QMghowB~s=_!qd%f-ezt6fPz@1%Vne|hb z1O17u51$G^;TIGm+}|{lfj9vIkFg@U+5V(N1=ay>_{*co!Qut(!{{v*$#Tz0ZU(tg z*0FFg)=$|)UTZl6aSc4L-mwo}J!cP#Ewg{81t+B-MrpzcoOJ-kz{1ysatG_d!eOaR z8fX%{zQO^@^Rw?#WV@jzl;{56q(Ki-+u}~T^L|~;ul!Y7nl`xJpkKBRWj7NDC7dez( zzxh005Hb>&nAwXvNa^KN%OR=ELmp({foYRGt;aB3;U4td2-IoHD+@xij2SnMlso4i zDi*V+!~m_$X?F$0+N>%BR1_@$gKqO4L^HhPW|Rr@^l)l>{U@Mq4?cz38~k|={r-ur zb4PZUvQUebvBl~)WaP_EH_o=*V{L6z9L;@}Ew{!W(yk8@@wPWMTwv6+PJpm{8G6Q0z=KR4CrfnKzP5(%9YtACoiV(rSgb zPP#c|zFm}-Z*f2X4u!KI1>wjxI6l9**9_(wwxa#?#Wyx1-&wieqsf$=G1_`sj=#Lu z8%d<(E%B+{C-t_SK;=U&fYYdfQT*1b0)w+qGevR50YT5WD%)+fU^LY{(o<_0+}A-x zyQxH}EHglANkP696L(9VBR(c>i}k#&)Q~owoNXkHsR;muJ|1|zp<-~L0ZVN%Q2z^l z!=ZYZS14e#jN>R11;%zAanFn#EUDDo3`<7>=8HXfCv8YAH$V{H71Zr**qtV+Xbl#TWLLgyzc~s|5&JAiT>SUjk+yWU6!4IvCrzhJ_S|cl@sQSm9^ilKA zS%t2?Rzlktf#mO`jS-(2pYhSDKgf57sdTX|3v6FTxyqy^BMQ}uKo)Ndh<>Ha* z+vb!mF;I*HznByvu+@_Y6X0q7ppouoP0~^aQq=D5(7I-^C)j47UAD!srxeAEKvf%h zC;}WE!7;msf;MBN<&mm((u z95i|TOZJ{Q=E6!4gmG%yty668BzZ}V1|*9(w)q#GS`rc&G9Sa=vi)wHZFd>iH{!l$ z=x+Z2hhXnlUBtCb+(mt8#&Ik$+Ez!W48R$SyC#HP@~Xl*%}aeaK|Qr9qoTBC$nQUM zyA@_XXl{h1$;u`u4(PxRh&~62>7UIpC3HFXkV^BLFw_DehMPFLGu};EIAK=UqG`dL z-K7*Gk~s~J?mo_L;Urq7;N{Dsvy}$LzFDvsDcAiy`>ym^JRY$|ir)-7Y1vk`ngBd& zt@4JT^CufPa3o7%<4HcXC!=)C->|Qco*&A)46Zc`6kFOHdG)kP;LTL;#S8qSR&3?2 z8j~sv1EdNB$FDW;4YSH444k1ZlYT~uX_|P??c=twprVsJ)cxvzPZ>=WKiFz^%Jj@z7*e4a z`Zsi0Y`2tYbn9zK#k{Ho^$@sw!cf-1cczV(t98A@Yne9@PwAs&bv^di4uMBF`I`vq<^ zmQ~2nzw*BHtZ4s;ZdT)S?Dq7|_XI7Wvf<~T5|!iij=IwH+@9Z9H!q9US7D|KNi`OltGR9hf| zBg6Giv{c{P1WL}*|Lc`q52_5Iokq+#wy!-`igFnfY%D_rLq4_r+AA`7&Ql8)beA-s z_aIKhDz?=jiRc$|!6I#;THvQGy3s~o55;L>{<*xHfp;5E*AJ=YUj|G)h8{lf&d>ji zvJJlm_jFLJ2m64NQlR%}W)*xsuf;dg+AD|ETTSjWVcL`wZcz)Y#&@FmzG%7R20J;8 zQ`GNiL>CJML@M=EmL!9dnu$?d>M-6<9y8ekng;!5(m?y+&?8?NH7dwJSmBrdo#s{+ZYWUUBAatg&@pG(-y?jTp))RvTq#Q7sypIR8PX* z-;I=VrR{C#K7c2NRP|v1X_@sV@VX6hdz0C!pq<|VFA+LQRd1`VH zVc}Oy*2x?`rvsX0mgJ>lE_ymX_znT5k4k_ECdoEK5g+aUrBdPezf>w*94v55a^?<} zZdR-$Y;4>dKmVsv!GL7r;P}b)|6M9x=307_Z4799LQWB*+uOmmP8oabNOA1f3XOa9 z!og5dPM(O@1KSiRTLzW|z6wtQ&tLw;#O$W$S|t{Bw1t$K%9dDc&{<$*!bnCI9#}s? z1A~k4#&(8-^5V1+tDUVyn!=^Ood=BplmtW9YR||(6YABB>d#!V27t? zcJ|hgU-}^_S+AaDyX*V(Al~R~x{2K}bsYV}0|X?q#baPfV4b3|Di08#SHnPPX>Ec6&7QBl7Ar`= zI>t!qVQJYCMmAgLh_ZA+zDn0xpgDk?Vy z*rf|G2mU)ZFcAK!S}lO{!`2`mXC25P6?yUD4JH7^qV!-CGPH#!IRszd92rK00r}@N z@bCS;exVYvIDyR6%?TncL0V@V1JpBI8$m!{zRkm#U)wK$@WAVaIvM2kienOJUd2G} zIj}5sSO_3I7GnFmAn(brw^aY?GF9#}$n9Hp@f(T#S9{yThvg{{W(5xX>!t?d-ppD1 z%$VH}-`D`Rsy9LW+5DE#^4&BUBv10dbZle|9ZElGNIAW$JAuU7o%o#ac(aNfn3WfE z@}FPgFZO14kgg7pj;}PRP?$TzSkm+FV|b(GIVE5vS>57e*YrU%H#G!zY-?ixt$$z| zoY~pq=be~E$P^efCue^SA?w<)P{M!Vko^<;AYh^@Ao2A^hJlb{xUbK__fFtiBLO*I zB9|XMD5oFYTmQ-P@quXE3Bv4Wyo9`f)FuNcf`RJif{_`B{j{IZzreN3zetckYA%U~ zVU2*a_mB>d+Is*cbf;SoW>EYOaK6$N_R%+q01S58|ANQ=1@HfYrSE@1 z-2avykb~V;zd}g1-M>hrJh#6|0;D{91!0mc@1ce7<-W&e|H;3@rr6DRzp21XSl^NN z!-sQ%E=g1$9pSN89b575q=#Q>$Kjggw)~vNhe89VbH2Wr3cM-^;B{AEA8U=(4 z--@~pW6eQzxj_Iv84w9&YriD12ZJLg>9|ehZ3*t zDB|di&8-mv_=i%a5o@G|-(S;v#E6yOxud`&7B;lE7n`dv&<4r|Z8p^GDvS|?fYfj! z=raA~KJa?M6?WMH9PGBkgP@#(vWptzf@i^oua&Muq`gUF%B2n{KQt3)?)5DvB>=6N zEb}scmf}N8@?)qxIxrt|b zt-^ajH`kFCP5&SozoQ588!Zw}HDb(7=_Y_s`b+$9_T7@aRXfY;y92NxWd zmzq04jkckOER!K*a=)qAP=PJoob;+P57%ysWcPw8>42wXlLFmm zeMmU`Z-?$a&Vp6HbEjU=e|xogl~6cfN35q#mt7)Oj?8_G@7yQ|mj=d{$2?6k9ZgCY z5hoZ6_%82Ja2v9D9Yw(sMJ*T(7evxXII$eb`WqkxD}XQbF~sZqa=8xkxnz@NeS~CL zwTwU9MOew-6unQU-&bGcPIOw}n4L|KEj1tMZjn-ryEQHXRw_JcBmirT5rpk0OJwMZ z*)u&W+IGrN_OgBkb>wuIcB1xNm#YygYwW}YghRb(BX@E?HP|P@#r>p-vSdMeL#z_b+#(78Ve9dTbJIv`AEgp zt5q~=pqI7cymIB;UPNrUI;@zYr6WK4SH^hhU3QRRfTHZ^WkDT?k2moI!Occl`09HC zecfvG304D-oym-HYUyU`HEU-@l4*0UCvp%c3~}?pl~cSi5^cu(Q@|UOu79N zBw=NUWm~<`B#78@AuO};LO%0qME&)8j#Qk2+Ss$wX5B^ z&Z#3pcl+}*#JG{9d0EG(Ga8Hv&ZRa={QrbT#X6YCh8HpYCXh016-RjP4&jPC#0zcS zwRn6I!X===@PmssiN~u;e1AOJfyJ9aVb0coPgD`JxcB}&yN;(76rdni%wx_2tX6XF z9-hAWm|TlM+E?XvNxsdDW-@-q@w?t_l7(K6Gg%Z9xWvY%N=K)7DmHLB254g?&tMl5 zhqq1@XI^Byz*?4Lk_j@9kp*{i1@*QdW5AHltoTQWh% z(;F1LsLm`Vsno+7x=ep?dDD{#^zMHpTaBPqE4G+;7FwSZEM%h=e6j6ERn_yy0b#oV zl6I^p8v)mPvVoMTP9C;|<}<^j4Je%P+J)C#%g>}kx9@l9cL#3r`K%5ok&%>t>8LyF>0{7OCFB{yuPFY4P$Kx?r?wF>=7Vh?cCGFM5sS`$gW!}+ zA!^8x)Gt^aMsobhp%3NEl!`lXNDco{32r#J(q*vQcm9j_ZqB5kk=hb#=qVKXS1>13 z>UR=GT$QYPu^+PH4KekGRM+<&G&>D_8WXd-Pc&plt1XO z>IVIjR^dq|oYeX>!EomQP6L&yRnewr|NdNWQ{%vT(37EEq(Ub|%Sq%;`#;h2gV0&{ z>IzSWTEx__zOEbmm#^Wqf-K)fYCoJvy^u1O4J$8%-IH6@609qFIfDmE;rsa7=2_(UdU!iNNWi%eN~Aoa+E{3#Vc~P3XG9 z43cGz4o6NfQ9d%!!WcG)4P9>z+XT{uV(?}Pk+^cI6=&REIw@1T{|pw>Mn&L zlJsm6P`7n`*t0-9Q(B{??<=GqH>!+5_xD|r{AuzBDbM^VZS&w|bHsi9+-1cHhl9elOKB+u(fK$iy0R2ae~0FU0zZ z1eKMt)S}|b(YT}9Od@{&Wfn4*SxKWXp`|2R)1gwcLbqa_T;vCpANzZ-{lQm6dkq!+ z)j@}@xYD%;a5hG*Vj!eyPfe#g)LV1p$=|wW-Uyr^3n-aT&|%KMC-Z#9Wd78%OwBeZ zk6rR0M5L^8wO`Nv@S8E+&A3?#cao&+R@##GK(b>#E>yP47>h2KXWy6i4haevd;tP^ChJC-~>u(?R5CQu9= zq~+|seS?3=>{-`Hf_@W%-akz-alE4_9@)@MX!`olcD5y-W z0j>IHCw!vUAZYRzq@8O>$S3pl1&hzq4^4Vw+e2C&Q6@)D7zj;hP^f1uHM-%!Rh=A@ zZ_TN30f~wZKVx8CC%EDLRmaN`CIcs03W?NbrL&iQ+dw9?@$%`AFAh3U6W zhxuu`CHUVr#eFvTGG=(M!WHQ?mAE@6@y>;6du8!kTc@;o3-l%%Z7?a>#6<@=5EA_- zbdf9NJwG{4!jt#n#4XS;1u=$D{9HM6ML}Us-fcjMg;N*O6!M{ zlB1og1r?!EX}$!#KJ?JiB-p9|koXa8TV4BIygwXEJpq{-KhEiBhJ|+4NcE&(UHE6r zd*lH;Z-V>0BI);D!0WAYt1?%6$34j4J!7?vEh%~(Nz5=90#v&8FwPKKlC0hjwcTp9 zAe(x0Md?Iz%Qq6<0v7>!sB~1NK*~^>xS>4Q6Kbp7s4Hehm ziUm2API-JHLOywoim&&J_*WxYhVAL(*F<_am6aiWAj|3>(E@#J{zJsjg%OTzq>Iw z_zs8@|IyehVKVi&7jB=;h4(#CBIs`9<>8@H1J&VJ0Q=hlhG?66Km1zxEIrD1X&6NiwXFt8om z#bPTTJ>BhTgG{yl5Lu-;j$mCb+Pt7&nA%7)RG(1(Q4@ha^I%22r{qP}Fjg~5-^acl zcy11(`=04h8gS+!0jLi)`LUKK=Na*KFZ=dx`umJf*Y(uYdmQbt;6J+Z2esP!^ew}M zAD_74&_r3n=@-g(5LR|xreh&u@wy$@GbQ2t>gs?0n&zo<H9W3}(+=7nmH7>sxwarVMDlQ4TnZ z(fbAL1$VPcR=Z=%2$sHyQ9oJp>yodX?yP(~Xt(iw3@6|X!QJ+G4tcc5oG<<=&_%;q z-syu;@-w1@Cd#A!Wm##dEleD*E|x14p-oG}rqhr9hwZUFT!<}%JI$~m`0`>uQ6HPEBd6Q*3)O=-q;;eYO&bh4??JbK|=*U1o2se?U}FeDcxTq=Wy+p-n_ zu6Grv!f)>5Ddru_@%p8sXc}9QkyJc*-^@;Ej6Sz$j}#e|xJ z$-t?CmG82ns~om!(|<9O|88V16CazZCD2^v-lzSKrR6L^)p0tn);{2m+ec`;xw3Go z=J1FXJAf%xDaS+5$xUir8tOb@+k?9lwbJLpM7fl%`(F@uEvLiu67)Qab2VR-)P(Bz zpa4SItE9f*5+YAr0N!t(0&@!*X6o4r`@pgec0`NwK<_Z{@kEwyV8@R!*W-F zt>xI2+^QU*lG6%68guh2;kt0a_E07$9OB(PJp!nb$E<2P4OV9OEZDb->SaV8a#wBw zS98o4|J9XiSE1TBPt(LG&IJ?v!IPMll$|}bx6x4A9R1$Q-D?T+7n@FFg zmGW_HPg_j{_-XJ!9qD@3U z`~_C#hkkXyd4}fESW40_>nRz|4>@kT<;}`k_nfxvw$6p~ve?JzJS*0ctzai#YD5`= z;C<5ULz0`pEmOP0mf~<+{OugH$i2r7=lZj6dD774H&{v)tWTlyUheXvkD$^+mPIwS z?8W7;o)1M_B5Ta<@G3Rbx1Q`uTcf&Ngai=zgi)5uQkfRVwhlix@KEoo_Gs)+)=Fux z--)Dj|Kj}Ok;c-Ay@u8WW+yZx99bGWj9<#{wQ{D;g-BtK}!fhJPl?1Rh;&)^^wi z%DoBPx5ZL(1J}^rS61Q>KCRm0L?$4+v;~yD>76<#?sTadi}LBBf<}WHu-<3mo}}gV z!>N%ND7DiK?P!~tAMCcxC-=@N=Furlj!^t(-O3X*DMRNA%T<+Ae46^Pc=Bk-YX!Or z!^K5h%3W|l&=o`Uh^ZJ+!0hzPx~G@CE)=!x!xQrXH+jC3L#CO#;cemELGu1d`*2u{`bFn8gsIB$K!>7auJ1xuEn1VjyW!f%lu|_H%hl+ z^TUT?2eO`nJS$j-{(xyBFQZY)kEGv7F}~=b%GwfI*|jJP>EBUP=DGy`acS0$Nl{aq zCzrW^?#WSd)}#zS%Zyi=$qxffg@}j!T^yvL<9IVw+lYfD48>n()cI`L#NmPV0ntv` zTdjQc)PbbK`UM@L@oW%@%D6B!%&}BfM-V7t*>~ImnbskJF)!qyFmi=e{~B z-Qr4}P-htIdFnDbsTWggoP4L24yL{(=g#a$C*}phIvEbP_P^r6GE)7J*vo=^vkx#` zF%RV#E;BJXzrj&vN(C%;+vQw4o`4|l`4OrN#pv6st`KqAp!~@KCUC;CAVNt6duK_9 zCtIEKT=_vR`-YVb-ogUDxy0aI63)FXx|dci@u*GE^HVaa4!$}t6T@a~|4b)|!OH9L z3={KDZ+U2_;r#aPI`w9}YdP#vOqd?qDjHo&qu&QQ%hHhx41n`zPy)y@!##B&vyJTJ zkF+uIZST+cevZwoG5i^+rXR<{W4BLnF#Pu1Cf6TAlZ8j`gq{|}R0k?T_}9+bp9$0Z zD8og(m7E1A9y^q^Y!%9H*^mfS5d%k6nABi?jDkjPK*{0YUku}XMu|U424W%tpF=Vzgexz7fNf&gu5XPV-qZ- zA+1}`Ifm-XVw23%T8Vnn^< zXzG?N&$1X;X6xzoHu9m`oD&ET;f4yp9#WV}RFDnePE6?V(5$Wekqj8#rFpNtJ2UVj zS`1v^0yojs2X5C{l8uhD{zM);uD(HRKzDobI{&ZZ{OHYUOGL+a5mBMq9LuvxI%e;3 z?NIKRO7?%Vjx`8hjVnwr!(FpEZ}p90sy-0XIw$EEjFKZ#*xhhBJYv0Sa64ORm9Lt3 z780Sl3NGK?zmp;ZDh%~KZWI>jT<=1;;;GaMfo11Ri~({(&Xb2JHyS=dv+4-lDF<_9kKb0G^M*P^VXs&6LpAuW z#SRa6=a%8vn}mX58}`3n{Sg1V`7+nH=S-uB-{O75-d&9sTQ5A0Dr)NRw~C~zF0Ms~ z0z3`s`JUZ_G>e0jT07(enJ!*qBOlpZlL;F?*`Bh`U+qH`)r=UA7qeY{w{A_G7={O` z@!b>CHPH*@i-9i8&&*zl@9x>?2#{P?Q*c zOPI&Fm)2yLt!Xrf>_jiFndF!synd3W^Y#ruUOF-0X6Idom$33?Rbz}3h>U<115%dq zm;ro$xz+Lbd6=}F-IawYCn&!y3Vi$Z3&V{ov*&}(mPTRpg05sZky88ERGhP+I&4qk6y4k>jyaH!Jz7_Lrc0$bggc^JUHcH3Is5NOfsFn_D#c?8y7uxcjFc}Q zMi#EQHIeo@Q^>w|$M4_xzn#(w)xq##Bj1S?dkpcblk4(#i|&F2g4`5>AdsJ4zF33Z zz~tJ1F6|CT85?9BEu*XquD~BnSU#hw@|~WH_4w>UK{*Oma#}I-4dMT#;qURRT}fd$ z=r5z-uRhWiUa=spAb8|>75jx$5PO2g&pi%ZDf9(z9plj6)UBHN!vvSyx(BPJ8{{?< z)?p+Ick1W8UQ>*@-*@Qy8=%vlk={+568D!zD$nG`rW7BUV|sIc)&B^JhP=~JaMaU? zn`cs)czV(d$%jPoABD;khMi7op77&k<=7*`cgiH-8V-kPSm!?ua!CIK&rF2&vxGkj ztUvJWt)0yZdiVa)X?l?VCcc;tVjE>W@nW8J&>2EmVZldFDO?3pivg;LV&aIM;1??< z6v?CVDaou3M_qCWIW5-Tl#L$SaWE43`olIeAGkyB;4K4g_VSVjx-PT7*vW^+RBllf zGdak}or<5PH`7Cdb*=brmZvaNrXxdUIx*E|X{MTDiuy{aQJF-$R~n3ZOebPyVbkpL z=r49ZcUuw{ciaXYBLTP!KpyPfL4z?0UPlA7gRM_Cw5}@cGIQ*tkZ+q2aDO9B*c&5t z7-8Nn^Oho6HKJNDQ0FHwmp{sOBcD$9m;`1F5k?zitUfUN6qPskaLe8JDS&+KT?CVY z@mTY8J@1pAOP%MD(4u?hhu?F(7K0{HCy@iRjmxOpDSs~7fNpu?CfWY^S@ur9x9ng9 zUw$^Gt`8L%*>MfMO)slYMrM`55{m;=reLoo1;F9Uyf+=^zO}ls!~P+{a+camzY9OU zgn>@q=$6WD<35lglN=vap<9cNu(zv*%pog<0esD{L@oxGm%Sn7c~|<4TfxEMg0-=v zva+Cy4sLrJ__OopwnhaG93hn3r)vmf3aS>u>139dLc&%{!7cJl#<&-QMSeqrs^lpx zzHHTb;{y6ZA4B{PKWDbA;?Y^KsEc1XA@)P7x}+=W26=EuZHS}$&F&SmvioKyNLk3+ zxP;4=8)pqXjKyaeP6rpb+b$Us> zkTr%vKVItnU#8N!WEnAVwZP8j`9|<|eSYC>HY_H*-3YMd)6ah_)b(vs-3Dkv zqby;Pe*$3`l$*>keo-*Yavp4EqiXGSCHA=jzKZY1j~On}-fmoEYII+tbs88V;_XNVq3z735l-U}Imvup)~kzT2U_j2Rv_q!+Ic@6 z66$`kxem$evJ1)Ph5L}ZlT)F^s6vYm+5ffYMeJ608;VtnhKOsB)ciF2e*j%TqQ6wK z3#)pf4C;TJSj+Zv*O^GocH|DJDm8n2ujyr`&*uy3974>S_;K@b@A(wU6xM3C9IK<6 zpt8h`Ix&jq#XxV`mWfIvBd;JS#kXZycBAYY-}+|ktctcDghP?0UxZ1EYLzA0cy}D< zjxbxA55ndXVTVNp{V0d}vW%cNf=G^3aXET4e?)&?zRgtaqdPjHHVtJ&t7HJ7$YI`X zImfPje}BuFFiGN@1xDf*1-icIQSB2fo!FNg#Z~th1fMXkdF^HZQg8mKP zRp?z~<1QxYk!s&B>)pJdwK|2;q<2vGgKN&V__MPdP_r7JbxZlWi{;?C>n<_V4CWAm zStEa~OGAs~J_}s=rozbV@hQFH2`wsic&~o{btU^b$B$)PW=4+vz*=Shtyq*=PFyEP z9*`foYAK(0`Wy1*4Bb~xPLEfZ3)MQR*KuLtRCeDGL<+|5AQucjx(T_RkHJ2+v9=7y zZ#6}!Ij0);KKQMCQ6|*+(Ro$!j@)~qK`Vcu^`h2b9z!-JN$z};!LXj}H6f-ddUyu#E zvn7ch3|fW<#!4J!?fdgJc*(E3=DrAWUPYF6h(R$vSBYcei>e*=Q^fpgqy*3w&aZ#s z@~J-C1Vo5E2TmvIF5@Ml&N`188yT8E)@0+^Q7xTJ1;o@z4@MU0PXLzSya{_wj78E- zX8IZ5PHU5K``0ZD==~5O)WO&%Y0h0#DT>D}NI$9KXwhQd@N|tIolq<8ev9SFM_7Lb zMByoI8*@R7N;?9)$wc))oFG|eNn!A*?w zC~k1j{T4R}WZie{gZ^-iI4J;DsC1}RxYJ#xnZf2Z#f)0f3~8PtE)+W5wF zKu{&w>RC|sam@suOP^t#=YWcNac$<1xkw$Cf%)6ngggAAdN&E0&Uc(FO1Xh_X3=!3 zkp4=nirJF5iH8$!mmh_*Ft)w}g}XulbF%RtiH~9s1Y66TIa1c4q(y>cc8p@P%F!&t zcyLqms{wr3M-gHpTsxT-ud9Dlcu7FzpZO&iHH5-nPem9yoza8(?J=Q`qs`XCF!ZQV znJrPe5sPbr$}bRl;&6VMcuxq~UcKnmzitp6>djeP-8MI!dyQ>)sXw2~hjH(H6EWE} zE^h*CJZ&{Q*w?1EJ^$QVULn`2G$r;{N7mA<8vRUQj7c%`I~(nT(OQ2TMqE079CU5n_g(YiszlINU?Y6(D(xygqcd+h-$ykgA z5nh(PDvyNB#b9g2hHbhrwi@ zx(n;*=5=IO)}RTL{nT)FWsDA9)J)z7>coc8lx?(Te9d5>}8Zi~F1h$ve$JO9usq&bP8Zwt=* z+OuOLP)HquPx61`Q?G)apU_XY#n#c;=cA0-kVHKlOKZ24srn84Clbk{4^3)cTwgyS zLY=Ubk^Gw2nxCZ_z>5*eAY2R8vppLkul->EYL_Tp&`u0Y)ZBTtAK=$(mztVAiDOLa z{G^eiRE&bI*raxA_Fa`^N||mBr^J9A~Wsl$l7O&l$|Z`X;kwYk&oxg@6^4w z05GIoZx{${H;p~f=$ez=mZC+-x8fu*zAtbT^7xf@hD&`ZJc%I;#?D;A&9B;bNRD5gE@S?ImUa+r#pAg%CV|f*H&z6dk8&l?|MRx7FicjY zI=kfaxc@iIXiP*3`<%QeSsrA%!yg7#;cS5A{gb^D0(^Q=Qyu1q*y`k8W1%*Zcum)q zu&dU^2D|!D#q?)_a*=EzMOYFM!+TT1oHkNCzq={>*HN+hz<)e%>miDWGZSlbDti4qi8Ed;GDY97* z#KTbPD|TCGdM<%e*(%8>$@e8~q2}w3BH^aTDA|F6b6X;SF{NUBd*f;YyWIa3AV~eB zWf^}G5=Lps4rPJP_ZD9FyZXmLF{q#fS7hGr0V&7^)xsg!%%ts-<0s;7zohJFJE)(# zBOAl2$ZAl(_8nqRjIvfmPYg7}UGlT$qoiaB!l~2{rB~EBpR_oG@6q#RQ~p1lBRGj=GDD9Oe9+)Dr% z^((tM7QC=n%fzKDTFry@2oI%_FMsLr)5D$?Y+gQI*VplP9jJVmHl^h(R9u&&>a%~u ziRsD3b_0@&N#1=WP9Y)y+GLP*{F-#H>!F~ zVfWz^u8e`}X(J!LoyP}pIM;}8OW!2|ld!y|Xg$1swNtCif!wZlwkH#c0_hg8x6Qa? z99O;jo1aH{y=BdWq2Z=wX^0!;s{OUytIeo{y(v5l#vz+IzUH21Lxtt=O zFB&DJh~p~0uRggOW`QDGH4T=qSp?&yrqb(a_==!*JeP?t!u}h+KeU$zw8MU`S~&KX zE*d0BrcVMJDW<#w@%5Ak0LXt!Aov?c_R550n~)ICeQR1;_i->|dy1X3j~1%<*Bfh1+rU3=3A+>NLy zngEfsi$lmcZ3gmrTbh;vHeT`}g9~L@n}<;IKZzrv;Dqg3xsJo8hU0&?D6m&T{Q`Q} z5&Nc7mzd{KDN15=VZGow(!@LGQF_7*RVb$w^WAG!c*Ez!H7{(1Y36V6Hu4h@_eIbP z!y71t``s%=89DaSN4`2jcT7FLQQb+Hp~Q&eLVyuU1vCT&p%|BVJ)0e1TQ9qJ)QvUh z3ER`-A1{#SYG~x)^_PG58@?OhJm-GOi|z`k4LTA##x?OB$sD9;;18`1_*~&75@~BP z#gA0=4KvL|!a39Hrb5<0pK1e6NbGK<+O2u>hoIf|=#P4@z+f`!j<9o;w|bH0Q?egt zwV2pEc*|SG$B+*`dLwVlc&sg(BN<4Nr!CI1>5vhNPHT@Iouq&K5^pmbm>w4jPQGo{ zSY)p&!TZu8wRP0kAiQo%J$c;wbg#~f0A?O*#dNW>`MsL-i+>eeU^KT2wRsm!q?#4 z5nVx*X>sqgj@8Xhlo@0(L#jYSq}!-ZQz8-+{-a)2rHEo$5Gf%Gsy3SG*XjP2Tc~VU z{24hNXi_in#pab&|AbxF*8ru9J@lijnp@udwzm2hWJZ6z*7l`Dx!x7=%ljx5LFTEl zu=$wKYim}d2nzsA-ZMTiR>gX3?Wc~Rq7)e>XOhm;`Hvd>yb7W{(B06b-*)wi)~DEL z46+^7w~tb51LzF-`dOZDQ$QYXc)o}MzQn0;Qeck4GuWm);A^rhsW!!jqu`*E1bQnr zBzLq9qAPzrY|U@ETg%?gVQQcu4dgQH0?k2tQ_7pMRO~XjlWJyw~9j zkdc3tWo@q%WnU~e1wPo;%RJ>S8Y0tvK*_?QYl50kc$W@*+lhR!@a3ZwDyYIoi_z>y zSG9{I<6eGApSkHkh1<`Ia_v|~iYl|}&Bv`PhdQKu&$RZ+47no%GcS7_7dy@@{RfFE z8EvroOVStflRwh47YEfRQ}`0MkWDxZ!)4Cy+*&* zro3PSgdGyFwO&YLoXSlIds3P07c#0jvCdx3MPZOo!U;x*g$0eDr`1mx&fL9k(uyqY zk47{M|HiUh^?1Tsns1UTMiAZ%9;5x*tLgjh)!vQM8&!(WSDOR0QoRn|yw-Zr7TAC2 z{vqU85edJ-hWU;6_p{G^7%)0FZZlyfOd5txO`R}P%Fdj^SG(R1ruLaFzgrfOWs2q% zAW@1mJo+V6;Dz~hW0V#DVpdD&eu(v&V`?+^ri$Y`->+Z0DW$5))281usXtwn!8sbn?wTfcO6TFFeJ?`=aC9me^u7y23DEpg~r z%j+84Afd%em{(JymFVi`Q6TMm;h`J8c!!1!Yx6Z3~ZId{furDG0_fEHs!Nt!!FSZ)v4+BYNEHF$aHFEUi&&K1d06 z72^$dVyfW?LDJo#sDq5fphzSft;DHlYd*N{dmK}sT}B*QDSz~f_r-`!6A50Qr@+UY zAKAi729N7LbKiocUa^gbff0W#>@d2TChWv4qzodWO9~nROST)(^tNkoQ>nc*k(Bp) zK|tNRUsk(={(|;#%l(PMxkRKgh-E(?+G20&!2}oC!0s-yQR~it@?_eQCSpE-OQd53 zZmRa;w75LnD_|zoz*}+20IV?VqkrG&sdlc?kQ8LHv)-5^CW4CnnTm{Ql!NJ*&^DaG?@r01dAVYgc=<%%Y=Y^mA2@(!{><>I1w_Y4*G zawoqFRH5>npW%i!cSJ7F>~prVclD#gN*cDI{!z6~e#^Fs(& zr^pT#zX_?wA>5&`hy=w;K!j@_zVMs01#Z;RR|KIWjXw0Z4oszSI}UTEU#%0eB*?JZ zik!C8-s=4ViU+uI;)P1SEAWs0xQtRthL$z|_-U;uqt=npoqT_wpkxBm?F;_R1+uO#$IP<5%Q_Eb?VvwILW=sw$0JDEjv%k%ebeCt<0d2>;{R~lRe|9GCm zd$7PqGY}Xc3m<><_IM6X;5cVn>p*G6Gu`fu}+5mbHsot z4Xx^cMvbqd$2!g0hiUcq6rIq+8_O_g+WrM;q5ZC;HFuXpD~NJgr(T&1oYrE#Y@|Yt z-=Q>{!`(gXWcNsrmHH{r)|cl3g(q@q#lMu4T>Rvsl}LYgY7~!rHf&y|$A zRr1zlZF7H}+Fjg;FAHxH?4z`G&#($fXlFL(Hs3J$*RVC315&>JJmcH{QcIrDu+V{&@U%XX(kf~HZ-|~Nec1>BvYBa|9v>zm}FU~5Ap9pXF zU)md$uMHwsmXVTc@St*X3ZGW^NeC(=ikbHwO(cIR-6LOzOtH!zDneV}DJpivWzy+X zQu4gELQC4OiL2@F#VlhtWFa=w!f$y#5FPt~={>8~v{8X$GNNg9{<@o~hchaMCAy<- zU*~_($ra=>VVIg7oY0*;J9(N+v)dE%o)OLV1G4Yp_xT<8u1rvjLHs>nJY{>qQyR4e zpi{s7CgCTw9b&>81EF9MJ%^!5A-uHHt$oz&A|f(Ff4_90l^NNv!~0mc|0b zR0gc_g5dD4a}8MYseLr3lQ8b#n{V?G2OfXAe3=mBQcLZrY+z76c!0yPG)^*(>hols z00`2>Xg$KIx%CPJR81w;Ym1+mf~Qu+1@x0lG(*~}CvIgwZiJF5QsCclSq{B2yCv}5 zF-16H0S9aVTvv+XY*l;J)HBCv}>0 zaq?c~ecv~pF_FC9mdrMbgPE)QsEX_`A515Ah@LDzKfwhqE@7@RBy77K69O^>8Xclx zD~@F2q7%~~s@$;AqwlnYw22F!{Pok%2^YoAq+`x@;(^c7!rL=U)VD!4OQ;B`2zUL;#`R1%MO2(PEZKM zav-BYH7`jj%P~94Hm$qd&6+mbCx1rz@e$qbSt$ObmXKk!71wIThlY?)Na|KKq#>3D zAEjWaTG~#<#rju~wDUGMtyQg}hW8yeZhDoC<&;hPur@>Qc~)70&&fqB;oYt6ZoUPj zdYc=YT7(=X^ykkz5mE?=C=-8OjGD*fO)K7wE9cv~?zKdjlSFiv`KUu3D@OI7X+SWnn+s8n8i zpjLnJ2}-qcIKN!k<+Vv0ytsG7pcvfulJw*&tkXdG^MW~1W(7_yzsrB#b!A@HlEoPt zl`fi9hT!NNeB?d5b!F!#%7ShlMC9)-`uoW5SIL6tnezPkWgBP?rk6ni`8gq-nk)f+ zNKgE=Tm+U^QoP3_{rI@EK7}R*bZl^oj@;pjj2TagBJ{O2yC+*Hz7g!&(;Eay;4X3P z{dM|p;k4n_EjE)>E2V$1=DpIaF*kJ){ISJD3oLz~B0kvIr#QY$%VRHo5$)345&HZc zG{=TC;f4eA%hNnltRTXgiqV$7VqgO#_W?dbag3Rx+sc0Fsgdv68_9bY>iWx($2%{7 z(9#JME6Ar(7y}SgcUDi#~Fx2oHyn0#6d9}&lJa&I<7Gdx5+M4M>(va33 zzmkb-!R_o;#&E58v*VBEle`NNhYBz#=*HtJC1q)>?b@OGw3u_ZN@hi3E}0^@`0?k@ zH{>B0jZJYn@`s9S&{+I7kNXs>Lx%>m-f96 zHQE7_hS}!ea|HAqU))hl9O3;-cbGtGq-q&X*^{|7oaTRKH_H6CiHm}`!icXHncjSf zdi#_zZo#C4lj7+o|3zhAfL_OQ7lj5 zQ^W5c*RvbtMsn&gY~+a38QDzcNtsH82*T{!lDCOpHNNvH^$*OjdA7C;`d$_r>5SXs z|H+q6QxJb0Y7})};KNAj!0yg$mzVgblqWac+MW0IB!^}MmUCuW22)7yxZVqK49~?| zOVUQjW zeQ968c>pFG#UCSyzowDuLry=}S#>Pv0!445gKvL^8TIOg20q%vlXf3zrb&r6j17+4 z6Qr;P%@jb$0?B)ZO!tvL;tG-k^gejKCRK|drjG^$Yc z8QOmr6ETQoO_sLkex^`5)?$KjYP9Xea~ihSa6Rt0FP|_8?X{)im)>g137ST)NsN*= zQ+6*eAZ+0Bl0w5Eaet-JW7F1~A4=DnB4lN!@ZUbE4t4*FGA8|9CuUojcQ(k{~qT;%9;Id^`MNV9SchFEpuU0Il2GgKK zb80+8q$_UFPn%xvCCVEE`jc1bmCPiAhp8nV|2}KTG^+TluxR$ZX7U$2{&n z23MQzS32Tj8{?xnlB|zh3-t8*9EGk8iCgNOcw5t6S)`k9q^*m`*c$@h-e#bQ)|3s1*KiEHW**DLa`EAatL`nVlKwo&o;wCi25pb!^dF%32cSpXHuP0n* zJ!_{b&Lr!*Q2X!0c%fa{z1^h)P&0pJOh!zeGIgMwU+8%RBXt3tD?gE+{A_#^GDI0Y zsHJgEW-+V8(LzK~wgWhb{3Gtmj+s$JpM3C;xK3PcVd33u$80vAj_>!Uz9Bcac_@|; zEZj8c8ZPum4>7;x+LI-Vd3}+{lbNwic+XcH(9$UwG+I|wT-WU-I=ypwrAB{vP~Agf z0H6L+5UFi2d5odrP#n?)i~?O!bw25l3xw11@7{JI`C2q2P`Yj(YE)zPTogFSGKVAA zeqfBV?g|RoMP)?8rS2pf1ACZ>utd_b_A!6shhe zN0$f!;g`J%TN6Lf;tqVu z@Ew>FSE)5kLmTKfY*7*2uUT@t^QHRHE@YqaDt516n^MrlmmGo~M~d-h$#ueUGgvJ3 zN0LUWbo3-jwZFd}2Tp(Ybf;%VnYQn-upgZfmXkj@C%h11%DM-fJih;ySE@x~cFN}g z(A((j;eSpDPEHRiu4d}4ne1Awox^igE%yM$5Lp?!uProGiO&&iZ72)!u(Cd|V6ehz zH^4pkU%+2+95%7zJgdT|_b8_A@|sTwY9po&6fuRKwar9rxA}i!x_ZT_J-t|WPm4rd zVs&+wuFN=P#x05@!djc1y*x z?eYSuuVRiQCm2SA#r1~1=Xi1F07gM$x1Sz9nvr>So;_C81yq4Cc%U_h01_yfE>(!{ zrx90m13Wr_`mhnbck`~V>btA|9{?mk+rRqnM@(=PWoUmog-I0Wr^NUAjtyf(bds-? z(zgs!b)T{epA~Tl3!B}{NO)7M4nHnDjJd-f%;KT0T1>q07+0=UdSr5h8CIKECz7`Q z1z%DRY?wl7eJ&ykUQW=FWl6*XIzI9*<#b&kmf!~m*46bw9vr2;Ph(JQS*dJYbjL;y zo;Z0wvIKwg%2iiVoteEAWMslk0GNr~=pxmNNbk{w#!=!*%^eDV`qlAM#JZH2)$W>o zvMpH8hab5ZnxFIVKEe*|ty16{)i(e}rHhalH_T{u`MDlmGX%Xt zn&FP`x7xcu5(7VPULxts#Hd9R$a|)24;UYLlS z{JolWsOD4=vWY_PpN-@SRbXC`L#rq)zCF;!8%t^sMq_MQX^|Y*8NcTDLu2q49e87j z=$x%t+-|;Ln#=mBrj^@BVr?cB4dBF@3ND&7HJM_n`Pxe6K_>Z+BQ}#>#eM7gEK+JC z=t+O7{(MdKY(z7Ksg>A#-qEHO!t*&4nskcG%-ECRee{j|h>#KhfIcu?;}|UXS*qc_ z;{7cs9_PV5skDQo&|^r<@#2wh@ai&pVMct{U32S7E7}%O_nYdO-&uD^(TKacJj!KQlJ;DrEZ$s_8ZbCPs-*|nhZpX zlv224&<)huR0Z8w_$Sr?@U>`^d$?M}!T=>S&3PH`f_-GFG^D1VZ>ikb&;X9BW~&uv zFMyEB`tG^dFHrX1`wyuMV%=^|L`=aoRA{C~ylh#&jlpMW1*Bhq)8MWA*Ig>Awv2y| z<-}@nchTSyF0F!-Ml-;}yzJW%xP+E77PL}>vh`yfH@RWWUJ=u6lgc&gMU1T)HA5Dy zkW7_rd(|4ozI;fU+;EOda3@aO%s=$rgxX>kfT;k+y$b$Rmzhre9w-?Z?T&4WV&ce% zv$rO&2ecp0HpBVJn_Y}tDq?l+jLv^Ki$WRu{V~|U)HY+?Py&Jsyx541ksIo$T-v;Snlo(LQtfPso~2>8rMfxlD+QE#MvEZY-Zt$c{(MX ztt*!couVT5unH;Plx2HJ7a84}=hlGU!;|f8!+M2Me!Aa0`ui#T1Qe5ss1|=C8G)2l zzhO{%n$6BFQh~vl;U9<{tT^B}R0}`=h9+mH)%AvX(^W6ao#{Irx`r&nDDmVq6r~jy zu?58&)?PjCxvp|7-@mnlf5AxW-0&jaBdYPhKVFNc_4IqMkW8;M2ZDUPj{B>^sYNwr zuBNaf$54IwyF3}bAa^#Z85w^I09#$tMgg5GhU+1Mc4Z+Dg=u|AdfNw3TiGX&3Z7p_ zFXrHFF7rxhR(cSpkH1;rT-ZY5hpE2zfT@jfL` zeZwK_jMp^cEu^mclNHtlsx?FB%y&wk9(j?*xahM6&gM?UkY9f`psX{)Fac?3sz1+d z%=X&&-9U)9b9{|a=xgkmJKm>-1&!@#U*gY#b!)BfCMjwdDSNjp`yT~Q(%6TE;&L>Z zJ4m`pTW}M~rZjDizwdu1O&r&Q8NW|AQ!eXT^3kQZ*sQZ3c8bOl`?C4UT>T6r(TBhi zy&nWsH9CQ0*ryNh;)r<@zEO}t>EKHg*v&r2QzjUWDg7ZU14EcANAoxy@>alc)Co*p z_pb*kN{MeuU`V7xgcd)nJ@!eoqzX3&9F?|Xw#z@--ejEruv34c46w~Yf;D>Td~@X8 zNLw`qHMP}Br(;(!ZjSN1UeFu*jt%X@q~4kX0>H0-!@E$0+R0cR$8JdtoQm;G-P7V$xH`QOmNyr2O0P zSK+SA@^Pq=D9BTddicoi3+oqIN>wu7g=y)|H#)KwUPXUUxLXVy_lpb}OEt)Oby%RZgai(kE{t7~co#H$FQV z+;k5oU@rgt)P_t3$#M17U~i8VK>uTtIgu-NFuXOL)ItYQTpZQVsf1I3mCvqPD^_AG zwit4W!h%HkH>3299v z1|(jD^we=i&9%$|e+D}F%+v-uiG;(5pPkyOAclYE(?RQE3PU&t2&J84H*;&E7)_vI z_NzM~8M9XG^QyJVD4=|dl;c?AD)?{rl8hE=VztXGnHN~Qhp}3 z{7t9C>Em=@1Fq;Vn7+4WXLV{-UvWS~?yi&bF)5PV>vZJ3#{mgf_4N@7>b`dbIBsLY zq4OexAocupNcfwk745z#ia z-*Da)g9YEL=B5S@bc8wLbC=4;Eclgtf??f~#0-?|rt?-=ON60#tkP&9%JeweN-g)5 z+R}!-AsL*0jy*)CBL2tx-9fNh2kU>$t=Dg%4Xl#AylMeYiPXb9LEqwNk48r)z8v45 z^&^~OGmn0}%sDYbeaO6(yxeYJEII`CHxqvZ+9|B*;YEH(g|QmJfNMo2j1L4)Np<_E zghPD1P^l><7p1e<^5qz`S}o=`toZuwkw_k{x7qNHyIc({h=_)mSXO}7LRx>2*_LhN zWtPZ1PW4KCZLIeTi!Z{ulrk_$NZS>3Dg4iIAfDa%*Q8;S4H5r9eAp)*;PaDGY24|2K1 zlrIDDZID?IC$@jDYV7)FkSzgF>L~1T#cJqNBTe27Q-TI1BVU8`rNJ$J zS-bYr=s}@TW{`dB$at=&#EFLTdm=xplk(#A8=hV8)&uP9emkV>l;l$V=~Ut+(E5Jf zH^zA;3XQq*jEE6kV1j?%$%|LsOUX-WTl>X*(S7G(bQ(GcCip`4n~&-euNRxNUFCw% zW49WbVscMA4%1lcvk7{b8`IECy)67}CP~=Qqw0f-pitt?W=mD28$8*3Ct@#ZR-XFE zNi$z~qtS-P2RmXy6!gYX8b5%A$X+Wc{~4O~k^i8)2H4P^jP8GeIAvKCa5qUz?)i?6 z_;%Miu~;f7s*vEa=HgNF#1N+cE8vuy$6v(sa^yv#_w9MO=#lDJi(I~MD}79f6q22& zl-itjBQUaWN)b_)sY4sKB$2d0h)k`Nf+R&o81W)*KzD ziw(bGqD@DxQ-w!*?BpuAjLivOJa|thC{2O3STFug+U?9?_TX ze&pwKs3(M+{Gu`sSAnK&w`16Pzi~Ks?tVE}-$?}3TvdN_6Gl}@$^VPUF%qz0&sABK;ATmpsN(?CDKG?BUMNomg6hD9Kxo%)7tO8F1_nHg-Q$wGM`ps+w2L^Qd zfI>kY6Ma4a56+f}#egaAoCh|^BFgg8Ltf`CcmXFU_A#Djftoe*P>2wsO<%{L%A^Fx zk)mLo`&^^tTx^;E~b)mfR{U zDB=y4$$6NC!@XAH?9%o{y6lkd5$NrIFQj+qA3r5B5pvoiSHl`9RZlK4{b)n@m>Ksd zjGocS*0Aav#{^)c2Eh&P_%c&uIfqny6q+T@pN>CIfHP+|FJHj=*DZfifW~k@-T;m`nt!^!4Sj*pEFA(nbfpBqibNZ9GzigpoCs=# z>Htzxa=qdy@w_K@QAI&;w%f>)u#OQ2l_ebO;2nQ~))NsD-YSkAx1}Hwi!u)` z4-x&OOADiam+gcBi!B$zR zV@sRdO6dl*ft&kgebad9+%amCB^anf+q7zamQ))ZKfEquXpx~|(Zm+4g~5rYaTA&- zBUkm$)Xw0$jC_vMQsANH+wWqZNGLP5mnHQxWBNFjd5M2H1$<&8I|~vI^Imoe(_{AV zn`YeZs5KJ&1!<@me%N%=l!Zw(TP$YcY^Y#|WyEf@LqX@!Tk4O_fCDtYFjxb?AQF|vG>k-Fb;621KE9!P3TS37s~015XMV4q(w z{^fiw+~n`Wpq6^87qvaB+`3k4Jm_hGbU>Eg7=M54Xov<%rh?XHEC}rophO9Um|k@g zPif|d(WU)I!V6!`Vy@v~)*2Jzn)lqZqZ9L z8x(JQmi8UmItG;g^JihfECDpz)OPGe(`KlhNnLwm|E|i#ht{UF;{I@2evv}vI%6qz zQ=5MoTN)rXlL?SAk{A_WatDxD7%jT};#*@lZkGXm#L8od2pN{$D{W-cKg`Xx@~_t;+>ybhQk(qeV1X#gb5=jS2KRyf}q?= zmaJrDf%#gkdLm;4Y(6czLxx?d##Di03mL;Ub@g7+1mxW=+s%rKhU(wJ72WN7soS>R@)yr^ z@FiOxgWdM*f%_+LgXC0&H5eb2zZsXvS8qA?NhqAT@EDQOy)4#JA1gIwkZz%Tr=bj} zLe<~;f+Mgna{OYDm9;VaO}YX47=3^FS>>1ghRNT3=+o$V`;3B!U7de)xh1}zy)xF7 zB7WFU9Pu!EbitNph(68q8cg(G>5bqEGYyeudPG$fYrH~%^Rc4|5Sza^aF)^BnDO-E zE4UO5iHc#|rqA~hrKPuZ0X$TF3f#a|CQS2yaDzYNZ$6%N?PHmt1AB3A&^CX6piAU7 z+0dlgpP7t!A0;^EWu}L8zdQb_Ym*EH_my76ygUgb=cD(+-qLu!%s{+40U{?di=Oi8S_?` z;pM2bXZE0%d-%VCbAh>FMnr!GXo4yPmRsD3qa59+GN5eA3ON?Z{BBua8{VYSVrmN3 zKf;H_G$@I3>WmV}N76?GwhWcb5pz?47{KK1&`@ts~ZJK2}LmogwFpxU}UWwa9(p*>$jt`0uVf?>_UZ zyNRosC(Zdlxxkj&x=9uL_(-1LZY!t4ptDu80@`$Y2q2B%&O6GKo@ouo4_Ix16d;zT zagqtGGA7f&e^#WVH&}m>8O2mV*-{|-Zoe+^3U>QIbbw@)prJ65UO%S`kQ~+0#F0dI zjQ{b~wmu1%N031;f^s@r7W(Q6BUqQ^eV?ZI7rv$rtfdnu#wY=L$u>CNp&?{^2@@Li zl&THA0Kyqyn)S_&ZB5o#7P(4tQ+>e-B=Wrhs6RDA9s!UD&6j^G0f51tNMaTZP~EDu zIhN4-LWrJ2?hrXSXNZ4Ha=hmOC;qj}e?9|8i3nsgU-8uC1P^!W5n;Yb>20$7N0p}e zWRBfbp}w4op^2yDy?HYCw?+l`G9Bc8BA+6k_JUn73^!G;`Ebq7<8Jz` zLv@v|{^;_EPSk(AD_@pCQpea(Br~hJV|31z$yU4f%Er**zHGX_BcdSoVkCX{1SrAx zoOy2`oE8?B6t1S1BwNP5Y%nhb5w)xLVLCUm_7XqLy>sUYx8D=wV(oXjKxyMD$ahIg zH1uRM%E=D4K7I`)-fW5ne*A4@RbIVlD=JujI7Nv>pyPj4`H800m~Qv`uSQ`uMhc3( z#QS<{3%Ok(Je}T4%1yfr??K|^XYCc3v>g2JVpV@!!wkDnatT?xJ54ifGQ6!3X9dU@ z*tYsPMxQNlL zH({|1Yw|BzeEX>uT&}GoWZ2jb_QoGX5WaSq{~fCmM*&xbh`FZlGYvJG3&9u<>;jd$ zJ_64Q5I*ElX*Y9?YfO|!@s!Rsc0;f%E+00PexZN6-YKbDT+--T9*y{Yo@aAyw@03g zTxxhbG`nvHAfQwXJ*kafIapKMM8>~NqXfGZ1kjF z1Og-Ao+smRfIL~B!|z`uI%owR(GG}VSyS?2k;^cY_%66Z)Y_5AL2*sQNlrc~X|oX%{X z1BS6*s^sD9l;Ms^EzD#7ML%CtnKGJ{GfXX^_I<3r#9O}%CGHw6bS?84&MBoCJYTvm2HW&CXzz)Wjb&0Y9IAf+K5+Un50C-$Ao$;n%5A21R#w2R&@8#39+>)*64-Mp?Fd z&4E)+YxtDVt0{@mf*-;3&tV=9Qv)(P0pwKk|$Gpp35*y2UWNN-7)BuMTE?g zKf^-h!d2aYV73<)uteqSb}1Adl7MCL07zazAOd*H z!8ti8%R6vawXYgXZcApbEz+RA`KwHQMqRkBj{Pdw%l{_&b`UM1B|Gjev&EPt+Gh@= z^Wbt;_5RTO`w6d=Z2b1jRI1A9by$W=|0rkb`YlI?LWW4{mA=^d-EOG-#_&U!2r->=_ z&$Z8CH(u@r7ATx2!BKJ&zw*zT1Tf1>t&UC^6Kna(rC$9e=AEf_7 z5LnE7RA{n#6UB{kw%NfzKB-KbmdsITtF4dTsi`YN_Esz-+ zA&GNUx5kF^q&;W*Mhx^keA+biytDrld>R5=MMCkr@h`)LRxCGvwi@^{7G_zBk+S1v zRlmC}j^f<~_qSTkQ3nqR?B*HG9E?$dt^GKSD zf>Pn_+nC;DYe_sf)pWX=BYV&ab3UM5VKV{Nq|uuWMU5Nq%!?xPewD(1EI_jiRWJF3 zX;Xt|V7Qs1XphE!SUKon?#rh|4t*ju>eBqeq+eeFd1@jmG45)LgQ}`m9X!m^;`XfG zKN$rtKoO{-E78fI1WtWmm7 z_>)<;U7I-z53uFK^vew_9TYIgGmzRf5v3eavlVo*`igP7iQJEG7JPgIHO~2?wn!$; zmGB{%@BH*sqzb~FDF?8SrQ~jrKrLR3Sq$}`-`Kkx#i{cx(9hm=sm6zh&bzvIq^J_V z%5$jW$~Lfnba3pM6Vki3kXHoHHP^q4JhHF8FqcUR-8GWv3E;;gfVnSAlxlmgG=p5K zRJ~w_wH)XzS{?Bq);9a}jVFT=Y0_N}{Hjo(5!cj<*U3Aet?s?8P&j$}|K;GDEgQ@i zm{7$^Q`D!>_K|}-#~w^*)a23sH+ovrd=;h_Jj=0vT>UTBV#Ri?Y?}018c76vUPDWo z_tc^TEZu}YS!SU?Lqfg{LtzAq*Nl1qFVPi`A z)de`!2LTliWNTzVx9E=X2<&o+ByEQ~p#EXmjO1EUsui14DP#*EXhFvhS~Kw2JINC2 zmztJ;PA3pszpr}c-u(TzafH7q0*A#{n^>%SnpbN?P5{6$L+TV>CD^XoI-doK#68iu zF<5b=g&<-n&tz}SBs}s87Pt-#`Kq*|*7!r`YLdY3(Sw$lnEj zXM0BvW?u(Hg)Pb1A5aMpl45+I%X&ia?39auR0F4k-jS2N?W5O0kbPaNU`vOJD_vF} zmlt3&r=zQ+X!<#@kWyfGcRBn*?lvUKmsG^IsIxo3T0C(x0v=c{tBvB62v>23ty}-g z$L@3t(10W;cH~SIsgk{NR{5sW8eI#2oE9EdbO96Zc(^&r_6hhF=wG3oIkN;ETo?W! zdyhRyT$OC}_pu#}5V`_-wlw|`3%#o2{;mrNdcItqa{}wc}A& z#DE0O$v-os9+$@%&02MKquP*`GwT2xJBeO`P3=R;V+F9#Lgv+*08P;XTWPX?mLR^r z0sYrGb;gxN)O6*S#djyMkLO$haN_m<YxtDV<^9R zl;U!5v>LYshKLG&>m;UwOu_7(*&=kjen(tZ+PcwCGJsD9N2A8)HSMYyjqgM$zj2%P z%h4VD5Hqsss1wD7@pBdj+b~UkbNSX9e>?77-`}0q(U1Ka38dabL#HG0K}yV9JekhmYjEK)|ib2}F!WUwu+q z)W|1~1BKdRtlu5!G>^T*DiWzIkiV)TgzziTAKngIp$Eo`N%#xE8vi@qGTeb6<}gJ7 zEd)z(Hw__`uyk1luNZiLJg5 zzevIj%bEQvu4q$-d65O4q&F2C6ehNa<<0Qo86_dD$Fck9o{jVk^|uGVO0iu!j2((s9|X4W*sitz8%aJ0~i+IsWq{ z#;!hqs9N%>!e7FDk@qQ@tX6`|);EJFnhktX2`4+piXHm3@)<=J=f#>4q4gULVywncZqsBe*-dW%^x9M38XNqoan8=t%q~$^ZPE%S8ayD-k)X6EH-DM4-0~I z0ym|q(`^??UC4F0N$Bh*){D;{PCgb-YMQ{6TmZfVEyCD-EDqLM3>`Gfo=170VToOo zH5)#$v$+6&ql#bxslm=^MopCa35mIZsuP(G*A;A2xmrlqn;;8B)dh2g*t+={7?)hC=wapgj<4a_R$GN z|IXrf43V=8V6HEn!aG*`+!PAoed%Ijrpi+|@Kj%a!ixYfAzKRJTbP2q!AfOk!J58R z_!whlvky9<_Twe8n2QjE84$F1`Io$o!nES7vSL`0`59oc2oD>9&ZNdNV^ycJCN(&w z>pwa=@>8nB*ivm#w@Li|$nO=Hv(+pQHG4mVQE=oov32z&c*{om1c%%mZ1`P~X?b;i ziS$E%k5t#`tsbtW&fMNEYEf>q=BFyU`QMaYojLw^`e74l3SRiy5gs;1BTzob6!jwL z8`|(JigvwxtqAb>_kRbQyc`~t-p9>Hd{%X(Uqt1}@$~s(lBi*CauLuFPh&F+WV zCEyJ*C-7$_YZQPHVC283Z5y^V=&W>PHa0tdFTy&?uo8$WzThM?-+cL60O;p+yD$g) zGj&+1IgU5Km`{^TW|(X)>7@{-WpH4Br$J>qZSS5pJ0rc-bsy2-@9=aWxb%JDX-hzc z1E9pL4#l!YfP+I{acyr-S%cIS?(R1YBnv6xP(r!(BxT`Io2OB``}IM1Aj%W>zp2GE9V83Z|?&4X_Llb^9K&?exTgqT9p zI_R>!%$Ehx?1)GyKFyVx%B{1{BKFLNin}c3vcW~#;_;O6tAVS%p7=TVD&`9 z5!O);sM*eOzx?Mv{VJM;_Kd=6O(lnaFY7u-WXGVUDs9JeB{oBr^Kmf1^rMH=wwYtXOMX znFfJ_MVlU+r@QoWG>wNl12g3YfxA*79wiM+?LkWiLN4}mhVWWNu~PU8>1zppTTLku zPu((JZ0{=If@75r=&Ew*hy%VC7uWrj0OpGc>kvjS9TnDZdF8v$yIwzC!9 za)t51&Wf%23GJejUo>MGUFq7UBROt(z~CJOnjs}_^L47%O;ttO2#To;>s&u#QT2JFE z{&XRvP~XL(J_GCOra;dl*qt0N5Bj|=*rTGP*nh=`U&M5NJ`luR< z=tb}h`N?iaXAxa0Yjgn)<7sUXDeTb(4qskQ)vFctpt&UrQpow`p$V%m`y5v13~&Wz zc3C@mTxvS&;<^{h?~=D1IrPh8LbIo15!RCIj#i_zlEvYwd;y}jEFuTCY!yQa**ObY zAXrs*!|QGXeSFtvu26)3I;IdUsSh@c}q2u`F(^eo;kaOoT z))1Dfa)Nd_=N?Wp)_Eb1N}iN(H`l$A7hF=yH0ycRoMMx#aITOyy7G6PBtAC*JeDRn zgnb(yKiWlPevux3<$SgkHRWL;yn#pew)9Qs0+nWITLiAYejQCR-Mm0h!%mUXWgWZneCz4>6@n!nARob=JB^#W{EXC`^CiR`G zsc_3^4y{;;q4PI1R6J4bx3F^0!!6pA>S(xw6Gs@KEnb~}?z!tPxRrI`cVbO^<9zRA zcy`}j%~p%TNxe$#B@zDTM(Om6o$O5s-L_OHg-n$eOS3%rz2y9156Vx;t+1H9KN~?} z{h0VMjtrbBd@y1(Ir2#@bw)r;d=(IarA=p}Ql8GUb&j3=p$z{j0fmv&*%YE>!_8%H zuir+O4Doq?D;~^ehiSZ9Q0yyBb8_TR|-yUM*xM$)*r1AarX3j0;M1ZIGdBj_!&(#`#_Lz(%0JHTws>Z0OQZ2Ya3;|Ku zeUckHsq4Dblh9GN@lnj;9N(4PDsYqr*f8?Z`P-Cpa38Uh8g>*^`GL|Bp!jnk^O8-i zx89*S)j9D-T~Ni*-*LxfI869td*V@Z*}x!9=DR?R08b^#j@#OEX!bhrmyO?9F(18L zDnn9#nu$mO=JAkLr$G0Nv@NnaK5K2UCaQ>qZ?6#yvRi~~$k}l1QNa&H$U$GP!7NvI zs`M@^*Yx1dtNt0JOIt%7*hvIhSJWxaeU`Ztx7&bWOaepmv%--eW(cQ|19{lxxGt_O zRIq!T{JB`RdS2w7+L~4PUlvn>xJ_~QWgbm`@hGSw*Rl{=bOmd3(}iawS_xl_kts$e z|F7^loX)s(j8U&~$*oveRKbjg>J!l0@tXgsWt*N0XV3g$GYeRqujF^Z@B4txpZ!Hg znSm}RlYKW5wkGGuwfJzG$;?q?r#{z3H@d~SDiQ6&HPCqj1vLb)$Oc)3?ARnW?Lg~) zoy;pwKwm&7K%S>=+wKAfC9P3ukhoh_CDbGl%+9Yx;jFZm%WsIRxXBK2_SQ+!#D)y4 z-GnoSngGupk1e^$Z!|+8ONJTgR4nKVd@-iy+)fy9`<6MzZ=dfdtf3O!8yBeNq;7rr zY^0qJnp$J3Oq5Zxuou9>y3MHd0RSU^C=nY?_2lcw1%JDxVo`~R$GVX`_g>{&pZ%!ozD<9pd$zhM`xf6-^vI`wsuPn) z9Gb^FG2=SmoBxEcfy4M*mq^5YGBdZ8=9{)~ zkJ1rVqe%@eYNCzH^SbEArLF6K9P6lsvLPy2cvM28=@S8lJ*4<~kskuP-Ro-Vk}Tk| z5yBE6m<)J<-x4FT;h_{a1?(~=iy9~(Ng%j@KCg6=3$M6n6xt&`Xn(N6v^9J3!TWcJc#bcY-+_FeT9!08Ww*?0`$>0+ZH%{TC_rtcEc~ z%Q$wTIq55Jd@&3#6;IPWj(N1kD9I}Gpb@rx+ilG;ili~kdFDiGe3V22hZROUQRHIy+9)^0GYE6nfuyU4RscjJCUv z$!{fGhwn&WoW(ZONDezz07xJI?yP8(@H~YFcNZ^AkFctaw1ROXpizd_ZMPA|V@MgY za7=Mhr>bfkUu=X|Q6wzP!@nX;Of2yQpCo~a53@^v<)$iW&Ze_}D1r_qOE0)Q$FZRq zklpuc9>xI^Xx!LnQ4>;6MUXxgZIVebMa3_&uP$NKf=-#6vku;v+Fng7$Bv|mF~O8o zGNKo|!zLT^ppBUoEN~!SS$p+4#!EoU?jP(}wP*lIZiLsvWw=8lW6+rWXKvSq?v{H* zg(*j6!*u;LhTtxL4>tt4sLv^iO>y&5n}=S9vz7!)O|EUl^N>Fx(->i$c~=Z@RKa)z>U$* zh*h}VIq9MzwE_0F0n?D;+_eE@xu_MZ{-OMAa_7(271+0we8!}^SYP^SF;?8>p2Mr= zJuJ4olDh>*kiIzhy0D%cn7y%zuhXS?5Mfw(d-H5*F|dJpR}9rL(nX2>AgAi)RNby@ z$eFT6K_KUUUFA#akdg*2u2n%uId=w0F`<4QSTWZz;8ZnoG3|QP>XCZ88KU~$#0fd>A^AP)xv8G^zbuJ4Fdg&LWwJyUf>iQ_;K zdFDz!ZLt2ce}5}5O2;afbF)><&P}^w9Ya7`0i1w;;<+f}T3-)=xD1_WeI@W!f%0$o z)$Pbw6bOK92QnVKtpW#v>N^oa$7_%|*TnBu4BtD4{xibHk8~*@5tQ0IchYJ`t54fT z^IbbgxdhQt6fn&4?17O&Yw8$YQ)l4c6-rac6z)F=N>$UtyVHBHGea*vVv}^snH;!c zdkP?bMMlxok6zY@+IXZpZXc|=92BV?e@(_!c`$kI=1%ZCP*vcV?Iy4a1!LSa`9J6RFB@T z%XyYo6osfX8nrbuPWL@ZFTQOsOThn?J=5)=E!AAw?{@jKm~Pe68l^!0NyI{7802@* zyk>O6SaWSBBw2cG*}coQSD%|8GPtSNlu2XPPhn{S8UjX-%h2Q0qEE z&PuH1D$HtG(+6dZg9SR{mtQ&2aU%_ougTtNiK8M-+5&S8rkV`h^6h?dE{`}Szc_Z) z$-RxZ41g^_8xj#>bWnu%3DKp(%%E@76~alNTL(6-B{p=uVyBXDZUhK_ea{rMJD=FC zi%I@GLcR^C+%Rw#N~?h9NGD9xy5jU6_t_0Ady=;4v^FX_(fB$Px;ZxgOh40_*#nwIvK&;+~%y`&LpScT59{i9#$s{l@`SJ0S+@EGKU>DYz zW9vRZ`o0&p^F_xiYb+6eh=MJrC7^f!@24@smVpqz?7swtQ=cv3Nm zXl0CG{f8zvj{8Wqj|odhQ7MnByus9T>C4Zqe8K8HAi@6Y2ra@su#)W zCeO9bVL;~w7;AigtWFuhLv9v3fg4e!sYWt+&;84SqBg?yc!I$a@t@to?xM!tp-gKx z{_;9UwQMDu-=hsprW3?fN(Mumz-DXvRqv7v3RJRi`g|A-JWGF3t^;S94`FjQgsjVU zDq~}bP?r_cR`8`eg5Em2iCv5B<#KKLK7f*Xcl{l^q&q$>Si2U)zKYzNyzZ#`~6RI~39JaFnI}2l-K4tn6?%Ar$bevg4r;*IOdx5<$WiK)kFn<#M zrBePsqOz5L;a4p@hgNGkFIs|xsxJ<&RDl>jn?Byf>bSVc%^yj2X#1k#VA4-_`W%+;J_^}Vc*qG461nZ{c?ouzifQK+~YO%VqJ6RPs9ROEv zb1xaPx;ZHUJBGf2TC*fvCLS#JRbekCZwUFVzZsi475NX|#qFIYa>VAaXRp;{#&$posEk$0Xc(zf0@X_ql` zHsdtni+^`O^6;Q}L6`$o`FN~VM{I7wf-9wm_wH+o!XX? zph`nBu6+Zg)gVtT;l}@E94f_=Q)JQa0JNrK>dEDa?5dyyCtDi@4{%*SyM=xX?TRx4 zv7K)I!3#(?wbM&jphqB()t8=4nX#*b-P2`9!1Mf^+R=^`;y$+jz_c>TnWs>+gUINA z&q9jI9;Iw+17D}DY{<^?ztb za_8Zipx0JouUrUfT_7O%sbB7Calt7?SFjwt-RTHr1mJM)4hk*AK z4f{&dn?d9JoU>!;LAY{bdI1rA4LTtYiL(EE?#GrQ|neB%VP}#)&C^HMltr_^#6Tu^^DAOfu`xeFkOD7YGsbai18g73VHr*G_Yx zQ5<}=WP~k2QF@E#wvIc&OpbZDpGoNTftwZOW9K+k4ORV{Dk<6G{vf}FI5|89bfSuvs#E0tkKLq(f zU4PV&3FqK2^DUma$iVOT@Y7{a_aF2SLlLAoW;DSdzJRQoWzzFR~`8x;cQrOza@<$@iJ~@VmCg_zyZz(G57AM z-V@@?+Ev5eLpk0@jT5QKhdQaJr1EhW(yr|c61eiAAAdGarda-VX^dYF-J!FM?;_4BudYz|66%dah?O+Qu%n@PcAfLDm}c!^*0)mvak{%q@JQX8{WK)!_%t8rGfkQ0A9lU7l7?1YXP zv)g@ny*8`0*=*8!$09*ph}XW)1tMFR-m5{x+Av7`CM#s_bcrJO6$$)PMv~@9y3wSv zN*$y5rW-`0goJx>Xt6{8!t{?)<$Y{30qdX}q_EnSr_v z0p!q*ee;Zehqin9n5ij$n09qJacf9#D>!*tm_@8LsW$bv(3p(I`3fS(1jqwItCG*O zhAM-%^jPRP_Le!}qx?yct+our_tK1~XFE_E-jb~tkIG)v%y*bpa<0M_;P>gJf+{$8 zvz+tgR|#^o2WqMWc0acr$uvkpu2M-p0<%Md*gYqI!|(WtvO5ZM*(5$yXd1o@i;#x# znC>Nq!@yi=1;i?^HQ+g|$zjTLG^NG~=l)Sdw>KgpM8Cgvl~i@JB+$A7Ci>|dNf;h@ z9#s_DQ8Le~=|I#?v2ihAFT@}UFC4yM#veMj7MOUzr0~%-Q7?jlwK&|W;hy;OPHcs_ zWb##iU|92GvvZ)*vC{rYO_f9R=Nb1^=~$`Za#5nE8brrUtZ%y1pWkFJlRVxd?*q%I zbv0?>ZvvUpnjEV<5n{DW|N5Jn68?RS0mw}m1cwho+hY_;oSIK$DuLU=p#|=Ubje)d zZ3Kw1fgzMZm zh4=DV0#T(RC5c%=Z2r$^Vl1*mCL$=qNS%T}ZQpavlo7CwUR2Ea{_9c~Ba=#Y&UL@; zn-&9;qa3jOVO|d>8@H^yq&d7*EW65z|X^G!C1 z*!T}@i69_c&0~h;W*e%ZpuS&j0dSjtl(PA?>w1=wan-jy?)`dys}4C&Ka+Bw$ewMs zZ%KKf$AHuK<-}~^bJmo@iU#Xs_=>O0s6WH@19HpbYf&)15y#&Ss2-IW#DstC88czO zg)u2vQ=%mXBwM%ZAf6s(QV9t(WF-m{T5`HjYGLF)b5@nLajG&3S^^wsE(Mr>+h4r7 zLPdsuwZWr1UC#jFNVIL&BQFQ>IvDmTb4HVrM&fNefI4kr%WWrKqwRt91~YpV_f*J~ z614E}KBd^3q->LkLSHUMyIIVD778V_hR5Q;ZgYz`=asA?{}j`l$HO!CSwNG%JvaF} zVN2lu2A$K15~A%INOw#)*M8T3aWu$-r35FeQ4~d%wXW;c^PI)uyt)|jd!_SV^fr=Y z5d{XYP#Cjqy^pDqiG@JU@J-!qRfuI9HYDrcW3z(|{~PEFi2s2i0f;k-=Vqyz`2s2iq=`9m^*= zS5K`nW|c(jIAsG=GYI<+e%NBID!_D3HgFAl8R?UnK&|pO+vn-vA2)@m!Ml!Rp(n;- zBUteNVbWZcvq-O}fa9YfA3hoBaeEsG2kL0`$b@6feY7@1|2F09Fjii(h{6h4u0ZFf zO2my|X)8JwEZeSgj+9@2i8EK(nyE+dD3}hWx*;zWH{Rpf(DEpWU2j~`7RxNQ1p6+F(<7F%(_qGxduM zz3F4`!=Z7w7&teol+4&_=%dxwrr(qfBmMZ?4#xmKab*^NHiR~e7;$Lez%8N)=?sif zMhb0gNMu?0BS%pGK6!9@(hV`!8XptOMeF}&Fa@hJ%S%@JPVMMNa`QvIDsY56JK9j! z5#(Bx$Tvpy?CX3Yc^i*VlND_sz9U%roZqlff$|f3&(~R21ZfAsF^0AOZh$9FZ*#?;&fNcHX_Vr%E-U~Xy#`V)hNj_yyS zKiq|NBBSG z6E%OY6Uf%Vjp6_4YfBqj7aNcNS8if%e`9R&r}D;5b_{AZ=JrlNDbfG2eiy<2Ei(mz z0E_^jJpkxxWXA9(-Cy1EN6hp`{9Xnx4?9~sfQf;XBhbs-1o(b{_i!|D1_D41PCze@ ze=7bH!836HjLnTe?~VSxS>XSQE@fk43*h>f_?^kWRsYihlz(khs`p)LY-?lXe+DoH zn!q#2+k)O3g7W{LefD3u#GS0Hc4Q|F$!*Hn(#7Ps{%@(E$D-ODS*bU~ORa zUpjL~adTIov7$N1$n0O0{g?b-xrA&?t$+YJCRTbz7Ix-;Y1IB~E35bZeV;+*e@56J zJ^O#@-doei(gx`02w-LWO9gx{f5(3je_w+?v;hp_TEe1Q5;Xr$@BC#ZW@BV)Y;I!; zU}j?n7&tf>xWO~NHv}^q8^D9DJ3D|u z_>bschy%bN^1q0K1;8NnH(~`aNdJxAE#>|faWVrKl>bKWeky;XcR$tte??sHeg=Od zP5^`9--rvqVD!I;_1(|teRutB&BVwEU@-m%WCAb%{{i1gnEVZy-=mqB|3m#A&+H$N z4Z!eE?K=~zf53NV>%Sq>dt#e^!1u(q|A6m_?f&NdUQs)sgSqWLm@>WRcK8Q;&+YgR z`0fGv2YfG)(?8%lGv|N6e|Kgs|2WHcDsF$l|Lh*cKU4It0m1lpJN?h)kj z!pZ)pJZxV7;cN7-N&nYmdGGsw<3D2-00g=Mjo=sOY>jxnStQm4e-(L){hTO6Un$Gjw7W6c3c7GUyiA#~BKT<4n~e~j`phD6=wJ~%QeSsi{&IbX&SUl)A~c5HuO|i#-wbqwBpxAQmZY6T z)`a2dvVz)jIH~aO5DAY@4|p484d%LFEG9RKyEkvZfBgy6w2`5PL4O)r@os+t=;NCg zh)(?d{;swBX{L|RlUM6Z?J`A^Vxr-rI=jR!4zj%N_#dW|B%7XQ6cZWZP}5*M1T7<3 zeu2s|$J);QsI#S}r+vV?e8B*w!aIDziRXzDYNGt{jR{^1{5r39-uU}(EH6F{4H!Tu zuP&;Pf9cX|yk9R_uT8~kU0IIMF;(QsUCo&`319K?TvIoM^aF=OfxOElpFg86lwmoG zN_j3KOeD}5>XW-~-ACNWeKxU1M$1k`z{3lf?f05>@@6Q}H>t7Mj$ZyigZMihA+BsMy zCuKi3!@n)kIFGh{Laly=a8f=6LF&MMC$#J6p_4`>2m929(2A5U&;hT%)N)Rht3%54 zku#?qx?LNjhvox+nDMz{Ukc@{`-vGhVPhydHEOlF1tughhnBNVn(D=02$1vu7jDJ1XbAG7~ z>c(d)7+miw3XYQ3b!Fq2s`WhE;%*x2SHGD&(k)_y9HUJ%Q^&sm;0O8fT8129m7`6k zx5{i~n5Som*5@Clj3qB)h-vqoWH<48?@*y2g@je_K#mkgzviAWG;=#jjd_%Me=b5y z6u%xLEmTAA7oq z-2YK}c0$xntRj{zbX<4C^LziG+^N#*%*fP~)XDWP8scxXKZ2I&pqL1-Ja5`J4F(M8 zObP<9UHkntxupnKe#A6`aeAP;M3f1^yC z#WdxY!(IrWN2WhuUna{uajM47Y%eGu1N~+!r*(C~XmnZkt$+O8GMoeKlP}mDZEewl z5_;3al|4b>j!Mgj>>fk=EYwV`@3Q6hMxG{rvccAD`>5yuaVwD+#}Gpie}h;&!-p?< zLK^IK>37-@}&GQPVBc``w5JzhgEzqUr8P3q1C84E56BCl%XgtIqWJ zz~!IBN#_rTQ%k+cGQ#Duf66KuXE~vnHDKjIyuR<$11-h+q=LX!7Oi#$F##%3W8pT- zSAw5w{eIYr;vJiB|K(<1!+Y+u1wQ_shX z8GoQvz0Re`?;2u=B=wJZb9}oJ=pl$tB|D#pP4mlDXIJny{or@Yg2ab<#x`cK51bak zex}}E90<;u>)Nfze&__#hw*2_cyA z=yzLt@BxEU*HU?FlyVLP_zhw1eC?$HS9_?2~LDvMrJz&Xct_QKKV@0xm zjl{YATj5aG3m%mrZxyBa8;>;EWX`9-#z8uwX@5+DpWeAJe=XsuQLM*qaqD=+94UsU zPajK>mUYK-qud(V_G~yCB@{^bcP?jWcTNo1F06QXkYu0$nnpH+D zg7{c4!OEm23UytKkE_v4a$;d_S^0*v4t-6By?A99sF9{vX(y$EbC zM4$N@;LV>Kb>}#iYzY=-qB@xNX=i|~Qe!rrCr_#ae?(_O5Mh~p7ul#>?u7UP3wBCh zOn5!9JwN-oA3d;a`>JisToYIxVtH%zp;$MAIasLcn%~p)nZu2rGEMsm>r|eqys%*3 z-A{oaMd%F|00la(8b|HVT5`k@^KnQg^eQQ(l*xxgCHnmmZij^39~P)xbP zw>;KCUp^%D#)EbtoNo~pbT|~gYC%1Fwl|pK^gh2uKTZ{6Zx{K5$Ivvl(baL|TpaBW zl|perDEdwJ?0?TO&|OLOhmi|gY)>%k!%$Clf4!h~XDvc+q1~QpxF!A2a~)Rj(dQ&< z*VALA{)18KP@MO?W!Gs?y|4A-B%?R6tFiz&c2)OM*u;Ts6z3j^iL$y&I|TaIv8S>P z5Nh0b%f;>219I9)_*@-Rby5gP7L3|o*Z>idnDij#p4N|6Az0~R_MyQ|4fYo>xcCe5 zf2BX`nvoG~fo%$}10M-;;U{=v8}-&^;z*KL>Ara71yx4o^U^f*^z{S*LpTo5=s}TX z9r71ZbsT(%zb(b>eY%lDwV)W}ROZm-1(;1wTg%z1IaPP%EPgUJdOS&6gQnGBum(qO@;P!Bc#8k)*Uf0ij*GInW5I#;M`Ds$;eWHKJ_xw!fa%KL<2 z5XjgxE`g(hj4YMG&23btXLT~Qu%2bJII3N3w&WE4!VT7FiT zi2L&$Y2h7lrL|vU_#x1UIMaXecSYY_=sg6tS)lA1z<**v4jCc2KEx3iH-FpjfBiYp zHYr-Zi8TmgI}pXO@KNAT5}~LAV~JSw z>Ex7Ha0To|-xiy!9BW4A$AK(Ae>$BwWos7b%LOI~|N8L&IjD09tL#hsw>)D^Mm9o_ zpzax0)~7Ta9A4JimLEZG>#U4UR|-a5n#Cxegb#aq+m>UDyM`{3#%iLQW|dzMi_qt@ z>48x-moXz#lu9O+(Fi9*2wEh&ApL_g*(~VOCG&bY7*i*?BU_=WWOY$_fAIdQo_1ZS zZwY?kpG8fvF{Ju=N}6f~B$MmUtJK}vi(-c8Ld2>=W$x>Dli3s^39NI0=1;S-@@KQ1 zS?2X$?yearSrx4% zU}M7Ia1ZwuT{+k+Wr(O|e=2W1LxV&pGryK8b=YIi&rM5=5WuZz8q-!X8`8rWbj~YL z*KOLCz#H+*v_u^rHy0n7!gi6Z^z409;Sr2?LKGcs9~qi|6<=iXhcr^f2)Qva0oMtX zp;QszGhTa~y$ENclssRkRSo&Et8p6-QnzC(J!ut3p0IGYx)dace}n^8(s7Nf9pv6L zv`4=>rqGDA=dkk@C=`&JJv$TPVK;Vn4A7}%pdjR>jXb40-$*#>ZCoRf^(mk%S^0I&T9G+f2WW1$<+Nm%vT!xuqo}kJqd_G^(W^ZN~3qGEIgfJ06*_8RIbWi z;1`JgN9kx)b7TSfg;u=tl%9jER$yO{i#wQY!U67tcH$K@D4u`oXCv3nCw@xC|VoW41ZRIxNTk?e2)}--ZD^WHsGw${6{aq> zDFlEKdH>Fn*h0?w%|7c9-QNAEFxhe3jEFhe`c~qz_BZ)TZyp&}=^i=! z9%dB2vr}&fpXtjli(1S*>(m{=4p`$`4nHOg*CwCEf0R4wF@7Jm(}-sb74kVHSRBJ` zZV`@{x5>qrv^5$_FtrvK6Y_O~lR!g%0#~d|ff~n+xTjN~CF}W_o!#bp`s2K)7fu)q zdWs~im#m7BCz&`pFULT2OtoThnKv@yRBjdGEU0!F{!zN+)g#~=8YdYLDb!HPrbNdA z8bYy>e;Iddd?b6ibK*5r!Iwg%{o(a>IC6g@Pvw*V%NeaPkcu;1yx-}xiE(!hyNGuQ z?kmudEo(ckrwgx2#FA#62B7z3;(U1PWW$LguBoz3xSB-NH@Q2Ts2O;RWq!Z8~a z_EvacgV|9eufmXrnNDN#J~AQ*xUgLfD=oC=3v@rv$}7b)tBdp%oF z`a$0t!88p?e}f+Jp;*i?STs-A`Rb>lJ$m2F7p_5J9W+b3o=1eU|>-1WhX1Zj|7HxuE3Ju5Q1Rbfb%(B^6S7bp{moVz3xp2adAit}VsB@=C)%*J=Tz_10h!mm!vAW0 ztnDr9I#_>Mq40Db%%_Y=7WhW~ln8N~bvU6gBy62ib0&eJMPu8xZB1<3#>BSs#kOtR zw(U$Xu|2VIa_T>l&E@JOz#(c2hpq0Cbw#UasG>#^MOvZPE>y^4x8p?!rOqQ>BHEFr zP<5OkKe3;wR-h^vYAtSe5(`kVbp<8e0gLhHrLIHUGVv9#a{T->H%mLx^qaNbEK-g( zy%g_=R_VB*W&bSOEj zjXJu@%qeLW0XXzwrrW%~o0f;tpOdTDrT~ON?t$c~(i64vC-|waHb-*#XA7L3=oOhp zn2=YS$a02EaMC^k{M;?$yR>lA2;gY0{X-s$+PfY1nKTYgE>^pi)|CP~9BP1`Fb9$i zZ-pwmEh?wk=+1rtkWeghKk@9&Ex$D{=}B8N3KI+lNg zIyTZd@9RFl#5zO3!rSXG>{AKkk2*&wHZL-2%sKEn1W?bX07H8Yv(^>+hbp2<20xQ#MEsoX+ImzmWh**8*GL2(xXrsJ9bXP;nKSjg4j1Jl}uAa#+M#I0Cg(6O;&Gf`%EFXz@-^7ZY#qW zUH{(F(kE;oV5-uYmiLF;jnOhTx^<~p(Kb4FG0|U%|6=ZHJ2sPb!dBQ0*_zxADct_? zIuR$l%bu#P#Ek6^qx^TL&&>-FV~91D;lP#p?J{nXx>O&CtKFm;23(yz)R%2xAa{z5 zJ5fN+tPKNm3B7l`CHW};+SY+xkR3l%%*lE$d*iP%z`FxY%<=XZ9+6zf?rI-vG#P+y zA%le|E+2`UAoa_%V4LlOFW;ywKO#fF&s6IuJbnN)VHhYk{S!wLwJb)cCV}ZnI9s2& zA?WcP$bQIn5Bj|%66u$mgGIoJPO(9=%G_S9uSHriBXspa(O3mpKy%^D^FlwU)(RRZ z#PUfc;7>m&{LQ{Pc#(?3ji_MXP;GMqW>!na8VQ*##Z$t*`3~_+%fs zGTAlS%DcTLkvq=CKUR?C<3ZI0eI#qjM(y6jJ7@2gP!Kwowf9B!b+Tbc(~xLxIJ8NP zKSpl$9^^rOE{Y@!Ecqz1=sDK@BqVu4u!4J`0HGZmMw0?Ek74$j7lQB-ww~;|@Ck>5 z!W!e2&y1^$EF2R*Mjf2Sy16scZ`iZ22uk<4+`6@c2g-N`c~aGO&JVrxQlD$*sQX3O zJzd3{_EOo6Qa>90O%Ww;y|76E)g0%r7dWj(8(4DM!-mlWM9()w+t8Y$`yb4P>ackn zAP&42OYQ8>rpPCgD_iDV0(~2%^9vj6kMlCfnM$HtTN6nl{Ok`gruQd~U5cMUZh`}| zK5WQOt=Cx?Yx>=o2%G!}VXB8sM!rtEfpS0S*; zKG5HDjauX#Y^TEemDbEx<{EO7SK4U-a2V+uozHPy(z1skV%4ThpW6gu?311hrn-!O z<Gz{%Sddabpcs#< zq!!l7fHpyBxvJvy0`jh1f4!t!x7!u0_SiGB8IK|uDn5Pu!2KzGlTPTsvOh50d3#%1v0fY+T-CNd8ACEqP*3DojGo}1%-@}Ha^(JF?wi9A z)2*ZlOM9yFjeH=C)U{vZnG7es2vFVne33x^+fZF{%fM#WQICDzQav07ke9GmD8tnS z(K|mOlR-1AX(_l@jtrT&J!K!!=|`m5Pj48txAi90km8uYLvg!+xKI~rAFv$Rtc10| ztCBY?HoBvwsQCN7;>=Q!!^@B;7Qm5+@ftZh<~}pOM}6>XnOU1P8CZ(UwZoIEC{zBb zIsBj(Ieq2#SecMoQEiJBV9ia4_bth}sHB2A*fcZV^lD=cT%ER7xu&5$VK{sNjrx$C z;#^k(M>L$A`ZvtbpUMh@P+X-LFYW+vHV5@cP!rO;vVQEPND4uCqmGFSoMyiL%~lM3^Educ)0)Gz`4XFF;<{$+ zXOYhL%?U1wI8dHu)NgE`x;h$B^B8#UG$+PC>x-2T&cj5FfzH?YohtY&j$nTLiH|wt zH8@C{dUMjkL&%JS#4%C_Q{vc+iY%ufJR;%%+c5^WFA_stK0B}+>S15 zn!*WC(!2{VjuG{eWO?2^r{I>^G;afB{$lYNgg+cd;eRpO`5!$2oxA6<4OXP_za|j@ z^DiZB&#*JxkPCI9#jjU0QLcKGnegzU(Wa8Y*c(yYFcL|607M>SKf6BSIN*z;lL_RF!jMIdW|vqViGh)V_{{pYwLMA?Yn|$HD@=r*$*SvaA5-gkz%s ze>S@3Ii~S@nEMQW?HAgIxZQ8rTb;w(QMgnzL=@%T8}p3^OC;;1dcWf~Dgk(jZMH&e+O!q~q0=%e|J<=%Cb+kq z=hfHwD!e;1Dag~0prZ)Ilc2NRJW>GlIL+{8c$J6CsOiXv1n+@yFUL;FE6AJ*xO~bd%jij{4 zEq*p$eMXQx>OoYN8!Bw(EdLg;p@NjIOnP@oF;ugj_l3C1|n;D{)aaa1|nzqiI543T`a&Ksu z9)Ny9Jlk6-M_^od0dF4n!s`RNHkPnnQuJ26R=Nf1OAZd^$&vA$D1u)x4cnWQ36Ss9 zxovfrHq5ysawSR!)mN8$5M0(|Ee(!2*W(b%cW%bW)LP#A?2SteekwnZcNloFqbRL` zW?Z6xnM$D?sz@R>+c8lV@~@x8!u=ySW{ZAaYoR@Fy(#asdV}wS(6B<^!>u9ftHbl^ z)KGnkt>CZCcCLY8*2p*VW+MIm3Gg=~5b7TfhuC$-a8d>KQ1NUihDqj<`lM_MIuQJY zqlCHxr>}h-&6zW>8lEh5L8uECBSFuUSH^~kGHhuauYpZ)Ft?2_I0LjzQZFPjmgh;%>N(??6!)Fs_Ux0A)hVFV zTx&wr-AVT7SL-RM=5Lru?$%%YztZEEAf>e^Hm(cxEa)uGX?0~p$OdoNuvKz13vy~% zD7J5bWU7N;Y>B5fxT(=0hl?XhG>LD*jXRR%bYr2uExIeT|e-QdrE zGAVMtL>&z3q=eprmdWA&u*(I--1R%uhI?GXF}YW17!QsYMF-IqGs4DPXsx{D+kdzm1BS1Pck@)i^7JlbT?U#BxM=z}Sw@6?_Cr|tBs#Ks(o<0G5g}daTiggN z9;T=W?Ni8M-(pj-qg>k~e1y|ikgRks8nr1Y;&pVeC&VSpY5_>%Y5sCWRCo)PI#j#D zX??+%D6f3JJ)!PewG#7-%|uY)JfJX~I?HfGkCGNJ(^}kIY3IkJpB_-@aE*FqLF$k=`vn7 z_7keR;6>(>S_VXi_^rwv=urH&+fmKfvYzQLFx#bigOqhz%=prY@jskXVrT+pb9b-N zqX~&|`|EaCA9b;3OsoLQpDL6)Hda3@!kHqvnQCT{(TpqSWsskHfi`}NVR2N;L1V|< z=u}zRtQg{hd~W$e4%5eU{7c>wBKp(=2M+(-Ep&Mtv;!#VmQ9S%08eirJdnzwE9^nT zcFgt%*L(N7$m*ELYU(X;XRfD1-}t4NokeudHXlKy$)0Y;Q_Fc(@`_hORDZ$MB?)l7 zii~uV>)%{GFwhb0k=a5|GR~hH+mF?a3WlDkzn2xDo-Waq5XVAX%DesT=l($|9{e=q z+@GHIL<8v5EHbK+Oi<-UbSs>^A|onYU~CTr6d)F9Mi?;ecsnS@qf#zCl;C~e3uhX; z*F_`!5-)XD7H{WIhYCbIuP85bfc~=rQ$&GD-PZCKm)O?DyMTcP*Hox+lzZ^A4Q*^R zgKQq66u&K57>77$9CRyRcv)=UEkLIeT*c!re=i{P;sUvTpL`_reJT{_sNgcid3~ga zZmQX+|KC99ogF;`D{Jd@IwVOLWTa22L@>1FcQYf!pPD3skpqLQ)b7(!cJ{bF+dt#C zd6L>8F5aAtH&&Vtn5D!^38mCV;rd|e)W!YQriOA%O2nBX2kB|7;Bk+z@F-YSYY=5m zIog2ILpVRe2FsWrtb-H+)uRWi=U0?uj&#cR|?Io$LWYhY#mh)I;+Go?T)y z79%=n-N#2P`912(%kJK)d_M%Vn_bv8A14e>o#@yC`#1d3O@tFkn&P7U=~4@0k4LL5 z=z9>)PHZy?-Ag#A_cqi&o)Ly(B(Pdw2dMzZj*}q6U0)RFOwS!7w+OP$GHw_Kg*Dtl z6!Yl)g;>I#-*gRu`5oITy$U893FnDNr^@4^qL83Hd9F|fFtPP*88?}Xc=!W13-alV z*OjoGSu>IIMOvng^!$Q?THM-6W%yzylwwK=77{6&tn8`{-uH*G=k{>1xZFTEC`tgt z=jNG}Ef1Y==4zUID`}6xncY6iMDQ#8pc>nQ z(kH9T4{@6ikI5Dq5y?pZJGC}4;+iv z5b8VXMdo#~EYWaTm`rKEsd}>pv;~09Vu13m=%Y&>39b|)jc4#0g7oZMPuD?g4X?75 z?}7jfn=#I;C%gD17=^43r1n4OMK3`!dSh&t<9;JRBAej|mFxH190XVW-iRJOjIv7a zNsaJs0pq1K&$7ykr2&XKCKwrVkLpoouR!Q6t?Gt}*lmsnf<`>$xTa1)c+CKO|3XTe zKmR1Pyhfy#O;@RMdN?oFzj;J-RtMzG<|p?vM4K|@$B;O6)GB*4`M4o{J}}pKTxHG797hK5!j_)4cu~n_JVA{qdHG`Fuq0nrhi{ejtD* zDd6dXk=8**x!%3SeYhKf1x5s1hNCQ&5|mMNEf`qAXjQu!QVrx~ERn_gm5OJQNJ%Om zT`$fsxiV=fi<;hBCH>HMH=}R1Xu>WTrRk(nvW~p3Z?0@7G@&3wb`Gc>X@p1@b3;K3 za8q>#j;(dsk#hgwB(_#0K}5gz7TXb~@+W!Gd7UhH5aonYug^tTei#D^MkE=>8o;6? za#Je;bYm*DJSkRLClG9#8OtN84#A6vHm1`;7r4AR$Z}CV)rg*zR!T=5TyyK^AUw*EYrX-74Uq9(*&pkfZBTR`D8Xwa@GCgU27P&TjxYT9&{(_CfRot4$E( z^xq?X8HQyoid;0*=n}&tnzOq-p`b<<2kr)2@%_VW5I3ABz2F#zqp71~)}f--NzX{{61F}-{$6rg68PSaa)ac59! z!ERn0CiQ7=BB%hc`*>-u_uscZ!8GY_$+%|U_HO!TbLYv*ka1XG+LB9eN2#5EGYQ6` zOp&TEJ<^Rx+-h7r^hR~`txtYYPL6>nU0`xqx8pvgh$M9hqsUPpyw(-bEXz3^xsSX< zr&v+l)02MTqiSvH3G7;9gjR@4y{>}XbpZ=#H8h00ImiS4ZU_}Q0|z!e5b{A?Y}Z$H zKlqaAD`SXJ4*%7!tDtF8JI=N@lQH1JT3Niebzc#>9#&Ns6%UveahQnpQ^%xd7P>v1 zZ5Ogx8{FDod;ZB0ImMxCN~^t(G@0E7rS8pC0y?`%bDR(;d5>_-rK+|WNtzlE>+*GA zSP9c5?)CvXy0i834p8eYF+L1{*O!!ixlPt2+MEv_znz-(TyzFgaOhO=q+EaQdeNYR zMY^HDRO(2z-y;xSOg(VXf4*o9_Y2mQ@56!*QLZ zz4dV^(-l(~Cuv7_!p1+TlnNH|!~_%QQ7SH2oFpI$5$X+IJlH%27)jaoo$$h4QTN&| z&U(WngBa7jjU<^jK?b(Cpo$d@UIg4*UA^d>yQ(_(5QOo7U(#U&k2N_(+fg}1f5Ln{ zTv6U_J&MJduG(D-uC~>6lyoSTI)kI;uRN{GHFabGBa~k|4XTI9ni?;2qiTR!dSSzL z05O0wJ9%8>c*t@8Dc&3bhsXa|W5%K|pQ6LfIpAL#hUD<8Rts8b#)8Hw}k0)iDWg zpt#ZCj9T2bRJ%4#*V=2pwO_vi^{@ni79$||pCfLqeR@ZF#%Lqx)S>76>CO`kKOGzY z7Y90tWn@FCHCmiS(|bvguBzDBY@*`DpvOLS`BZQ8D=0 z655NoRI@2#ky#`%ap2P(sp?JxR^8W z?_9LsZz&%0O@S7DjQN~^v)`2{{1q|8Ie|=ShxzblB5)CQ);7a^BoC4nb3)}-i#HgM zxpw0Y8A7-ebe&C>Zg_LF{H}mif-+HczH+&AAtbyDUek_yV#@ExkBQg=EU4O&AiSIEz%2DP5%{o{VQG_fQFGL|uzv zd5tRpLy;PDiLZKy^wa{m@V}SK0ldCva=(5f_C>_Nr49zYGi(sWoY&%#si>g5T(x7-e%$uHRxtbb`whys7sZ6|$C(mk^6g8$~&i72@F z9HoI6@|{x=juJUI)^zrFOsjmmFs=QL4qe@7hC%qcFKuQBn$)m zUoUC0<;esf4Yp&nJ0?H-?Q1c_grppU8#tdjrj$qMq!<50Bjp zd<9E%PNa)NX>(X~CU<2yeJQ`v_H57C*<0$OrKTVNg^|JduoAfTuAj`8xYvAI@!+Nf z4l3DyNtXVe0v_6Z#N;e`)9FdS;nJ7y6$I4!++j<6{t%Bo0&H(4G}_DtE{rNtgMHyV z_>Ap#iEFm!oz!UHRfP*8G<1^E-azz`BDX)3Y#qF6jky=QscOv!kMbs0Y#eV2kLRnK z=vM&%&6_*>aYP?=2F#R#n1n4kLBY8jOwS6cz^0wGxRo!~@w_va9E#bpL4z@#}OlpNL%TUt&sz`-vXL?#Qt7w;%PQDgN?cB*#W=Dw^Q4) z#!t_2qOv(hYHg!16f#xprN@BvU=@Dp_aQ**P>qmR6TTbl1qR} z?_;7y-iqVT7krPQfg*-!l|vJv_B~1arQ$epQ6UEsW|V_P#5r(1=m9vnOtm)ynZ~Ul zWzy(01LaIr^gwHHacD#Eo9(1e3Tv74_yrT+5#92XI=a5|7`GdCswx{| zP6HX0p$|)0_0z~?T4iJ?F<>u?K~L!XJ6)o5XyantLF-kVz%BId)@dB3e-u#9k9xLw zahiWuEYoAkUK|oZA=0(Ux)C%}C^P^|c}+-=#D?6xRnB!~gc z=+dG22p>Y1RVN#qJm76ER3DnN2HnaL%ge_rSuZ%WKqdN{zM$abG+6eyQm7~hkEQa# zTB5?YZz)Stj4@s&GL?JGJP2_2_c3+Ff^war^Te&oqK2BA(gYSl^P;^=R)zY?r7#2A z+E1luX7nn;LImFLTOAPm_=mn3w6-dmW9pR4_ucW~E$1$Q11^tWe=L(`8e0jUG$(M4 zds#)%#i5MOlWVTH3sLdlx%6x~%)3Tft<{^084AjDJorU49-<7L8KMpzyJbtDVj$p6EtV7Gqlrh%xMQgFFW5Yo98_qT zK686G%BIgrSLJqWqaJ`AEeuQ3>zG;jg_R_#xfp9F)33ywU)~7^9bZoLd@C>PG9VTT zgiDx(_d_^^)rtd#p}nJ`--CCaHjDkqGErq-Pvl50_2Vb3A1kqxQdo#ejZe*(ey#)i zB&0P88qXbB71TiT?rspA*W#lBxEqnrnp^Kw^ZaM}l@1;!y zY7l!>z>krO>;e3O=Uwzll>TCs2xr2_j7fu((i}Xk>Jld_wn?X4$8S?f z;94lFsK%|JPZ(ah4`i*ET1Gsr+s&T=2+bV+sfct(=}tP(Fm2}tc-L;WxVIg?$KSUj zsS+CTR02wq$J5`2e2>-s$(mClq@vK!eGUnBwaFb9DfNLly}rbrBBn_N>7WRt3uRK6 zIS(j+HE|-^oy4R2fohzigv+vXE_!xRN19j5S^vowv>!&zRn<>Q^G&FW&W5wB|4H>S z1SAZMp~a^M{48r@-2Qoo+K7`Ol)a1OKFi~8L;}3R6mEaUL>s)6r~i#KK0@%@vnu zNXZ%gWn`3Jq-9^Z-F_*s>qNMT{-M3P%q4)S&z(X0%u?Ux7Fd`pBlG$B1%V`|r^Flp zV*nD=jlD5opSxi7qo}CTaVZsS&ZAWdXiQiGf@Llw+X@ih!7fXzS5j1LAYPKb($R$} z91?bsce^uzWc2B(aEwR1X(lp==uD31vXhPB7SY^C;&WiAuw~~JcJdrL@&_5O6>Y~Z zMmAjzEkK2jfM`LVjAOR8dQ7OuSX-?1)&aFbeVqxIvkY8r*#<4UBz;~lOZIyEWFF~IBtH7ExrhS z2@*;jmH=cD51V0lOU)#L1m+>sYn`-~a&>Ckfqs#m4(~_%F)Ywl>pKtZyW$H=zW57p>nXh3^)tn*?9 z_pd5sDrO6A;mIW9(|rRbtZ(#_v~$lWw;jZHZR_@ z(0c(z9RctAdZtHH*)1+i&CvMKJG{E5$=RF~m=z=#kh?O?do}vlf zfz0ygS`sEsRX`m1*XmvtA`1bZkld10Zh^saox*M3Cb2WpUq~JvT>x%yUHf<{&V8ux zUL=YSdc_(O&3C#f(L;f9t1ETHw$0#OH2V4T!mH4zuT-?jNUg#7F_nE^_DOmuY?;lIHzOhcZ)8o%vA`lPE%UVEWcs<$ew;47o z^v6%*qq~&h6)u705kMhxWs##h3+`cUCjrmXHn=xaJfIVR3DWS}oFgB-<7cPn*>8FB3==ayNi_@cT$gNj%+TXur0NG;MT!LGzV|@EB zp~+{C7=1i$JzEu@uuhXb6BR*?5^_mRAAMam0um!UmGEk#L45#cw*`FRMs=EoTg69nuy9H#W7%I@`C%Qe0R$T6rGV$ z*x^R5zf?{%{sJ{%-A_qGL;}XRn5>ual_XKuWX{84lK^j!%~wY%KQ_RDzJ+ABaDhaL zr{R^yJY||JyX@rRtLl+t0&Qor!C^I%e_RWhc#A;_JpOLr5yf@Y(7Wu-<@_17K2bNL zS7A9to}9?-KSB$-w&RA!{6~ULHSbfXB;PxAZ;8T~2T@bX?X`HQS?L23=kc;koJtNO zh?9&!aDd}!D}M76wDu){vDquGOPPAU%UWz1+HO)S-Rry2I(u=~ z{g_hW-%>EBRJ7Kx$TA=(_Feg8?)5^5LY!qt1VDyQP*xZ^mZnY;=4!_JO5$xQ(a>bnt1I03WC_(38c*_M4TzY9Jcw zu?>^l!osIAENe#<$dIA&h6tGj&3!SNXqQ*i!=im|h@<3uEN*tIUHsDqHM>9dqn=vj zS5#-N$`OktMs&zBGnrNU1!(~Q1uQ#aL>PIecm%#_6+&2|_cX-ClR5grtT)~SKxSTc z!)Y|ho9a^+KSkI6s@7Ze4*`n7L8c^xve*CxLLNxUlo&kB+P~*ic5%2ODeI8tlMSLi zCe{@NMG_EEm{ep^bRaRQn4Ir5E97N>aPOG5c4ZkmgGO)x(FX9-Tqc*L>IlVJmbEx_ z>P>+(GcaJ~Cor#6QP?qt?2Bds3kh{gN_ z-1#2^7oOdDL8FNY^`xI*Y`@F&a$&z*{YP#r!*b`&%ZFK?&tWr8a7Z2iyQR8-SV;e~ zc-q?Wo<^FOzIsh(v@XvFa}rfEU*N(+Z`xlbwIq6C+JfpRB2{QeAoEghx9qNqsj9Ej z<}6HCCu@4(#K8h~$uiss8vCHyXG20NS&5GRw?%$|AtC~9Wc`v^r&`LTQFFg!58dH$ zUZ%bSn(DB9yP70`|B(>`fVd%{u*m1SdeDm48wwdfd^2yCKI>?2mTiG}2H^xu#9vfQ ztDu@-pjH0eX*Jif+u}F<476Kz$JQGD0Ga+1RR-CP!i-adt#(7AR^{8T5wUiyKq^9L z5D{dc)fvmNEYI^Rmu7FnSq`Tn4SdH^-DfV%5xXJcW-+&H?&^C6c;Bvg?9YiimJIAs z>lhhu-f|aLDLg4}l~a=Hh`XGTcj5^pu)p_yi*XYr+K8%s0jA(UjcZ_UkmL}jTPo8l zsU3mW+>+V@To~aMmC{kUf*;BRj0a>*fEE}kGhPW$V>{y=rkAiAQyJ7&0OPugBVJ4O za(qw~Mfv0uO^Yf3KU8(2RM`+bqjj;vC7Sl?;Gx?)!yB4KVYA5nGXX3FXLPC$*4*a9 zw~2r{IuZo&ps_J*yT6*$2#$-D<5@WH7-SK{CmL0Fhtwp%x(A<}K&EfnBI~xe`r-6| z=f{Hpe+-fh+Dvzab@L-U{eM|(1rw^8qaDSuF>C&{qol2qq4 zy?gUxQ}n|bz^h;Rlto7aIOb_aHE!*mDo$(5KN1wRN;$g^S0bWR%Y0s6J{XhR^(*vRXYp73!Zt8o4Xyzx{BWBmUG$l?B9fE*My zX72wd6*e}Ei!Q%&8eR{CokD2AA>pF_sMVsmRN2Z>h_ieM8z zu}HAtM7nAW(Q3+v3vd$dJ+Q66O%xp+g%aSyoS|A|DE1W<~%8svN1na-?!kB-#yFY1Yd+=yJdY z5Ij4y8oPgTWnyR}4m>HKygoCls3_?E+W}ccWi`GKGCYc^QU=IB^8tt&p8Zh(uoCb@ z^o8*NB<33btw5-_o96-!?TYM*Zwif3Dqsy%540ID%b*U}_`>9rMq=&XSr)7~o<(p2 zumdUTbawLGu`qjjcra~!4iMNVkepN$N7Ht11Bkn1th4>?5prC(xR?~eR0nGP>)0cbJcL~Ee zkak9HORF)qbXJBx}rI zfx8Hu02e1$hbQAV0cTs2f4fHi?g3ZF>Q83JfMiZ*=ih+E`Zgyg(8=&`1-9|vo3+D4#xPK6wsfooE9G>U2p6EkOZ*+STxc{D!HGJZ`^UEBp{F|qoEnmgz*wFq0oC!3O zl!=NPF$n`NpUCrI32czX~7Y9pyn76%}oL~mguq!G80_j4Ee2PYm4 z_~%74H`YM)bb+#Y0UiEC=wDHcl5YYwz_OFZfJ~Z5E0=Hc-(N6E^_;FuO^t0p^-c~T z(K$IWJxJ+7&i#`Ez&A#~G0oWX_jN+ROw1d<+5?q=V(n`EfSe)Cg9WlCCV&l5UQOOG z_raJVzGB*eWQ_r&c0frleaM)s{!(}L!ytxt+U*6AGrZ?u$8Mg3C-A}LC5)er=r7NYUfi>9+P&i57z(fd-b}( z-|DO97akE$ZR%`4;1p8Zzbb4<07Z~}^J`mPv9B*zkD(M{_L~erdE}FD2&i%&@QMg! zb?}G?rCsv8Lnc(QEqDTG`Ix&e|2T1cSe2Xlcn1E=J9#kBoLN~PfG9ULh47Mil?V@P zsA@_%p){Lkp8#nL^80!g`24mIynTlMja=caEs@FvmxqeU(@`GCNJ9o6lZ>2 zM7*xT!*@!4VSan+0{aUp{}u}t1m++2(@lavdRprHekT*Xlh{Vi$FPTD*iWklyGMVbf7~jOiHbY3 zW3vY~p{0UM@km+5{&v21ja)Zx)K1cl0KzF6Gv>Y0i*u|-{dcNSQS4<3sOjq!@vy9oJao}-ka-T|?h6=}4QIbb)+)_mpF|wPNeOZMSPvrDy*m9MHQfsx&qk$*R2=$!|U+HC2 ztSx6NN5SDMM=m&x1jq6R;IoH94dk3_C9N%_eZe*Jw;C;F?Z;sq39gpaDql%6!x_%u zT`NsoO(L{CU40EqFg)oa1*Ykc2Z}K5PO=Z(Q7h1;2GAKBGuL}j#dONNn#wa%9#+Up zH+CeiVa;)-_nUJ5U1(=6uLYZpMNM5iY=?gxf*|N3iFB;JT`-OX%zly|c#uFJ@(O3< z>%%yv-6!xpJ=U29;)DR1l{&S%x@tRr&_9dWR?{Zb^rtR$vqQv{nX;W^D3#Q)Lm3vX zQIWo5EH@f|hOr=4`3fE-!{)98Q4ae4yX43&2G4#zaF61giT`OEBHKeE-sRudJQ1fR z+6Vo-2Zn8M$+x=!=D=6bHzMtg#5lNNuF2}VcOO-{CT$ub2*d(}GBt(4JKN00&5R3i zZ4v3P7@!wC$uqfoNXnTB3vgg^G+Nk00c*8l6aaL(9PM=;B;G?>xF{~YG#^t!y0z|{ zek$Bt6{7hF`jtA%o&^~3QdRbSBh%mm`U})8G`Y>A96U6@WL5-?a{ zuSgA8s~BW!E6;u!_f!t7)}+7)ZeRYMnhEFydiXqyJAzNj2tgU!{H0O5S)&7Hn<7Fz zmw(<^WeFjskUdSA>Dov**}jucfEp;JBxB+IpT}j*;)R19H7QJN&Ew^n!W0w)#I4Ry z!E5^3kcK$GqHWe{LC5Pp)NPe%cV3zPo3z2$1IX;ZCU3@hJPFSQ z1on$AJ*78WQWVFsq`LUdHOdn9e-rjT5H?~mOG;#_&Ks<*R?KU?rKZKN(A7vwt5rdc zuTZPK7Gz<^>{yR?bz++|l>+-J)m{7w%6c(-dpatB1Atem)C=Bc4GqJSUcvK_UqI(< zS%wHBQ`AH3CQcMFbs1ZotLUtmi0Q#b0~~n5{Q--fHb_DaWwiVqbcR1ICPd!g?P;+c zuMGBVHd`+gmBzgw_{E|iG9d%%{a5|y1r8xeBLS{boqnE0m82vsoDAAPWr1}9$E~h& z-%&kaR4u{Mr7Fl2XWKsZitnkLm5;T$01kUywsV zx+W3sv7`=SmwkT`M8pY9k0iuin|4FV{^%+>z9QchXAR=S1=yx`NZcgwUOUiVy%b}W zbb1=!48zOB@Kd0_o?v{zF-n+Q%IBAa$kOQm-gp20ZnK~^NKE`2#`DMhJzEgiJT0jQ zw(bMD3HkH_oLKI)*nD%E(JPr{yKPSl(p(8F;5?)v{n?Y1aFe%~mNj3dZV)G0Ab}T? ze}Myi3d+qGQ*5|83BjzzaqRehMYb;X2X)>nwbqX^$x9Lh9TVK#C;xL*6!gRpXE&Q0GQDbs-R|=E`eazZ@%keR084$a+n`ct@NM77RFXvA)!Xztd(`=5Q8?9? z6W77}23E%7j1)G~Z}FR7Cicg`+lX2ff_j-KT;XcZ8D-s_y9-r|6o)BxQ~`Cy^JeV; zsktRK!?3ncWLsA#!1w+8H}O-}kVcSXRx5&$tNKdw5&lU=j>lF(Jpo5MqAwOY;Hx_z z%8FQ=E@j^U|8%4tTN%_pN+_HCEc-Hf>%+?P-zh7TwfZi+^AX`|*-lY-w}nKisic5z zcaNzzTag#-n7|(zRL`0HY5e^>Z^n0q_n#dOPwb7khBp2yc_Tv%5sN#n-Ac zY}=mU2U`U$NtN~C`oJfg;VN@dPj`}&K9&rQgi4`ro#M1t03$c0#u@5$?9SzPy`4%K~^+q z^7N=|iXgo)EMwx~>)OhZKIZdjvi7Wo5|c9Y7k<$-fW#w+33e+V&^<)q3-ae@KcOX_ z0~a#6_m7o%Dxkx>zAkb7tb#DalSl5?tKOe^S!;7WSd=UA5#9Kz$IjzYTH>c-G%gou zV*mrntRpi)jom}ck7BR|GjHq~K=y&9*X_{MS?56;B!G&!j*tMq{-x96H+(8?*}1Ne zGqe6dYZ9LgYAO30AlkOhQxhery>cU`BJL*PWH)_q?Sbg#tQCGP_fR`d)1C z?$Tsinylc<&Qyu0oS^dIC<0F>1N||F^v?zu;C2quaYypN-t74p$w!BZ zlY;qWaH@>m8Cqlekx3!<$=4U+(NG*Fnq(3%W^r!?1``s!JK;&$a506%ubqxev)TLf z5@c@03IChSkduC$nu`gkJV9ov%jE+5ITm8e6)Z08P#4ve2I?E>BhEM z_oFXu6qpwRDFqS5-RxfJC{!ix=f-S8e$)fZ+eO<#>Xth{L6F!l~yLx|RkWTl4_ z-PI3N3FP*{+@P#1Sqq)KJYgE5x{V|WiGrU7N*3}nUl)1(IoGsnwn-4 zH2xcn%)NEo@=d-CQ~5bz=P8@VbhBSMy>q+b`R{$u19#C(bA`gZ z)8VmOP-!yuT>-~#Kr+{#_ zUq|G@;>S)s)k6WS(pKu}v;&@>-+nu~qjZhkeDwnk&nJJFvvM8MIKH5CvsTXJzcPFY z!aU3KVstmu(zw9gxlF&ZDDG{_&9o)#VL#2v8uoO|ERSG|WW z^q9go@hay*Mg%6G5YNGn6bHoo&8pr75B`KD-aGX9S!Ao z!D~lJ|z?XZMpWGUeqfqQ6|fdl+gJhK@!%YA^D^%Q}%ltT4=*)DaGSGEM~ z29qO=`V%rLW+AF92Jy3Bm}azf!KI^!z0a40J3W8f;!M<_lfO=7VWAc=5;?m~-gglz zAPRlcmOHm9v2Nf$Wgnz0*fOO8A=v%W36vIrCxa%A(8788X}NnyQi^IdYut-N=_=Sr zFD~lzSw3ouAaLAlskJ&U>DId-n6WiY-nPiYOsKPOB+`CqZA7RP(Sb%=)Ey4>$ya$G zTLOPs`DyBB45X%FBOWp(&WG=(dtJg5QbjtaoA7g48?TiuVb+qfI?sxP`e0C}Ax|6Z zMSGJ~Y9p7uL}i`g_4~R#T(u1R?(F_po? zmx{T3u5v2B7JnVw%|@`C6UA01xoe%cCdq$nE#ML0mbMtnjxtAAZi}kv{y>NqJQMF7 zQZsjx*$bSC7iTs%`1RZA7mv2wV3B*yAjkEDmhEDe^u1okWm6pVM}Xg*v`v17{qLW{ zoSB>tDa5=SZ)C&cXoshTebx#Tf#l#A31r4(2+B#f11tD<5c$tKSILvdvwMcXE~bBJ zz3h_weW#!N+T8#I^GM=hJf*wKO24mW1e9DSGFv|42ojl5W=zBqrGlYZ$LEJ@%P|@w z69xLhlGTzk9q>zO3>rU>UZ_Ni?62AyYqPPWefggpqG&IEjULwiPVp}!>oB_Z-wyMB z>+`1olO?F2c*Yd(7poiF6kpsw(pP_>i`;zB7!3J#vrRlehiaWf+DXYE{3Tw6w>t%qKb*RG8luO zgN&J9(+Cr&jqEXl;u^=FsO29IZ=vT`7u&W)-(ZQ4!M+q4%wrbyO6Q;5Yg&JNPA}UQ zFQs_(n=_R7ZJq4w66s>_)f)|=<0+~YJSE`F%jLI^Fnn~yNn=T+!XtSp8yrvx?+h{f zH{E^|)-$yd9TnZT#d6I3ax<>kl2-$02gxtB4*KXmud|az_`{wtL<5Oo zVOvDfF$-vFek7P5g)XHmSAtFvc8KEYyJfzW+g-~s3~OcB^~6RQKG`gm21sYFb2PuD z?}Fa0WYlKUPqgn&dh52#P!^*W2R>O@`#$%>UniUG1o=C>!XeLeH+Fxb8A2VuY7&Y< z^4yzunz*SD>iB*UU`A?R9^4B7Msptzk!)OgXe^|aHMpEt9_zVMw}Z=0eJ z2h(kh-+*tfX`^qKE@u)o2vXk6@4+6-Cnab*!td7&8jE^8BFYa@=X(*T<)1zdZ~2s{ zk!iqS1_(4>P$g1vVq<@^uilBs~t47)Ud7gBX9bUAik_da;k}eU4EM2s=Wj zdigDp(Z)4ad?$PN4Oe1qdy6Gat-ELS2j8;OAW6act^83&`@X22HRA47p@daowrPSW z(pofx{Zkm0rO@~9Ff!7%V06#vmQ~{sBEs@Pu9#C=$#BNB(-?mwpP@#?vKY3CDPaF~(YFtgOaWY`@t^3plX<9M zChzbw0F|NLVUm9Y#CmWy6zDEsctf81MfWq0Hcd(fLg0O`alhhD$f7CK3r>bkHJs@_ z3~f5T9aBn|R;H*b11F~~W!pbfFl{6H!jji&%*v*Oe+1YS>#Aq$=~fT5q^o6$S(K|o;g9u zucW^6Y?&P0aZu^SAd8U6vBZy?SX8O!TJ90zq89_P3AVX_3kQqZl28B+33JYylMAZ} zy=P;{@mqiKj_^EGC+cw*c|&_dj+6YaS9gUt)o2c|$&> zf19h+|60Ia$Z=K@`9T4GE?~V%4!J#Br%hT7!05z-}w%6-aqk z&Dwvuekq6QJNJc^8+RiE^lbgJxd^>7eR~~=aW?@$$Nl22K?b!6B6QSGqDA~Pe){S7p z6#V|Ew!}xN3X!-o(hGaB4`_7o)gw~T);8s_(+t{^+K{Nu`t%=~92+IAUDuNWzgwpC zg7YLZ{kY9D6%9|ONh7Rv6DD`gYNICa<#b-EeQS5b+MC?^;2`>CZLcC}%Oh8jZj*l` zRzZHTv(>e89`70M1WPhG@i{SpRmQ=;#6%uao;|~*`~%grmD%Huh<42BRTb`O%E`71 zLp)O+?k8e5oY%OWtESK0?bKNb`3j^Z^od0r=2A>R%DO%61|ujIntm#0Iz=`J3hVSB z^kDef6zE@V2ouSov}WsvMPM$F`R9M)3FB)XSn`H8V(hdL@2)z)UrCuC8U-9x*2u+@ zA<+5dLXSJ|R&{8^u1!ujYjwgMm8Zn2lP!1oPZaCRe@MvNHYp*pzylg8tNkFID;jXg z1m@a^Sa+CXp*vg`d}Ol0aMEl?p)71sF@HH-J91++r+;>*L&qI@G5D#PE{lJR&EO-H zu~HmV?$x>{r)U^D9VDfL0xjpbbB)xh?u}Q}C;2JxIv5R-uL^wWl0xdudlxHs2V)^E zhQ-jC*FQM5s_P6lDla!T*WUv8?2D0YNQpya@N`wimBZ6@`H2H4OuJ!=7&kzHB`5Am z=|1$0(GmV;#;A>vllg@(i}-(;@4d_p0np8jw;{c|a0P)YU7p)4f*`QXhycE0(NFG3 zn*b&BJ)LIj)__o~03BeVL*cJptPFsI&OO?d|uG-)> z=&hm)M={IlWJT}{k}r~*;Gp*_oDioT%a5DwlTVV%FrTY}%2!NOt=5>c6L_hbn3(HI z$?*bmRNT5NWJ-Go55IEM)mG)NdRT!+-4Y*Z4Dq0_7x8WdcpS`&3UtD{KVd{HZXl06 zuUOo^Szk?xbi4I!;O2i_R76YOw)eDt=$lb;u}H!Q{NTH6gyNqB$-9Q#Dulf9jPX)ef-c%jokkkI(iq!Zkx+Q-mSG2zod>n6Y|&GgU?HoH>FVPH+tjYjMU} zFvYXPpFa=m3lM*vSp1wzM@QZWl3i%q>Eo1}9#Hy9{9zDE=IXf0AjK(VwZ-n2RM=5a zbZQkNqtTAR+T`MUi9;qci{3`c!YPJ# z#z@qDjH6a!K*@#Vf%LS#3iQ}Sk3md>ptrZ>PxFlE@ZQx~I%eVdxr=)YEWZ&hQzu|uT(o(fF;D)H|AXU| z7r`JEiKc&C1bn|?`a%C?ytI~LsD3d(>2^ZHVDC*a$Hl3Y=D>p_00{WV|0Af7XZ1>ya~s_Ra){y+P%CS~zvB zS?W?>W+7D8^z3fC-+UiokTf_rtG@lvfEB)Hgx-G;f%v$@6-J?yOWicCj%VMVIK1QW zI?Ir{wd>sfH$YV)L(=NU@vxI1`q}3BnG9%KC_Hre!oNOg6!%A3uPa&s`>PdT*RR$R%S7iI1G ztNm4Jc-bq!j)Nnt%8{hGcMPlWa~#H1VCGR@;dWgv_ft{`oQ+uX&62q)dibukc_jh3 zMMG0=kg#Tc<6q$hGhz)H*dsHn-IVGy$%KFHAzOl*kkNnobR~2eHQvkMk9lpY5n+G) z7V{v13hl{^2MMkN{>n?gIG++TY=O!}s&+qIJwKM0*BTv^&?$(wbo8w_)y4-5*t_1I z!=&9CPV>Spf?7ITNdvn?S4f(Z8%IW$AD7@ub@ur)vpgPa0NlqEGDGLVF)820FARTV z_ReSGlIlqO)b*d3Js6MP+v32SV%eF|o?iGC<;!N&=>i?$p@kDYpja392oeP~4Hs*9 zLb{iVhu3=I7f7)_)cMfw7Nq#2c*h|&am#*P2`h=Vbs*W`m3Y4FMJ?@F)v1UJ>V{R&N}KMelvI+#9nJ`X@Yc=JV4&Fg%R+(Ic8BH z-=Nmj=)tLtT@d(g4}i&Nhd|uftdE_2pL2>Y9UjiI7>9fFqBeFn}WJiv`HN`Alx@n3FKEdC1`pTD8wi z!h9#gf{S0vUJ^M_UPM>KoyorAA2IfQ-m%AM`kxjH7sQA)dYfzM6loyuXJTZNbqa<@p(8 zvRuWm>3PrzH69T8gz9%b;gOM0Xj<@{PH6(_%l+qyz|gx|!;%^&68Lr|A>WW94RfgO z1qQM#trW+{_)2ucyx1Q|EQ+f;w(=ok$1sUsfxX3SN3)#MbCRD}I!@jT(<5END;Rw8 z#TBNj&sY4neei$evp56z>EA5`=W=5nBaci3X0{CpmQRl6L-y#|BM0ddxqb8pgg%Gf zY2Pm8B}Q=)<}Ln+Ny}O$>=Jucl`#ZtblSonSuf~jLgDZCJ$yt;szWlWWXc|7pm99) zJL?Qaf;eSL^~=vZ7SHSP4r^PGl6(H&Z;w`xH85j>M}B`PSkzJ`s2uw5!6fUNCHF1g z4gt<6eYj|5nf8*>^SQ3}0h!_S$zpNiKIvC>+(f-zH2RJH&uxR6B8zSL`2=0147Lq5 znRhEvQj;3LFtKBd0oEq&i%wuX()Af)P%86oWewwFA)R*=dS?L_FLllq2}p%sou8*A z$i4+}AlrXmK8kb`L_n@jJT5w{d>@^}p>nVORQZ_|Qbc{<7(nAk`c)*~chUd!*X{g^ zxXtO+4;iAZg0qyVN24%8bY3b)B05yl2TI2sc$ZIgzQPGy{5c_XV zw|pM#oh|J~S5ogslAEYw$dkAcnUo^T9Bc{zdn12e1+p0&n&H9iI>)SXQ{acWSOpt$ z_n>7yme5c@T@fbJtMf1CLrzK1jA3p(jCqG+Y39UmCeH%G2-Nm~ObfTynf1$&p zBy>9bVuG&S!-Q?ajKv9?h6rWm@2;=MOJ8(7Lm8ug8LTasknmkBXJbHEcaZ~7UmShbV%=EX&M~CGm&SjnF&{9bH`8A=39<1kE@4G07whZSfnX!Q zjtd#V60s6K^Rr=w3d^C$R`R?PTl@p>Q1{J9Y&vocu1Z1_CXJ|@6#8{=B)qZP)#F?RijWYmJD$YID0LT~1rB*z;@s){XiXNtur zLZ8nh32#HDWEK(vq(@`plViV!=A4*J$OI{RvXv2K#m^H}1u1NlLG#eyX9YrC<&xBv zE*%FTS5Sz-cQpm))5oDF@ZSUYeUX1~@vXfVzH4z&(GW98RWlAgesp?yNQ;&f`*Iv! zhp~Q-`AsqyR)_8-|GiBPjzz;wkX#ILf}m<;>A3-h8nRb>YK%24ljy9kYK;}=N0!gf zJ!F6wFG7pSEr}E$E@O(rfPlqswn}{066|f+w*W#lFv0q__m!FS8``4F77Tw_5O69N z%zM2UK(OZApRCaMmN$(BR_Ab3kJ}rxJkDbv>%j9hpR>DRq^6PuZG*pCr9sF27l4-QfHaf>_9G>F6oPWwfN`zf!B5`}B9avau z%ey79(_FldKgRG~i>LcKEu4S(&Y*C?)Xy-7E9kKnue6*#O!&ks5vM)qgGwTihzpj^ zK<;V%0w#fKM08liU9^>3CsOB2ZEvqE5Os?g-9{3(;0#Z%G;?mA_|a`VdUZFzy1unF z)lZ*^c{<&8REs!i*?wr_r+r#s7@AU+Im)}3{P?fbF^Z#G`CmD>%NBp3B~u5n9KLTH zz?BP@Gm52T3%flCLv<>4(Wfvb^Q~yB`MNAjRb?T zFO(|2NQ&o}C;0$R1MP|4wy#Xo`VE$@BS_tWN6U-bL*s#MX$)};T zVhE}%+9do+bHnb9+c zC@$uqvY%YP<#M;hE;KL1DfJ#Aye^)dy589w9u~9h@<^WDljvM40S{lk{AwpCBsrKr z!<|shBESsg+H%7ni0zBGKM&)Z-Xx>F?kjS6(XvS_>RVY@7&?E0t+HhLYzRJ8Yf+rL z5dlAgEo0Bm+fw(d6M%BAh~5;Z|8Yk%(w~Acw!oAfDsjY|y{qW9N#uH>wM~NI>i4j# z{8wdY&GVc_cR8hSzLpxIATdidn7Kjamd}6B#rl%D5d!EVE63O@O2ix>^}U?*tQ@oEyF2vz(8l<3ROt0(pV+4m16-HB z=q_jNdKNcE(8qo?!O#($cld2*kP|DU^Pw};wV1x6KM3kWdH&74m_vb0;n%7t(Nxg% z*%#rJN`7kN)hrvDRy1iAyo;f`;NqZCa+86nI|+Z3z-);(CA&7lohc%h`@ZHxIKoNP zB~X8N$2QRxR>K`tTOFVvKxDE=F7KmO)Xpol6N@VK( zyR3ipGA{*z4pDkkRx-|tZSAxO>T8#mVzKMU%l*cygR~Xhf)58@;S!_4 zm>*?N#PpDmMB7lGr*Pm}rBr57L@4d;TVgc(w(M+-apdl0xV<~>or5YK_}Kyr&n=zC z0&Yq+I15xVcSa%&hbR|IUuv+vG=5QzF0g+~_K^spkIy+MBE2&OYyA3&YK;A5)A7Lf zRW*jGSGEPS$8sn_k^KTMK@j_amYIY&NLZHVwbrFkaUraY33^|RuG7Pw5=}-hvzFK( z?bc7Zu%OJ5I@g?XEA2;jz>oQqoCcm_izL=dQXm%M3F~1SHFvu_7>O1>f4?xUGV*^f zuNbYyh42s9L9wssM68%h&sfNfqO{tfAMvE<1Piw&5XR$s=F9>}VRtUD)d=INKK`hi zWr9TdsL#t(Wd`eE1b&}2jqSBFhPL1;I6 zX#hsNe7yejEUt6JMcUO4mH1rZ3;lmFyB~!hnhQT|#aKhl{nPTgiXgHb=J#|k%qO(3 zpE-)>ReYEkiH?=RPisybB&BX@D)TMsH;eJQnY`=1tx;MVxX54{ew6CVoDO3;Xz3L0 zH_h1ie4&@$a@|=*c_>u?78*`~ox^cQtCsW03Kmf{*rh0pDpn_Ql^6FM72SW%$OP+C z&Soo81VX)BjU^9zjU0pbCLIOO-O6}mMMi#O`15b}qQ{fms(2hZhyb|fSjDiV%aEBK z6gQlq08$du)6YqMbqHwqAtGBfFF#ga+n{E>eU$epbF<6j9M|vQUSrUN6uID|^nHohy&=R&)894u;6HU*44|(vAdO`t8LE807ZDYLFhI z+#oh>9&?57W87$FSdr$1RE?SRa;{7w%6xyA+@Xe>+z?p0u9g`hoiTrO?aXlM-E2m} zd=}rnS_$e=!yU%LmoqR?$A4y$G4fnEa6Y#>3ou@t460^BDWn|>v(_D@kmEeVw!#H|O}=^#OC24E*RT&H5k0s=41Nyio^+| ziOa6ffzDiouFv(oY?u47w<2Tkc<$5BTgW?n)Xn&Kk~trmtgCHd<-Ru4xP|jdD8_wf z6=5=&P}J>6jf_@vM!RL+bkN|UV@F{#I4wRdl60>Kvt4YNc(qg*ZoX>;M0EdasYP)vCOnBL3k)PjJ5 znesp?*M|9kbSiV9(OdS!#ln;|rlj!dlGYa6_>$1|ST+HXai|nbuZ$~Ca+YMpo#2la z6GuUqFt&e;NcSt29X?{ESRm~y~dzct+JiiWIdG>*};VtD%W8zKgix1^&_7aC33cKIWTh)ZwVf%u%l8Gd66m5FH?7nuG}p-M z2G#n71B@Zb@`)*CeznFa?s;Bj804_OTHSZEDwgFEQ{n6{mCZ#*k9MK4va_NU*-m2* zfAcO!Ong+0UjH^2lf-rcM!*+k5Ug-2Hf(>87uUSFAEso2d%kCsop1F6afrjE_&19ucd0jCnb6m$dTbYh&^zqlnmXw($A`X9! zmE0HbPAHlam}R812LZV_r=+s`YZz2G8wFm1vNl?pWq}{@eX`CUrPp<8XOs;B3*vu7 z63xe(BiU#z(m5=?I&FS4djH5!Q4u05wx-vBcC#JCZFwi17@S=y&^4C_-eL2&jWBUH zgv|eRHw0eEn4e&-HR#~cs9>a-VyvTFuWNz0dRL)o2CiBNj!ZGc`LxumnO#|FG>8w4 zK=?Z59_jfIeS{M6>iHoGQ%`StK+S)I`sEOHPEMv7I&*S=rF~$_JdsJ#WIboKlE&`){xUenk>_>stmqi8vPXg%;?ez8zz37nDSc#Su+qm|Adn(Z?B z8_6RRFMFjrL0z$Xtsla^kXI$Inh&4BzEl_oJurzWE=vrrMq)%pL-0kgTu1lgICxHc zVZc0ydT%gMUUKEF2%xXP>F_{BoZ{>llmhC?}-?*Cy1C z>6E{>Fg3u0&>r6TQgf$!GvQIGge*(17_@pu&n^-ZXmCrQClnMqC}X~IY1up-bU`0* zl~#-J>xt{9;F+6=tON1J_b}YcpqtLP-!4HD+y}%)P|tgWk$}sWp7ejX4w|Bw`93)QP7db>mKl^|pHw}wR1$*r?t(%>!gIeOJZv(t zo-D8Xg8jFdjnR%tg^YjYTo#_qYR~WT!V$rGt8hjqhd6|DT?Wo8?*nF2?4ChhM;1Xn z*xXvp-eC)Zj}5=sxr{_ulI%qt(zh{3#1krl#ToGkQ?2c?aMF^UWX@%CQ+)I$9#W_wp`@n*X+~dzEQz~ zKk{7+SYdgJnhMDEaVn=osVfco)uK3g#mQd~lOug6tQl#){8nNP^eN}nE;!rB@UPE} zOZCT0;3oL$a5%1iuvg?K3it?bXPpt>5-l@<1BLH(frA_*k!_CjL$)p=d*D#T&-{__ z>{zJt&R|y0E}nm7CP`5^+pI3Jy5rQhO@aeF1KbA(sg>8O+4JL4D?wf6hcE3DLUTuh zy1`1S1Xfp`)>DlMxOF7bi*k8%R|2xZxNIbu34e zr6}WhzZq_bN?c#p=A|_Wm!}Othge%~T%(eXWw_qLTa43r- z1LMUHeysk;Ope`0X2CS`wN=wuwt+wxl}89VD4H);Z`-T&Ft>{sMy*6&?K~(Z)@X1_ zo1h<~Kg@E-mg;oFT`tx(9JYcZG z<0(r%Pw!Z3E0l)KWTOL7Kk>V!w7J*Hv9=#)K1LXXXuW~}FF5%Po*CP`%$0D+w#_tZ z2F%Y7FyHczyZX>ce^gTn-=G`!WOdFS+}a{i3Ql*w!XBa}m(UTjve)1UD2)W;`;?qg z*6@GCk|@qHVtw9Z>XK@rs!)dwV)(woJUzGe_67 z6qY>)w8-qkw&}=T2$8?p+xLPGygd?FGTnb!<=-GaoMKM~adM&eE2-~`RtWA@%3%~A zupL)-)Xz+MWd=z9+GkKi{@z88?88p&`TO!Rs|#@Oos52({sFv(61FwNeqXwiFEwDwc+}FYN&kP6d(fXn>=iQ#?QTJ9#kaTLw^o!4Yx!pVW=GxY z2MktHL$G_6yFY_<*vGUyT7L|<7Y8_epxv~XeG_yy$eBiP+qSD3hC}q*>hLv`V~k5= zP=qXkSr@6Rv$dsg(k$aAE<{2BE?Jb!y<&wplc5$V8mf79d240f!9VBWS3!TV`TbsK z3|qD^_bjhD_ zEbu`u$@bFAqkKi)#{Q?)&~g#lgb8Dy4KNU+%o@HaLv+K0!k2mLX9?=^I&KE$5-%}6 zODW^becOjPT+rgg)&hglXr6yxCt zU`XLAG?#9A29*a{^T!s*PCGxyraKjeUpXg~QN;M8uFsBxmuATB&|-hGOe_szcBphV zZo2W=i<0hse6YLqQaH%2N-QIWgA-YRBaMY8>0y_Q6~UxJ`Q{Cshws`I7pWT zcVZ_^HXeR0O5*qPi3g#4D-XIP?{(EGwC05KbFXLn3tw@&iuP65Q%Rg04<2@)%x=eK z?WQ_Xt|{)(W7S@b&~7h3<{!5`w`&G@_2So}NO9NW^Q3>wldy;55--wd4?rWg zz(2OrzCd&42q;k%Bbk*m*>WGMUEU;C^AwcapAK^TvJ%fL?c+puPqxO$U;d2`O}c#! z`MBxHpUE(Z^`p84CF_tZZJpxR7gF*ExU4l%gI1=|PZMAEqaxbTg%I?!eS#vUO`|-0QU(5dtoYh+ltW5@%Y5@yjx*L>`LBA741n zz#T8cmOiM8u3_5gaK}QgKbNyX#(BrEj^IKxsgYJEAW0-H-3*2Hy}a{U&Q^0T4KzS4 zx!?7oh?pX-@r1r4i)|@qD>^^9n;;L_cAdTu`igsLa0t=Ro=TxRU{~qWsmtb3mJE2@ zWx3*$q*s4B?!xa!A68-N>7Q)o^c~`jSff^m;voS9xbjBI>~O#z=){%y%`YYPT?SpN z;D?kk1tTMC;hQd)sXj|_kjuP_Ik(UcG8n^5xpCv{EFc->a=@Qym}VukZ~q862XpBTylu?U0BF( z#7!e%zF$-!1!}DWK1^93_$I3*{kwkpTPj2-=0U9AsdE&|KvrzG(^(-g^(YoK8rN{~ z(lUQClJWI7qk5NK*_&*mF5V*ibI;IYY)N+dD<6ByINc1Y34@2h_^{PDB5gdyXmG~GNPs@`fsBN^bUO8j*y zdL3~fT{X#^=rshZAlYDP@Pe*64)+?uKy73r@lC#4B3Y%S+=dfpeb>9_|FD!pq zf;Ee8?_)U0I#}1u-Cr|0%#g3E*a6GZI(xk5!_#YSU9u1b+dny9?Wu*V2#p%pUREu# zE6~$7_97bE^ZH4DHv>YOsI}Q=zVCI4TG$P6OpcGJL4YK6>eC@s_M2l7X41d&uWMsBh7Zm0j=8t6O%oyW~6+cRgegs$^)G6YD39GYp zioE{VERLGuL*N%qJ`zkCMRT~ML9QPc2w}NjqCqXTS18$SQMOE}riT;#1A3 z)G}JLehs+~F}(#@F_qjx>*ar^*f*#bTX%>Gpu!f6@0(M6=}2MVxouSV=67~`BNcDu z&Z!M)`tIkIm^n`0yR(YG%HvuP9L0+4jUEigCg7Q6D048TO2B;Y3t4W_!%e zK)7JFsl^)bz70Y6HRop6u3P;z&iNduLd)?nhg7)F$xehUKJTnxK%!`{-LhNJ zg#CHNe4eS4?9S`LvU3-dL>HWa!43DVBek0~dC!N?LP7Y^vd?BS=8G{iixyTB2YiBLZNiS1N*S*^(MY5@P;m)e z`b%a@U4m7tPINxD%v&;6Q8b&11U!^Fc*$g@9cEFwPnV|fZ|Y?V4>d3Purag-8MM-R z>Qbo}_gYjPR{%}WK??Md28jlU%Ui@1+6PqYVy%DGk;2wy<#vB**+8({akzryur1!9 z&djVxW-|y@IOuLC*xb3dN_O49KYfVKSE3$7ut!bdO?P6UH&HB#Ek2=q;oi@vU2KnBi|1PAx zXrdw^hs}ujH62$VGEBs z1}v(#+&KQI&rwk1CD!*vx@2*;q@91KGcDGkh6fa6?g#FXjP!mkNmhV`xpqV;5Fz1r z$WcBbY~=QG7xVj2eSa>hpIp#9GVQ%z7W4~zdJAc6!z`TyJEccqWou>%hFq&o6}SEe zod36L09)2caRAqD(IW}JySN}(U(@);vMh(wm64*;K6Q{TSJ%YgZGFf;994gQ&t2#@ zNHC)Nz6nL^5zB%Hb8^b`H+cMr!C%yhgp#$!+NVZ=zA(h*am|*OKaDrRo)PzpDCJ&1 z7Sv?4{GwQ`GQF9(E^cGJ8e1M1zb4UC5zR=%duU?gLmD{;(+((7!fr^%| zDJ+M_eXeoI$nBxQ;2-T8Xl}3L%M$f8&F|L3&U#DeZc1kpSb9ocK$H25c%V$IH5D%& z=YVokKZyX$Iv8obNHAIox~=fpl?oYOWc@M;#?*U8w{VYzSy-#HI8J|PLUVq#W12cJ zx5p-vOnfDXk!P2BR9<-mX%2xyWb}pQ$@r| z%a~oO?gw_UsgJ_YK*3GeaTKYrgx5OUYY2h-mC+iN@}e;l*G*wPe0d6#M)l$|5#=MG zD64ERhY6gqx>)Pw}9=*bZV+rWO>; z#2rA=d-0CyS;U+OGU2n5E|Hg&X;9Fqp&g4%UJeWHe{IEF4@U*-;&Hvp>%Z8Byz=8D zrTS+movQc@4_>F)=6Na4K|;+`zY+h?B5hO!C-HcJU7-oc4QzkO-P-T=>s>c}KEzbH z=;Y+LiF1b|Pjd<#DFv!T^t-r8x{30xSa&m>A1cmnQ~v-$L*0nP0f;v7mY^o-fkYGV ztW!R?_*w2q+=wa%wDmimtzka{Du7X9M8iDf{e_B_mZn2~4bAxUkxr^ULHB*`Ej)(5 zcYL+^&v17w{aAmJvV*40XD&bKfb^96Hqm6az9sqi5u|3$1j5c*cTVFi@gg;+(GsdZ zIlYwf9nA&*h;uN}yU+t-ZfhW6Z=evqN1n=Cbt$^&YnWYZoD|s)^#taCF04kNHQ7b- zk))hx3bQG|sRmGcAZln1ceTLE+D{WuPFGKyz< zea`&VH)a+fIMA|{z9nIou{&HM6eztK0*#nEDIi@(#1C&!mGvv7`6wKx}7Q13(1bg@)}uN2Dy=?ofn*wJk}- zP>%}Q|EGT?C!tpN#B-@QZ&nQ^i0mO*6Pa%oSWB{;>S9?pB7D)~VY=AfD;m(V4O%NL z#HOc?S-@kX1|M{9oReYl3Zbf)j@Hu`)M@lY+|?l$2_B__1o<`VMNhy9odc=1xXcM+ zkaBfHPIif+7o|70nn_e;yFZ}^rl`O_WVF-a1ki& zYjASSt>Nr@$Mf@ONEp2%(#|%{RbeeP7MS(sCCh@x%yVKbsPncEFB#OrY*Dv!6&{gg zpHnVO{5)QlnhbChcT?bYw>F27T?KGoICdJ!1Cg7Ng%Qwsu7&XNQV4@$k zBs7_jB z$RG{xldxCE_G*Q}Y~ij>?=&?-s+6lK@|u+j@F-<}j$po5zWngKlP=grikXh!cgR=I zh8U}{ac119QC)745t5*HD#^_Gk!NF?bhQchtIVxzg6j!#1a;7zdjaz&2l{N@d>lll zCT2Kq&nYP|eRTgu%MslF5D~FwyqFF(RQ`X0a=B$kbHuR4)wK!G9Ihkf#q-z8Y*3ToEJ7A#HGeSxY!8 zWNtPVnt`C4GD)%v3{Ku2e+b||7LCPZsrUihE^{;}0h7u8?RIQWq3sINm)>Wub!}>xr(=rFRFkhAON834?4Sk^L za-^^Wa5Ru*($3Hgkdt5J4vE-LXaKM0X9tI(2-S1l)Jp)cI2F<(q=D>+=lvhjjNP=~ zNUT_PFtq4f5=()iCYM~&;Eh3lCZ5-d zWmg#SJ>qq5=^HR;A~(3|3S8&rg0`qx*izCe-N>}6>GC4r*`wC$Oy{nK@Y;&%#Dpf8 zEDYOu+a*th(dMh`x@8PR9mADb4oR7#bD*jt2&!rue46$twj&Guf#R%JsO3wBa`fDa zyq}JTUAJ;tHpGRUQ*b6wkhYUd?0gg3wrz7_TN5V}^NVep6Wg|J+qNeDvw!VwZPi}v z-RZ-tQ{7$F{XP_f@GOZuc0|j-OUXF{Z-c?xonc=9d3b;G-(^v=WGRZd6X^-zoVmH~ z*7jct6AYlWl(*4<{_!e1#R3lXkEp49$!O1?;6`J;L)Z%AoC=fl@oEjdxNLIV>j@A! z{lLw#;S;gk#$jD;(#e))RB>THhW@_RzZIVt()+T-Jg8~)NdV6~F@a{_jWoGsVt|sM zy$cl@n!jzT?n$N|f}?Y*UaurCO=WLwg-4Y+^V-nS(M#Bzdsxm7Aefhv13aYL)iW%- zddZ5<>zi=cg}S>-4F15cO<@P;Vt?;vEfHLU0vQ-L&}XbF{IKpK=E;3O(8^*DMz#dL zWjtl#P|KBo;oMORmyI>xWwV4ZI_D*|K)B`RPczXW}g-;N_oZ>fQ7(2Y)`RL z(35K^*)jRJwB!l|2WzYL_YFA4_Oe=M)12i-zWm53_*>hc`pDS_ek`#k-N0q!^(al)?5i2|&OZM{dAr>6Q` zhV~|4+5I(Id)U*?fBe%d#I|Edn*U>U9-%j8IkzVV>$C^RNa%2wx_PohW*9=^UIVKW zWb&RKrxVm2GU7A>amXZDO(T#7b4) zi8UIUNLSTllRVtU!Q(%x8n)P&Q*HZbMiKfnyPFdvIscUc9}?n{Fk+)Xxl{3b?E`B? z%$TWtua^F}md*nHMvSQlv!gE>hR%eVndv++0-S`4`Q7tdTTvoc#0+k|hx zkyb6N>F3-o!D2h`w@dMZPQ2_z%2BV07h{NcwUH)>nD)0}XUF(stw-q{W(k5LL*aN} zqt99JWzxML_MF=GrLP9U=AD1$aIgVZwwd^o__W%GwcmX;grI=oJRO)$lH;~S@hR&6@4+JC@|H+^^dC+GGL*%=ubK6w2yWjXGUpkw#KCj%g%5mn#8k z9fbOrL(q|z?Vn;5#}xft%Q`i#taBc(xg@P#>|WOt)_=4epZ4g7cn|X?*qX#EX7WRT zQcYxO$NI8n>=T)ulD@bJ&or}2fjs_FhRNx9OhDA|u6_IPrKw}>xO$s<=yP!%#@mz| zlqg0U$J{nVF1mE~G9w*jgd!%I|x%C6+Em|+r(mzd?c7Z4hV^}Qi3Fvzo7UoA{SWr z#6-;u2#ukUf}K~TuyB>N@r*>-$hH4Z`2id^xH4W%NDRvB`duckI|*X>>6{OamPoVN z-Au@rqabXPuKwrX?0rn!lQK->-%84^nnHf<;maRQT@7NR$mma+jBA(}3%Mac(osZ4 zEX1RZ7~XtvE@bt6`2cx#=bs?uIZ4we&uOklJgt-|K4%NU634ex84_efCdQg$t zjYO@9LZI3oD-ZMSCwJDk^m9yqiiQZ=L?;zZe=2Z=s zreQ3c1F}2^<%v$w-Zk*=^+W+Y8(tLyEm{|A%*5%riiDnB#DWUF@^-h+0|`-jC=fJ= z`#R%&BzPZMS_Z{K6ZdC?2s|C~lqIgG?*jy7Wl=&5PB#mlN_D1@bMP}zwnW$gg*PBrCs*CQ0y3D{B0SfiVrhIW3S^cph8;)BLy2n|4_`#?>NfuMESpm) zA)y2OywJ4oEFSei+yL@4Px3z1DNizn09->bOO%e%tZ;FfEj-3#52q ztc=Ft+w$h-cSD<>@Cj@4yRsFh!L-_5Pka@ATS*ZT<$6qUS<(l>+GPIl+^O5%!?#UaJ!a^4Tcu#rqDPwEdkMPCOT?&^mm5+$k#cXFT}h6rYCoPS zci3DuN@G`+-=_TsPUlS9?UKJiF9|0&=uT3s+0(KiN|_NTy-Hm|{v6Kb-{jW>hzBSW zVW9pr?S|bqd%pYexKhzvg8n1AP|F?%Yf9p`qzJ6XZ)n#W#~iXb9Im@zAt-F*aT$ah zhzU2hW};1jJ8S0jtz9{_`Dk*?T?4l*Q>pHuuJ6j3UZ!Aq!(te0_6-yv?(#7D1xF?C z^mngf7ye-2zk&f~YuPfIsE=H0LVGN(v;~YJoad9C06+Pp zsQ9z#cmK}7Lr(HZAFJf%UVzenc=zX?tT8s@6j~o#|jvHNNJ#)?8gaz3hLT8qB?Hzj2 zdF;e!KRBD8Tm;35Ffjrkn1j30_@+l%(X;%prUysTYgx6Mi)^cEIRt1xQJ9d2grpZD zb7tDrOY-bsLs8a^d5aEHsg)VBo<_ zLewN+H2GlB<5t|1lIVWB73Sk~h|1d;%t|QneIPxYMPP`Z?`9YEE2%+Ab43mZ)Gp)z?Jm0p~GxcZ^m}xQVI_ zzKIJi^BOhtzwfp{9+Dk9JH+}R(bS2Ig@ctpD68fOtG&9nBPFobGKM$5dm?k6P__k+ z%etX;cc~Boe@;7#uS&3ey(^C3=yU1KzU$r84w>=wZZ5_)P>r>+P2xl&adM~dr$l2@D~53 z5X{tw%X)M#-GQXJ9kl|p{EWl$W2WhIVi{WIIi(^f45kEa{FlD@{@Q?CnaV=oLXNCd z4@!B{PM$Wb3Yt9fK8q{;6|uB~P*d&_WbIOCWmlCse#p2>U_;$t69X#=$>5KsBFuL! zNgcO(+%)w?scPSS^ylw@8~lMcFaqt6xR3wT@p$~A;=1o35rQKpOn0KfQOQidHYM-l z?Qm@vZ?(iR5G5-QDP{Wx03HgY^viY&8+YHKO>fdnIiH8#y9q5Qj+o{ErsCnu1OXgx zsmWn)CerB-Wy-8}=T+F5VA@g@zeGx4XI3qm@ET2z93p>X<1;Z+WqRH=FDQy=woh+| zPIBw%>_3!06C=R4X|{x5<ZSDYe9{6AK2pob9m{dSDZI91+?iSRIw|1Ql zMT3xjODO*BGJAs4gXw~0@GWg2y1O;(lSc{bfc|n#L)O^0usrhIn9&QAqkjN^{|bi@ z-(Nl=VeQatJv}(7Aq!EnX#q7`12G50Zu8hZi#X;`svING+(lDDCz%4JJys*r($$7d z0llL<>~}Y@M=W5v@?L%mEA1ailGukC{*^njqdDXmZt8;18?|!l?Jv9%J3J$O)}D&( z_fiSHWgkIsnU9(gNPyQ5>-6o*_17Sc=Ltp*TXYXB{M5fUxEOI>ruX4aT~I&LUxmu6LAKD znQaEaE#7u?8A~zqn7yMpun}N?rEP4Q1M7yfnQpPvk{f6(BcXmZ+?Jbh#9v&^LoR9D zQXKAgM*{pAzLNWOV+qDlZ}7*;)8sOHK8(QV@^lcQ~lF7 zVQgm<`-^%Un`5rwlpGPolc^(t{eXgiRoMJmM;gfIJt?E@AUz-B_PWJ84qyRZXA?Tz zAm}2F=1}5ite#^ShZFH}>l+sj&NBY?Gh8Z->vr*io5zLxXjeaQA-ED7Gf+cN9qg=` zUGRGvPU3%`EFf~+bYor^wd!fzh-bP4+^*G{LHPPEFp%QQm>NE}4XBK=az8an$VxsX z6aYohbQN7_>IY-M%1hC>25--te8c|6__0)w{kDoLD}9#bgFO<=BCJ z2Di`<)?_KNL_sdojyx0(8gJuwgl*)rUC8GZKte~_EJ{=vx;8b^vZA8qQE*sCy;x0$zw=XrRK_XM;+sV9uk=%`# zgHLD=tmQkPIupnnBpSH)N@hp(X#lP=^U2>?-ev*xf-$01Xpy7 z>!FGL0A){F9)D8mMZ4F5>%_c|%v?>1tzD^1^9JgQ!)Yt#*NvLsK1>%%1O|k5?h`$| z3Z4k+(D~-J&(H{8mRlB;GePP%4>dk8_IlQ+#+X3-5cI~-g%fo71vP2wkpHbzxH-MF zUk14d-%1go0xUCYKL8~6r%4>jSMRK!#s4{qx@^;)JGnP(A#MS{ zQ#%RtHbsO=^=KfHll!X_Y`HBgAp>NJ%J%A*(vmy$%k36mq?#acP%F0rZ-IATT7u2c z5pHI)aO`_4|1e9m(xhgR=;I1lwVFG>tJ#~cEpMh*6F0hKbf2M6zy=E2E~op(*+4JQ z(ucORcxg7!m1goJnb8SQ5xo(;qHeV41PgiNT8zPtBIB)$nK+<{lkmc8!{UO{2g#Ba%A3D;=8QgHB2ZMmyP7y z!euZmW}E2(?7_ifKS$Rwr%*dt}@tw$iPuTF>lMx10h$^Z@FD0b{3Qic=)u4*k% zdGP|?)SLQ7w=-OkbGA@U%gj0zyAL#kb9cuag81psdP|dwlfOYxyUxlY-ZGIu(FQ5Q7=xo*slcPy!xql z_>%>3%v&A9D4WVbsD(fpZ7VXn)vf=HC0JS|Z2bZ1DDkQ&%uGM&1a4C?LR6SAIP;>v zWuH$mu783Q(|R*-R|-vfT6DjzwvUwZshq=DX-K-%cMLv*;AIroyc0rC^btmZuY=MW z&``VV*AmiFLPA&in?UEjZLegx&wZf+_?>5M({hCM>q#7A3A$SnXvbP}ovg<$2Ea>g z;!uIdNLg@#`zHHU36O}Y;K~;tkxM2z*_JsVyDpCiSpZd7kgwj-)9$=mDgEWUKISMVZ)d_3E>RzUpTP>^(ANwkxC@q77xAs`N6M3k zYh6mD$a=B!#<9^qQO7L*{nZ4Y2-@cui|znwO2*Cx8thvCdK1A?4}1hGpp;_P|HYx1 zuY(pZ-pwaVDA(HJV-b!)Lv$U?*k0+!&a0*MF#K*g@m_>+SSfEoko&R$RTh+nNaeUIwD>EYfQi8nTCx(ULL!Yr*wl^$^7>m^=Juj zGb80h%$GNVL-3P6E2=_C!E{PZWByOOXOkbnU&&v5Ho78I>>y%$8 zv;_k-(oS}es|U!w95{_Segi8alv5C}Fz4AYTJag8vhsYVU;*}@a#E4LF-@qjC4e{w zXvn-wk1c#WM%dn}*Zs#9O9%z7H%)o%JHPn}s9OlYFV|Ci0@h#RdW7w5g09ifz>e@de)#_w zscDh1=P}X++YXz4ew zRMV+9(XQ8JX*3zpG#P6n?W2&%F{djY-PCsbwBry$Tgnest)-}j5kfn0VCw@BVsT*& zsr6t^sq?@xMabAP(W?Ir*Ow$d*e55|xv3?g!%nr$K~5LVflfCuOFsq%)(2Qk(wz4e zB`3jsFRQx&my(Kolpz`lY|!9W7OP-Er|sQ#bu*7>N zWMUwO#WXO$5?Hj%M8_nRjfcSdU$+LsKO(h!+n}6*QB@wQQY3<#0d0e`Lh@QaTnuHER9NH&?E%3=BdMOFzMwYW;Q2xx# zM}a{E!uv52vobnwB>-B8iDm#Jd4)nNg5!iBf^;OLupwg_gm63Rnm-)ldpJX!;sw8p zP0V*v9Z}ZCWXe4N9S@L3^h9&~xzkFjLPWE?nS6pHe)=O6iVzWsaH=U>U8DCWJ1(Sa3ULR?Ohr!fn)mgs_ro5lra6r}DnQ4tbycu)^yJArO!J z*hQL!5SndwtsagpN)U6t)*Fmw8~~fUAzSEWu}b$K{Mv&k%2RMQ0a|~Z7UO%q?d=mh zGOl@eL^;m2UYoGvb(LW(zJ+J;@|LYwe3<$v1&qH6IWPS{aXNfBmqP%XSusZHsr0A= zrmACNGR_+_oWD46;Vi9~7|%O+^JC-6cdJ};v1<5P&`%q*^ujyQ1jZiVPZy2jh$hg_ z|3)f*ZrPbp+4}rM+|%LOf5m5Ztiz3&W*?K%P;nwK>r$Rr^ZCUy0s8#5^+Sg3nr3H_ zzCf_GEAV}js#uiZq*EDjDT3eLyEzjX_yK}@Hd0BUFdCOES&rLVxZs5m{kygg1+D9+ z1nYJT2_6?F1hacHyaD!t3zr7)4N+a_{7vmGOWd&F)G>8Pp6l>cr)=|2Utf!n^3J${ zotDsV#dSfj4e<|GBEL8qQpR_AQc6KaqwW&d1H!}BODhJKF)Wfd7hUx24XRlbZ*3rNIcn5pu+=(_Kv0KNL3Lzue5AesABbQ_KiZ^q;iJFMr-btCfoo94UrqyoU{%y84=C z@3@~18F=6Gt`kq6(>9LSV6?K+0@?qPl2zO|=hTyHYTZ{#Cp&h3 zzFXpMWV5}p<8nI|y(#bK)vVueaS2T3{-VzK)^&1R-54siu&`)4j&HQ_)UL<=NdCat zr24zwRL4fM@Ns1(wAAIo0{o5E6!EA%V=!;o97SR|N&WP@dDP5FQP0a)LW{SYb|_-A zFVx0%#O{wsm1F-AZu`{2+F|f3ImZGwhir<#z4yk<_CfGgsCWjdB>~4$McOnoL$#yL z{Zrb?^HwiOsla*Xu)6t|K-Y^PY2`(Ke;||&&9&I}ZbVsE6+G_(9MH>gd6?r{O$uTR zb4qzq^TLYMOW-fFWpXDhL7WoblaEZAwUU&Vr178!SKB ztzJ}0TVBF5X=ciX42Zr$jW9(dMlIeX7=0lXw_k`VRU*Ewg|N0ptcN{Q(}PqF4M!bt zyW>q8#e%F}1C_}nrd=V1Nac6(JBssL8f%RI5C*Btc)Lqtu84V!M&^fPc+nsvZO}0K zAgH;CyC6B?5n!L71#%l~K;#FyubLn(7zSpouN#a!1?nw$4Dg;3^Ze!kX%B@s2B2r$ zHq#^QW*M#b>y(qZI?J&vPCzl=!ooggR>s+YY+y6jt^`l}C;o!vZ4#I6um))-ZNx{q zTQx!wYFZORejppCeb7$kmY6#3cIor2gm3E9NlhMz`0+d{YSREfHGviWBw|S;f+~X) zx1oCIvp zskB5GrkR*#ESS>65X^uFZ32eZqJSX_rQ1MhH5hIlG0+!f>}bRTYXRB>#$40o0;3IO zvF?G%%7~D}L(hC|;HARI(=!`^OTEMIk#nv?sII4;R%R^a6{4)XYb3sdOnCFh@>QVm zud40DV04;7vZ?ShQ_l)dt)v^R`9^xuHU1>4+GFi8len=4+vBW0O20QoJu96>op1Lo zT$hVL1uXw9anzviDK`gtOaY>fH2akKWHUdU(@`-FhHrq-u$iIqOub_idT<|~Y%TYP zv!*A<%$O~Csr{nSVm5R<7ttoC$w>gERpi)>0OT!E2HL<%n5YSo)>tMI_SAHBrCM&o zuyjjWQD2mDsg##yD|SC(FXYjS6K9jQsQ%yHKj6y_wuc^mH;>QVNzQb%zI-**&t4?N z;(bSCOeFf_r0DpNhrV|~Pi?#rgPUQi6_Ow&b3PX!{{Y4@dmz3iV~e{|41M(+8Dgk6 z6tQn!aYU~hYllE9?lRhSt$M`kE88dQim_b+pa$Ps2f^%L+bfhy4qjcz1qzRm!iw@! zCa}|h;sTAww7VAcZ6w&56HkHSB6BfvVHKhsv-hhwT~{`R#bvZ=B{XW4Eu&`iw^BV- zhOm3T?TzYJOP9vLT4bWR_0-X0u}9xg0fTifx_W(^S;Vw*lS$5s`BRX$=~t=wJOK%@ z1W9U;O8$GT0mfYQcGmmnKTZ<5W}-`E3!nl4BJy07;MzIjoLei>9Iv(5E2x1tIH!Q| zu=RlE@*~SE71{2#tToD05~c6@}rL3b$7opp!2!>nFH*SNgJEa{_8tYDgo}|9T zl>EZ}o&6mZe`1Z^JFsIe4TzPAbtT;_0_mWP!D@91Sq7JXb_}{Qq}vi(XaJ-73P$Zc z5L*#OBI)TZB?;au|8=5LvFvU|C^?E^xik5D=8`w=Bv;9nIcq|fTc>-Z8>CBRf^ti; z3{VY>A`^lm$R#qf(4&G&Ce%grwR^9!7bjDSv)tZE%UeThn|ju$XB(` z$|AJ=0d(TzRp0r3iuj+ZvRdyu&t-)YUmGMqmL#MyIxe=u{0ypxMD3KqM4^ETU^ zm~TnYWx%=Lc!s=L!IAuKY3I1SXP3No61y5#*#l#vQiPkG^k^}RHOh}+Bx^EH zLIs=DN6nV-=uTC| zi}LdP!_%QA?>pyBbc<_%b?e)i+Qub@Y#WE}Ew#C1tAp4s$db@QLkW!1SdU(3NA}*X z-Z9n@)OB!+0|y872Zso6vX~uLj)SU{TNgGRogo|j^hP2Z%coqu?57{xok(LTAPQj~ zWQxF=1|>HSZbDgun)+KP zg?8i=lI(UU-u(+w2h>y|*TzrAhlU{d8U zan9s88jrm*d>ZvSHy{sQZFj{-pPs{#5g3Svt8jypnba`Cv8u^-gIHqK9+t66dX=4X z>K&;4%=^7&TlC_%HAN%LL&AzGE+ zJaj-bY&E%bHHYg>i&YQybE0b-jK$JuOIXvckIt$zp|)kZhIiE~i&I2@&S_1GRVRxn z#3%o4=}rk{kaXbR^%-L*G6i!1SPzsz}H~Cq{GnuMJhp?oWU@UYX)bM z0~4mgf9V|_T4^^ZdPhz)zyG0=gZq0cX&!NSWlgM{cJzRqbQD3AN0GTE)Ae<{up@?> zj|;r)>H)sdL=WPpE&~X471fh#l^OKL^7+|b9^c!H$2z0DbuY)e@L18J>O^L*-dqw# zRh`xkybDx2G|ulV?ObRbH?j-|NDTLWNZz@%3dZs4PHE&C2#JMkoG&k*lDAm7mkxVbW6cs!x#(ofx$#SVBx5M)qts_x z7lHZQVo;83i@hq8-pI>FcbjaFdE4!;t~@ObRZx=Jd~;F^?u*A{uqK&H-W-s{A1`!m zwM4HD8)QFItMXfpit|D<3mhyJ#KyeB=<2C%3kKBE0NLPR@msiT9Qelk)1#Yz>n5QZ zsl1;#NmTbH9Yo1DGWG*yy}dmJhe|1Gpn#+b9HS$fa^?ZD0mX=7h#`p8Z2fQ_Q{r(! z#H6-sPxZ**^WqsnBguPt$W+QWwB`V~O|(sS6rod2Q3-UmAmP7CTx!rUx)u-`TnX>5Oz= zNl@p?$N|0GxLKZePRk(DILHa=5$b3$W8p|A^w02xh#B-Wj8omZ2<|OvW6#I>kn6`A z|7Y&F%d>C7pXJ5Y(c&gZ-6`n_aF%Cw?X0P$MEUJ!i=Z=t2V@&fz5~N}RirKz*wQLH zh0;6#`wlCFCs()9)mxn-=#t#SSwR<&rK*mS2eYjOn~N7$ecTaYf1IvGs?U~RsQcUA z@m|@%m%X)&$dy0dDui`u@+M(~(=3L5W8HhL6|o zdMW2^W6c`w`?N)h*d!nHb{~sRYza0oC zA!7=x00IG!ol6&%LD|B?l#uxwo6De1sLe#kLdf(zsbpv8^gYQ=$VjLJ%OGWIX7_Dz z{HG;KsLjhR#L6tj#m2@aBqqek&c?>h!pY9a#`dktE5^buF3iMF_xqqh~WyDkdzty8XP*K{T6M;+-_lNTff08K!Ap-Pwy z;!=;+2*-?E%U}}>?5{wx-+qzWW;^T~hil}N+GVKwzys%BevdVvWi=>Tc?eH}?S)g8 zUe@2hfXW(#CJ`v7KExH6_U{hs5PBJJyS+r)6}Y!tsMj6R2Q6$bf@eJ60DEXe5dmDk z0{cj)OjBoE`5rRN8kM&%o8leTOg|w^e?v|<5_`m{eo7tR2u8>hzvZ-Y&GZ*+3VpwF zM#P_tKp164gykLX4Uph?V}CC45-yn-Wn5>dtNx0@!D$D&BX+1~1AO#}Y##+PhA;bw zy)Lo(!{b_yo<`0yB3^&8@ovxLijA+@xYZQ(TPEr_&CmHi)}yi}d0l{YK%!rjsCqulY*JY=un~UHJF!4*e9dPtxlc6J5^4{i>z*n6 zeKaP4N*49#*5>=J`;x3BQ7imu@uoAQ@oHgk3jcdI%(9=0?!1J~WhpkitYqO2yn5gt zVV}Huh$!LEe0rE<;Q&58qXWsXDF)30%KRs{El@%UI#9WA^1)ouRAN~sUd7@PWhS61 z04>ZCLmB#;PH^gj?rQjg3c3kBU-8p8IfK6Xr(xQf(){&htn`t4& zi-w%`>3WRDZ2gGhy&AszB%ArhR7;iv&zE?nU?z}cDmIIVN5m!g!fA)eBKU&IHSPHulp&0sm%xj0!lVadqE I6vScwAL}Y}ga7~l delta 176053 zcmZU4Q*b5>uxxDGwm0_1wr$%^zSy>tjcwbuZQI8E&$*9hp89#JW~yhpdoz(OHiRvn z445NLz!R7n;H$OmeAEIDnANYhPcgR{V3KsW-AnlUagFHdd$v%6!6r8YSd;-GCIX6DXh03?Krg527(4{C z26S4B7L4r3T-^Mlx-emT`n3Fm${R_6{bOQj5$L;z{~~!yFwr65rvh8+)sJXam}brw z`}9;>MRgjGTTG}k+a)n8-YDa=Y=)q+Az-Rz)n;b7ZYD3tBE5pAzV(6qJ9f+mJ}h>XrQ|{74b=*Y+}i-Z)xC zZ{7NE)jpCCzHNqfdQKg=dT|kjc_FG6De}uyd`scL_ot{VHhkYaTAcJMsc&s#5zmE3 zQ@r6#Y`0nzx{Clb@yXE%e7|Xfj1mpfFOn|;XcK#9KVTb5Fe=sC7xpOKl@^EZ^VxB} z_G{=--kZjfuaK(Q-RXP<;uIuEzQ7^_&9+Aa zE-DY;C9RqIqbqjB{@SRwS9$TzkCNU2Czv$Kml6sQhxt&pl5@u7zQnk6TxL1XfYIEr7T{>eUJ?=TceyOzNtVE5$M2K@5$* z8VsMEbj?xG0;#`rYeMj^lgOM3({D4U zZ&iM-sTsUkE*|D%us^`!1(97ZhI8mtQpSgsBHL#?LbE!vWl6rufV5=917+Vo;Fr|{ zWvO1UEdCn{bHhXJV&>R#?b&yliwA;{Krva@JEMacRn;sluwIY(+a(S&a18%|@4@Bx zm2s~6P)bo#=D&Xh3r-@J%PMX_p^ zFLJP{GF1=xXT9_1Oq-9&{`u(d*k7wqy&$MFZ~)pEEzHqv{{xX3fS=wB;QSG$!0;V( z-R!hyBC1T~jiajHpJOKV)g*`s?v>9d0MAy4dk4lFJv_V@kM)an7A~xLY^jVv%x91c zf}RYLEJ3GD4(Q!?}UV%fAC}I||S6L91swwY7 z9pwP7l~1UsDn%aucd#Z*>hV4msfhB4>IOSJwE5b$F&=Pj8+nm5f|~i*v6)&dzYkvO z^-&(V&tCS?TWcFF0N~Ak8BfO)O^=7DQ{$MR^(o@{rMFT12x(c+_fs^WG=b^jo?*Q&*-&mBIAEY2t5cja$0bIUJQJgw1Q2gwYeUq_I5rE zrDfa&j+%3@t<3286aF<8pa->!{1CQEc}ikc{IxfQKfoveX!C?3%P}a30YZWklbcKt z#~jymK@zs+-=(w_b6()1#%Tmj$3E1cCq zx_B4ReS$Ae{fi}`hl^~gAs;EU_^=j1gDDD$n+3uGnwcvf4Iq+{{^U~8t-qqrf6ksw zkEtm1k{=%iY73!cz%0$2ZQzEW|8# zYB%mB?v>8EXpY6_K)fWs9nCUqc$@3~VQIUWl9H+KO{bmQ;yxm=p!&4>oC-RWFL z^oKSeiNuUBe|(D`fAMJ0D)va{II(NTSw*VX0>OcvvOmcQ?SS}@%6@J$`bt+ZG(2x> z3;5$fMeQLfGSSsU-*a|+%PP96RLc&mDh|W|Fk#h*VNe%zk8q}&En})78tBm3X$ql# z3OW#Y3a;8pMkt|tuVU{ePNEJ$F;e%M=)e&EAvL9+menFy>~XgnQ364FZ$VQS^F^#% zs&7dkByU`>g8W-edAhCE@n)fAuJ{&{5x?UW{T?Zw&@mT-h?CRaUA zUrTuwWjAlp{Ou{fRHiQvuL1_Gb`;ff1asu{?m)zNMuEBBcSRUUB>Ef95JVD%n69ne z6+K^n0SqK4hHs(r-E)q$q2o>>m4`$RfblgdA;x;wbn>as{C+s-6ho3qqtf*@4B7?3 z;9zxGDSaID3ca{+=`r9p;M2)C!L5^GwbgR*Nt914>W48d(Z6!CAn8Nk{(~0~7yYQv z8$NtvxfB3y$Gn6$5%Ar~E^bOi<9@fu!`r|Wx##*|sZFwL78op|EuS#2d95=B0Cx1q zbpXlk=}{;s1I@1dliuxSptx$OdH0TEmz&>*R#R?=8KO+b*DkTgub3N<4)byP50;CZ z#J&6G94C!BEl=DW>2Bro>_Hm#!BDl>m1lF#*ACe~8U=Q|#9=k#S z4}pEcqxSyUmg$WOF``0dGPT+OAf%r>iLyV*>C-T-%CVLoSRhkxFWnpNXVWXN1NTUh z^n^gowU=n;wvo+Uv|OVjoQMl1)E=5+=GUi(?=N$ukT2~d^P5FG@wGn+l15SuM#?~C z0@*EG5KAI^N(v9!j{QLC7?AMXzuacv)aa$aS$kBC4RHUvJ8RVAhO0VEL|;A*p|Ie_k$7qg&^q;*i~GSbP0X~HovNz2bsbQZMr7I(4ql8qoiS$-06?)ykuZ7hKKg!OPNkAKas6mZyvfF-L&of>C$b@@&4BB9l^j+XX z*Rg9kyAGJb=GwxX4fjl1yA`s$4#Pj_(>X8^r;RC@(L_+p|Y0{>}!^=Ufbh@Vb!4evn;{z@)(dyulFjgwJ4%S;M@0mZskv zp(M4N>834h$5^q`&vF=R%IgozEmDuCqkWb4P}1E=KMh9}Sh~ZOfFnL#%Og_TCGlBl ztg$zrWp&$Cq&o-awRJkYdG^XL*H@5@)3hj^R|HhmbAIp-9tRfXml(T2K}v;7YTU3D zyWgsEuA%_|fBb^)eBt zto;nJzihgk`B<}89o+KU42r<;pRL`iw((fGqU&Sr$M*-H05gv-Iq8Tt__Oi!#=)U*DJy_C|vDX_$y=1^g4HPW~+M z_G5eg9JT*}ys1^bNMpi>MN5J~L<3>tOcV43rUImBN!uTdAoqN#Ptl_h&6_;>Hr21M z0Uu!LfkPvm&XH)dJUK~Z9930)eJnMu4ar(g3?4obM`YoICJK9%P5qNf;vBCSJbb^q9rJCctr$bV?0i}Dcs017$h;v~ zI0IyCUXD8&yG@#EVE%Y?u(&>r*skSAU-_hoeY^DEQ$)4pDw_JS8)GRpXI!GrdqUWb&lw^NE4(iXvo`IAo4CqmyqAE6FQ*wOO1#LDBBU~UVFtwgBDxQP)^ z;&~s}JcHrKuET3oPja$3Ei#e~TWWrk0&7k%ll_`H^dev1jhhG>*Gu$cZv9R7)C5SD z4I#J$2+5?i`?h+3k96ix`=wzpd)JwZH>b(<1dE?<-%h9 zG3=UsmS#qEd?!nRx?RB6uLX9_yZ=Bny84gi8)nknx~D1>^f=OBHz_{gnw3mSQDKR~ z#%wvb-_j%S-|&Xi1$OJoa{*LcS8{slmTT_|_{cw@t%EC)>j(wnmWiB|PPIc)Om;~) zTo^d=QwxD4vdRzqvU|Sy))}F&-j;J^ydP#Ne4)#WQB>Sh-l7<#lXLfT(AlVB`1ezr zigG8l2v-;dql1rECL<&4R`5HDFnU=w97UdPy?S1Aj(|)RuHgQ zED~Z&e`*eEHCE?$FfwKk54pY+OJCZp^R(pmi=>jfu7QyOr5 zN<~S+*yIbv7?8~g^MhFT)W;~|?(e1GH|Nb4!~lPxaAyciBuSh!uyGz)!K{t4Lr)0M zKG6+yUN}78U~b?uGq+(VHCRc9KEsixPo5T+i$C-fAQ4~ulsmQoUJUsLj@OK)%^iSD zVgLpfRCVtj1*l0kxAvE_-_Wo-Xh@Di@FmBI;r`3o*phQ17cYL&AIAiGIOv$Upa0l= zOsxYVz^yUBQM2g-yiYL@=R&Oy2m(+or0sw??M@*9tLX!|s>~qeuErRaM z=g2GY*Q`0IidCaUC%p8f3>D zx}(b#y39ZRVy#w0fBZ=hvi*rWyR8^+D>GG(O#)c3eOmnB&0=EJA_||CI(U|OA?A;$ z;hB1-FJQ){ctyZ0d&Q|2NDZvqgrFRbfcus#Lzgs3CIMQeKF{*}Ox=os?8R{%#N-CD zPqW1>wapbfESCd&r$#C7Z~pJDEvD)!o&2;$Jt!lL{do)c*>x5gU49iJ*VUFvCo5Eq z@eNkLJ;T-c0^;@w{|-;aXAaQf99>rsFLQ86EMOX&epL^nS(gO@m22`$9m=~W#K*bM zUy^1F6NE3g{3eOjV4(+td}TiL^fwl~O@^eEGs+0bQ`6aMJ-JC|u`Za5ZVM zYP6ENZMw$ZbRdkdryt@ZdFy@=ad&9zgA#=F55zs8DXZi}6qAqtS*cCf3M>*6_3;Hp zA3!un6$A@9e3>^W>kI#6z(rW4CED-nj-KO)!sHsQ;fB*MDm>TTPT%v-zbuWmO2XwY z>_UA)QnOGd6`9~JLww)fcX&s|SPT1RI*^j0Ir=_3Z2wW#`Ks{8KSaXr zrWfP_)Z-}??D=d0z+GmKDEt<$6K4a;Xn?p03boK6-p-B;$An2VD{y>I6obx1k#lB1 z_mm7|-}fg?z^}3S!J3NFI&oQK2*HZ(n;sFdqcN@ZRfB(d z{~M~;F3H9|23A^GD6s$RhbV)zT(zcHzh|R$>ERkc6`&Q*!(i-gp}tW zHcgenb3Q-Rx>|cIO1DHuqd{$Wzd1GokCNbm8g|o>d1WNMd+(GW_oS9Y7=U&LoE5RS zvu)E7jt%udkz~bmlT$Li5Md~>5MT={^bB+#6Bj3>y340e;l6Uck16nf>g5H z#yX|z8bsR`2TexCM8c^YQIRT@3Sg07m05f}^>q(ovH)?rT0U^@s%L;J^}Y<&nMyaCNhd?u&56&cxU0r3RM9;mZ3-{x09bN$W01XzuELMo z-Vuqe<+0L`gPn9nk|#=;Q^K+RG@<6k`0r|G_3aGqh`7^H$UwUx8qk!@IR*ylTbp5h zn_-aoc6s@Lv%$q_YxXKwBdqzPL>^yNgtppL;r@u?4APW)kAEVVHkWseOa*>?>WU7$ zNp(jAf;0fm|MV~z0J?t7N{7yy)Dx;L3=0f5cr#6Qc;HipwmhN&PSzVv7DmVIyvSKN zrfh1QW|(AqQ@FCKB7il*@AcGIt%tx`MC&prA7&IEqmXoJ826VF;a9m%X;P$~+x8vP z2cZ|#au^oDNZ=|StK{K4CTLw_y380?xDkh82CumD>jgZ70O<*&bVH=zqWX^36`b08mU2eVv6FK8sn}iI?kB^gC@yKjD}%XT~frN5DH zKh3yOYU+?V-2x?nF@fIo(v*fZ7e>=)!;(>LcXOXClb&$|ObIN}+z_TL(t{xna_H$t z-ow=QUkY$6c7#Xi0+X7@MHhqrSY4AmgGJnN22Ezo0S>O3+h_E$=1iQx$W%-VkZZjU zq%swPhQCRJDu(7>ab!kl7+}>{S{=7u?Q)_4$bH?^+c^h|sL8_#zR8HjembNw z4zH7xt9JR)qrtgyzpVGKRCq3-Y(A5$v-^|=vow~?z(3D8l3u}pEW+5@moD7wt$kmGT&>YA_k!mO0x-rxa{2S0<|jA5@xgg1z&e^D|E(u(tDESd zcE3#3<@)BnhgYXoYl}RyOdN65-DM7%QbyhV7!=GIj#;7R*V*DH(ug_wn8i^LIp#>a z=xeyxGTr!wC3@>b`{gsy*!F4u{m}%EN7M=vS0`A85_f-+t0+9OIPRt8zqz)y2B1_P zDM|$HUyiqO>gbLu?f;IueS|lr%z^bw2s*|cwVYA_@#}k?Pe0uw#@+*RgGETep;qt+ zWCk(AJj>Uo-yaXve?ezXz<^*(?M$3qoJW)1jhSM@0Cv4WJl>aRX>wZm1OPbaBk#iS=HW&q|L=^N#rw- zkPr$bLFebq{eIqzvGXSqA2$_Y7v%l8e_oxAa68p)`g{`}>b)iK^LQk#aZ}YecI@%y zypHrx52uPwGZu?7Nit|MIk;4$lpi&z82&>zAvB?jJ09}s#{N?|; zoQ6#BevD*_bkbnih|t9|H&iK}6jIkXYs4#kG1aH&2QCTnd>ZN{{nfMoG}eLmW*WCc z#QgHPIB&h_ROGNSC|@Vm`63{v&JM8Ka@+4v^+xvD91U5oN6t8WXI%Y$Ua7bxCa}N^ zQB{3FhCx25UprRC7BdCNPSc7%GS@63%7w>4+J*0~<)Vdf zd80h}sr1UQb+G?Ja444`k`)b{ClIXhxnTvqCNQe-VQLwi(%Ulh$L*ZZRdzry2B#}O z13k`uUBgM~40{joucMH%h<@wwRTBPa#lbykP()qK^DYqABjtHc|5I(*d6HzE&E*db ze{arcc3kW!K7HJvMqPw8SE4kT!}@&4&4r-kf=&Zx9HvKv`FAklg$aww&cwaZy!ej9q!hx6VAZ}LeY68Y&(E1F zWPmYxya!BpArti)w2Tb7T0PXnBGW2RX3;FOhd_FYd{r_Wgc8j}HHO#7bjQspIcb0{ zUi51)@LS}xo&`RnWzV*_mslh~gva${w)eXw9a6#F@JFt-QoHODK{=AwW~XgQT6 zD1RF{#?}q^UZdUS=RU{Ucdv`TSMxN7B$N9g*?Tgya4d4`$9P+~tW41`Sgc{pR>%;c z(OM%^C0$g4fCvv3MyjV+*U?}P2*)s9>CJo$DHeC7+bAb|Qg->&V2Jx5GFVj5ct769 zIM)wgT&`WHf)VPUG{5}Zs^5?*e!@ls=TxGPJDR8lIW#@hz%$m97`tdnn5xPZbNCpi z^cdc)ne76uX50#c50H3t5fs%*I}f59Qz!wiRFbS!$KtlFS*^_xndF;yvM5J+m}57J zmuvIT%h+Q#f?ayg;k2U>w&BFJW%X)GI`%%G3`C|Mnt2EA`#N1z@$M0RPWo3~%b-1p zT8+z8wxX}K>XNd*7w~ibeL8y!SjRE2%{#hb=SO&O6I5?$L@AcRxZ)Yl|tKga7YhRQw4 zfW*EV($~SxJv=yJ{X*DxMECIouDV14+b4nLmwiGbLFY7vz3q^ML^q9tIM0d z*b=Kbpq3Rn^(r;r{?nb^Ir6}t|Ef12Lu_%#syqKl=HcP=Pats7pMk-`{`(H#LUeC? z7NMr00(#R8wtUZO!IUeDRa-ziQsr1zZxWfGU~d~p&(wb*OKoWOdQbY(*@n_TfK?e& zD-JYcdi=>}CjP{`TWOmJlV~|0Ct#<+&2a=1 z2LD^J=Scyuj+>YRhR=?TiW7%tg+Dle;jVh;>tItGc`#vz8eowmcr;`zpz-*82G$1G za1x$zLZ42{+nEQafRT`f^s;+`T!;k}M--)H3BKEQ13@xK6TINB%57FT|6#X-@$MUG!L+RVeuDV$``#>Bf|p_?P1oQM1IOnd2V7)y6&N+mG1Q z*Lh*;1t(~M&r@$Di4h}~<~5c$)fQ^6q!{RKFAC>|S#}jn3S^+EWH}izTv!CtBQui~ zr^rjKzec2W@{WGP`8oj-14@@tpWJJleX@iH(0>q7F;VX(@?j)cdZL!1O@ z3;hItA7}E44Ko`F$5IRG2E1BxaI;#VHUxRsZ7d!h#|VNv&zrD;)ObLRGG&W#oh(tw zqyNH3YA}zH<=$`d2L%eWrmCx_zPwA`f`OwCvw<3;`z%H0#jye~Sn>lhCzh(e4H{_B zc=lU#B$V=-1K-bUdwqTq+B<+t&)I+8PNvs~yex4v;|yayse?EOj|w4LQNy`NIB(+e zzx5p=e*KvccYJ(bp+f}l^&EKR6w{B?!N|mC#d5xYB_Hf$5rJ}mI%lQ!L0N+lAnCwG zFj%)_GFsE&XQ=?GELkb@IWe!ymJux0mkFm2z#S`UQ+G+-Bw%Fy6)5jd{VjIcm-84{ zPQF9nBdMcP4OCgh7{_-Y03dpEP*$RJGp2%6pV2c|^+L^v8fyv*3A_d>gdN6qu-uIo z;%(#*|ER`u{yy;HXtaGOE~q$b3%M2B`ZJ7Ik!lo=;R}HODOrhSVl<((BHBqMKgk;K zrxp%~20|)bP`ovutsoNz6%kTf$Pwr&%H9j-O~NKD_^gbySd=U$1|H(?EeXbGfZ)Yg ziQgy1zAhMyu2-u7T?(FucJ%L_7{2A?-1f;-W8E-EcSalCppXe|NeXrcO*vnL4?gpzXfByxI=vm`T>1~WJ5EarAr}jDZ=TI zWS6L6=k7Oz4>f7|wVeI_d@VJoMb|^IDU+(cl|w+FZ*6IXg2tnOAZT&mb*(77BsX4I z0H}gCQz{pEA`D6=#WMa{cp>kU4eEx0Hcls84h&_3Y0-EY=X=HxgYH6g1cApj@+S&e z;!|VtetBIE6c2czOjHdx`HHq)RyYwnNl&H`6Ll-8R-zIa-l~tAQdd>T)L@Qq6b-!d zW-fr8ek*%AQ(S`Bnl08p*=2SPigbiF#FC-wrIE{%VGW0?HF%$I!%+?#O_{2#vrizX zmTBnBrAFHn4~lXh5y-Pit<5oES&wK{7DfRj0bApJa2T880Y*VlNNH@nqorR^;ILV= zPDzZDMPP}wkbLaQEYu;Bdl7*!*>f|6^$7stm^v18@UInKCdll`MkdY&N;flx^_mwYDTk6$%FxHS-t*HVXiE zlNg`eYyQC#sk0mj-RX%x>__yH>R8Lj;~=n`kUfWuh(VV$%&dHN(7QiX%#b=A`M1&i zJN#Jr!a)PQBSV_na}E`aq;PSPIz43lzh0EAm^Skqmo)00aNm`{(SX!=2+Wn6cW5Xv zo$4n>-9&p$^G`LO;gFUUQLI6}R&WRa_amint^LBsqDg28hng*WIb$6O$K|~L8 zVtc;}vZ_(qGY%?v+EoPH&@K5819qhQDrZ^!-NS6HTBL^Ly=o{-R~(n6W7Es0O|aUg9!)YL)ba60s<(wrGa4OEG15wI+sqz#UgO*JHcCmO ztVAG>|D zhfyY^5}CCZ{WR>qpn`d{URImuE$Qa|^Ur&_+kpOA%L zX3LBRH_)45EVg?}(Unj&erBy#IimhCx5d*qRAjSwt$zqGlZjlW4LFZK`(eMJFzc%K z|MwzeREm7-F5maD4DQ+J&++_$&wn`?t%t6>vI}TD z1D9lb0PmV9&uvq1N8Kc<(#g4AixC93d0UrdCRMxi7pZZoORBl8*&?-QW(8vz?&3kG zu*k(&Atb2b`F$Jg)msKudVbE!Bj7xvqwmj&FhH^c*8c?@BL2R;a|A%?%U2Q3eWm!% ztu-^6__h6VB21bd0Dmf0>4pmJo3J&J)3I5^mt6!){3!Udk*}LQOLx#Z7~ji4BZ+p4Y@ZN**u{C~!^&rhrQyUg+09 zD@icef3W?PRwEvz)hhQgBw2?@J3#y4?4#6{-1jO^LOKmY0_rJE zEiC~94OK%%c$^a55s~>l?t@;dW{H0W1wm0o*XAZl7E~g2s)v?Ls9_q=w&tC9%0tUn zpl&!DEHbQA^`N^IZ~7UX;w@M$Odmm(A=$yT_JwX^Otk!|$Ohaf{T%<>-4FiR5jrw> z`Fe2RsbSphc|aXn*%ucwg3D%Mf+ViAzQuq><%73_lq%vH1FCRuzmx#ds?GI?91j?^ za3v(iy|91$ky4G!RWODR9qmUZBCk?>bpJiURL@3L^3b6KdoA)g4t116&k>_A*P%q8a5U#LDgxiG=Uk}Lui*}#Q%9C_&T8MXj+ z+hl$)G8iNP9S#;fcH*epM(8aGLy!84fZZ zl)d%T?Lt8oqKuPb<<*&LfT2%a)DZx>eMS>hOJ$xmu5Fq(MF@({VWmvuVF2c;P@%B? zysBb((|b1|lPuaefJ^V{SI#|;jsKEcEv4{uDhYSVjc~EvP=dT{&uJD12G&?Vf?YRPE(ZK{(7c47U!gYF(SU?yw0VmDTFIw1c%fFtCb!_6r4hS~lKxtU z4GKksH-w~J#~2lfJlN(k7~@9(93%`PCh`8Fs0W`+R3$^P*DwF-VfdAa2DGoN$Gv}} zBX5tAJy{jhpoTY@!kRO!gaZJoI0BLSVBQ&dy@gYySxDOQQUYP;$Yn(c=LT0YZ_4hj zn?+Q?9T2YDJ7^e__O9<0d|h%9R4ARl3M8&S!E1}gdV{6uM%cf^UV!TOAtXwo!m@dw z*$$m1gbafChecW^no3~|t?xeKigaGbv{yr^X_Rr*Mj&7tY4+Hyd5J*34%prKirGSdiaB_mmVJzb#TMqxrz^YpXtn^&{ia#DI)VC)>0|QKt3@1OSBmOQpWBS&4w}BF0eYd;F!Vk z8KR);CC+ef39SjLg9{*={`bnO*#wHXRaRbc9TFH0*J6_~RJQ!p@1iVTK2N_5Hf=(7 zBeowuXyMB5L4oEsepvuXSGeaM?O`fHsuUK~dF}`3tnZKrLPSuY;5N8x=+@BHd#^qb zD3TqbdkG;9q65d4ux0 zp)XvB^(Icpx9kl@9=m{&;?JP0k2EhZ8rbBJOB(aC+l*ivGjk%QXQ}^_1S9;h>P@&f z&4is`zK_q2JaW z$tx#GoyBxF@GC_3W090e4p`cV)z|<7x5PV)sfY|}m`s-;C!?&02iQ}rM;9T8C)n|} z*=L3r@Ub1;53aZ;Py4hO$EuzErU+KgrQkuJal&mG*#hj2EvP5Y{|(8?(j)Ry0aqfJ zei${FYW|ZvkL~Ft^~==QB7GM_k^yDK>lfsMx{PAo!S>hB-}hN(w9~qn`IfK6m6k9n zVCx&AP*k z7$t+&=KL=U(bB1juKW$=tyqH3MS@nzgJxJ@u(hk!5jn5lc0mm<3fdg6*NnpA<14(MZF-REIM{AeYB<@p`ddOW~b3b-oT)7gi^(R^_Sk`L&pIz=HRMd9(?P^TedoW$QBh-#RMeuIe~w$ zOy(K)+Q>t1Hd5gQ%mA#cyuL+= z;V)iq|5)$90dbb~zjqC9cto=816~rzTd9M|o()B%h!IYG?5cuJ)R!K{)}O97W)G1M z-5IlE55e$E`5b>ZMzuzJ{Jzgo?npe|$~L--L^U3=pUj44^Lju3*4c1&{9pdd#__+3 zL|SMgI07gK6Z`+S;8kmC#vMr@`{n5?+8-u%@bNETq98WH!AGTE@gbpW^Hti*y_PPk`9YlIYRj?)7v}6Y1ncsHBs2dkxb~ zxQgV2DQsHX=IHDN?{)Ouo+)(w{<(bxc(`MpcG6Tmkc|#I0*jXTG{0N8VTUy2%8i#@DO>g?30+|*I8U5``{pz{e_1nxkXT@Ihsg45v?<(jYtSb#w3D0Yd1R za+%HGPi<9`{?||DdmeeM@WTjE2E}$18GE{gs>bC)paaoz(~ zYbdeOBnsdX$+`HC{bc6lBr-z~dIOU`#9WCVjrzh<8d3vcCN4fZX_gM8+G|}^`?S%p zzTa~&js4NiJmXV`<4;?e>OPzOXD92sfAx0tRSmHt3|Y$@mKU!BaPiD*O0mxh4f;WD z6zU#<02J(R=qJ2Ri;Y2Mr-<885J@x^F$7*XoDY2myrvG!Lc*^xGETq1YOJ}d`JaXR z6zh~ef;uhwYwTVNF2ht+70uSP4%bak((UAy^V9U_oQO83&pnvGfr3vYR@v)okPa7e zug%d`UEx=?x^YU@Q_Rw%q$DPJYeEaNr^7+&fGGcW$2t56jgC4Ruu_a3=G!C=z?wG~?koeuP!l+w&_}=XG%}F$={r zg|?;|t%XVOGGd|cj|noL89Wx$TSB+PjN7odr@)1m$WvFt)L#?xEiD*gIizwimfA{R z05869C+3#kV@Bu!?=LMVM;tZFVWB%_e3uwzQ~I*KT+>_#qx_w3D!T$3%S6+>k4bsd z3>r8Wh(t*0PGYWxO$rh&9pYRF;6mRM1`jo~ZrpTL*{7y9UGxFE#lRAW`CKS0KX!#X;pnLYj2c$Zo-dxv}GoeqzZhB^f&S~a3$ z7_vTDNsPp6gqq_ukZWjqOIXL_g}( z8$rndCg4K09F$Ogq(2tX0x(8ayHHr~#1jRhN$A1CpWN5VSW}(*6GW%{LxLVZ|4=;! z@e71PI>)6b|HJfZH9YWnoKPLj14s)nkMTw zZ)@2t26L2+NmUm_rLJubHy%KEPKjI2|2X$y+GQAdVdUd{;^(9IZ z1bAg;Gl!aeBjI(%M*)+XfUj+)y%(X=rtJ|1L|-=@i@E01;qBCJyDiT|3P5VI-#8cc z2DTtF>nE=K?&20I!s8(0U{Byf$%J$&5YPZd5dy*}w)Bl1*l z6~lz&%ddDL4|A9$v!>AH=xZ+eK_5og*~DR^CJ;Q01D}o~M2qv5=j8x@eG8T`r$YKxE#Wg!;yc=Q6BlNpr z>+%+f))IOo!)wx`B+nUTZJX9LJ9b#k%TfYQjHO9B;6x&bl6@h;D6mN#p z^(9-7#z}yv0cs1=|7NHfXct6>O(pL}Q$!QEQegL;8U@APOu@8rZ0*`}2Xa>j$7)%t znwZ}do7}vSecyn`zrg`fFFOxb$i?EsbDUH3qkT@O1ZrVb{MPE4C}NJ!QEtzNLYz+t zht)%0(~SfWzD-NfF_(UbP!6F627+Ab$|3~=tc*V30N@Dm$l<1;3E1OQe>TJ6*q-;@ zJcuOSMD>4CZ|wT8FAhOV2q++Efd&j7?XUR!Pxr-Y0ZZ_N)NUwhkAI{QfBs2(lGAcg zE@;n44!)fYNw0I<`AR~l1)yKjb+lNrJ3~?%E%fnbcwrz_sMUs6TAPcOIobCcXC)r3Y^dS|-uqhA#~N0ju|5ljo2hzFdFvR1<$Um(R=2ymrtmWjlHq7QYd0mO(}ji#}hkkAhzqAlVU1 z07eCgUZG?>us_sf;m1hdZrlacCwq84aQBN(h7dH?$Ss#jXhB&c=3YWIGs=YlKG z`voqx6U?!df|Z(>qDqllPFFol*`qjX&w`DbiQbxZ_Wxt+ouV@fyJg|nwr$(C*|F2H z?erU?W81cE+v(W0&698M|Lk+I|BE@s8sk~F>zOrcR@Evr!kEx`rS&a#fW04U_q55W zIjK`Ik`DU-D3OMjyiU?8HGLvdLb#+P1pq9+{u=sUvOK$`)m+wV?zx=iI7boCze{%a z170ag!{#vd!d_7nl9$&=xJdrQ#b5Y`Lfnc&GDGB5@{z;t8(kgJd&?JdDkcET%?`U} zVGLV0G}yoN9E625s~!A>j-M2RD_}sjd8!V6pA#ay5?1~i1E9rl*1+^MrP=H0fpC|t ze0;CwQH#)h19jJifPHXB(15_2fJh0hU%Xs?W_$xhv{=;tZ!pWrmg3(4jta`j$&}J3 z2u}^r(y93gP585Iz|1J_Bb^qAWxbn|YuhjV2Wv$3D{?`rtYEcJE z6l}1dbmocE^Wyavf$#E^8D4tZ=Syz+s0$+iG3hS>;gdGy+(aM^X@}DBEZm|+`VKt1c&vrEn{k9E?3$4@ln+wRzM&Ty_ z7d6Nd%2twU8F|zmNowIE8A-Pnh$qE`6Jx^0zXH1!7*omJ3<-I|u8sWBWhYYW_7}kG z6?9AD((B)H3Br$(zapaxCi2cRIxzHk_W-bp*4u%E2k16A3!?dgD)kB{Y|Lc@JX-d}X=Y)`ol3#3h+Ei^prJEx0eNh<5vH!Zry zz!&#yZ>G!xz__`u%c*kKhWU7>i$+*=ai&6f_-)basvW$~7qRrd_xfU4f#O}L`_V@y zpJ^H`7u6>(Os2wU`5d$4BCTjznfZcw5?io-X<4vF@$spJVZCn}JE6q@RJ5&G$X1D7 zg4%^$!|Hn%(52-ksyf{ca+mo_SaJl*LQ8^QF!&;04M*udIE9J!g@NIC2?S&nuj<#I@^usEC89a>mOgixa z|Gd-yEe|MP@9)1MNbsgEKE(w0RRb|0qU)!R%D#a_zR0vlPq>b{ut1P|Se);3BD^r$ zlwQHk$>@&y%$X&JY0>H3MQEpcQhN!c(8Vyx0F0>0OoaPjYa!)h|3KHjBH<6X;aEbu z^?lQsnY(?M`=dJmP+Cv_if})=V7->4DK7yH+r$d>!>w4a;)d26wBZNr!mRWwMLsUn z`BUA43DFk3RdR~;jDO{^^8A8ZY2?)KT5=-O!=vtZ_)^Z0IhQZwmG7ExxJb?r*yd>t zWIom>0;TzKdACUMD`)AJ$;>O?$Z;Be=oXQ`PJ0veVb695HE&EzRu)eU4FUTZ za(OpUN>&A&!ZAHu;6m}dT=Fssd;zafq+$UZD)wbVZz7~^h+=@N&=7Q4d;$p#V>w48N zpbqIJ3~y%ul|Bip0kzlkl2bk)jXoFpJkxt*Gs?|+Me|dn9k6S%0^lF2V8C~P&iJ3Uk4U#eHu_?uI zK5JVuaVXdxviLmlpw4*wH=Oz<(j6Lu%3LnN<_LdS-Le73gPjoZTWS18akM2!!kNHt zH?e;%-`ehBH4I*aaqHdf!S&NLYm2K zToI$cJfm?P34( zB0-=Z{gOZRFAP%WH!fb>>U<-?j_{OPbS7>amtC(NYP4Y$JhsgWE@ABK^6^yWW(-)O z+dS;$b|6(8Rt`ux(h?2z0@4@lYBMVE_?-tpS109*m__=*c%E4%NC+;GkB?t}te-5eS=yR3f=!`!v z7~pHO=ux5Od*MyGM}ZFELHYC~K!v+`{}5LuivZ)u0I6O6D&MbMO9o1<9b*74Fg>rb z^C2A12Q6`g`aK%4JIzq*Cn_B$?=w9HIJ8K=0hdP+=T&h-Gt2c9i)GsPAqBsj9o=JE zPYlz6@ROS7!W6T~&q!{8hi4uKn)G%j!EFfPv&q-Oq}t%|ZianMtJzpjiv9{n)(2Bz z?}B`3z3bO$Jcjj!$YJc7SvGv&0}i*uAoe%QKo%&XgC^v|@v#{Id6xpsEnk)fNNE3a z)UUUlVxpfG8&|DG(d&<0J{0?uI0`V@ZCurZ!Dv)jXu;CqhAq=2IB)ITLQ^&b&~=tc zK_+57U;Tc8a|0JQkyh0lVr1JYXl^A_=J5fWjZMRC@1A%1?EPG*kBmsHf?u7ludf+2 zXA-)CL;8-HWG8L$dRD?j?s6~%#7a4CIdYD_4XOzbN?TDi-?OVxB2HL+>n8uRMghId zt%Yzx^aK&+$(*|bD)P_hGGCv6!;k_HZ1$iUuJQSolD@Y(9k`-H zjCtzT=if@Gq2LpR?qL?lBnCT_vW}zBzS4r?>^9}D7odru&gwLXnL7i2iVeJ;GxQLES%0p>yD|muy)3M`>Y>J{NdN4z ziKhNapWv|>`^)e@m%W~V#Jyu}@|pJ8l{=BcYGT;Xj>3e+I)F8?Ch?T$XaA^E5|o%M zo|81f0xlZibnZ=3_BRT|t9H_&KHel7#p;=vLMlHw!TRH0TTVKPbdM(24db%q{f}S- z=O)!|yB+}u;gQ4m{2;$SR->3`fxsv^X`(@ik&_2n0q6+SgllMkYbJ-6UxkOYF8Y4N z+<-Dtni4vO!%Q0Iof?gAMW5`b!hUu^-O9A(R-FV+yCa_5%KP1VG^%hx<()pCWS%6} zMV#*yU^>Nh!O|r`EL8mC-k`L<3yBFg#;lKuiuQlN(?4o6|2GWgX8ZqC<|$zf;BcTE z9RC*$YsGA|BL8`=IcZS?<;PK8v?M}wv!TSL!+^a26OM(4NXPI_ZzfZ4@!*5H+g2T- zQTe6losfbtV_c^>qVaF+$}#2Q>w#kT`n`j*C1>7CW5hKMqoZS?TF$*lYELY=qr>ev z=ER(}qU9a1^?CS|Ko#IBt-B09E9GWP8BYgLi1*rD85r(xX1%sAoXpo}9}t^*tt@3O z1JGM>(GV;T41JWP9fhQ1Pt7+i{#n^BG##{_gzoV?SUS~r_eN$Xg?ESH8QAx6E5sE5 zi(WF|WxrUUF~*4TCh{Jhibj1qs)d(8Z_2qiy%deS6N3Il2ZW@ptl_9q%jd>^F|-1H8L6+Prd6nj?Ejc4< zp00j7y(I0mTSYjmHQe>un9ma~s*X3oX&$QWHUOm~@FP_uN z(4QjBRx!PL)n5+_L*2D4lrz@)*FiF~H+ee7i^kMxu+U7T$?9j3M@R+MUDXW0$!s_G z8#a0Ma|I@3^07(Xup*!JTy(L_ZV=H8SDRa~AQ&m$o_?B2=^|AYUtlS>rpYOCyvp`9 zb5_iPFL}anA*a5QF6XPe7V=JuL$5Y`H`!q32l?=H*sP4Oj+L;c#2-As|rO!>&B zTyTXr6;!mBHtfJv9&So36w03P-rO3fc?5=}^&h$$x`=!}asjYy&Lb|c^OzZSxm~E# zECjYUe^9c2GtI?I-%kHSHBhEVH4vv*orCaRDVi(&(ea8oZm%n~z- zuMx_Q)dyYm!JMxdkM0bwV%VO0Y?`pmp9Imzp$Qs%TM0%p6(ZW^&qywfW-wGS6r3VZzDhX%Q&2u-62YXY><{$&^@{Q&au}{tmU>cR@_XlA!8%nl7 zDkv(vTSsP}jpncrZe0LgNLV_0B;|l)F~K|Gk|VBy2ss%u?{?!+Eqdcc^iD=@WUb(2 z<+^P!J1Kq|LaG3&zdAd`jCPXx0pE{|z3rgW2X{snzE%HZ8nub)Ow`*_&5wC)b1=Q!MH4NJrE8^ z3wTJCggYF!L19DODJ=+QHUsTA*KbMpJbSH&>`jRZ1$c|qz9DzWY!<}qQpqdRcBq8} zY=1?}oQ$i}jamTC^v?Bi?NvaRyOb0{le013F+N!^`e{zaj3L+RX?R_52DzdI1v}Dc zhkxl0@^|-gce{WnryI}^7D2GJBdSotT+|o=79_A~ty`?-=GY^0u>;@}%}x;^Ev>_* zgvh1Tl1~|HF-3pBcY{cF+oXav9zZdhkn$252%-`v-N|*q zes_LK3rZV1;v$9(C{=!EPv>%Yri0?jKZh!Q=S7nn>h#jxTaC;yw=CnmC12!F6Rk)1 zHszY}1!)N3L%&Gw#_DqGPfw_R3`$TJh+^SunR^wsbIHSeP^62Nm`S$UQ_bs$Luey7 z2KcTbr2yKQDE9k{3<#;`6_!EuUwR3B<*Jz|SXqKKA&61C_nt)!sP|YfI~phVf2o@s zw4x@ZRD$-v`xb9?gK)$^Abv{VgUSAFY1Q<*OcWphS&Jg5Wj@W@q_dm4^K0pQ3eDNR z?YoCMrdjVi};TPDWv7MgK-p+We(Zn)J5!wGKEw_zY^|A+7zv(Sn7eJ z4y*-f{~j=LV!u`2G8=rts}$t1S_>T4F@!&Q5N;xtS7Y}G89(uS6mj7NKxSv+(R6#4Y;nG^DnS?S6{=sIUHNGyoaeMH%ufxOQ~CbBbt`b zX;VQRtTeqRts(1=9Hq@Zfq?Xz<7)Llkb=|03~&ImA4D0q;s|R2>HHvpz_3)AK}b9l z>JoaAf3a2xjatDl>U!!x`d)0XgafG25`Y13AzGe49MKj-$*d*p43WJ{eSRmy5%NAR z$ov+K2sG$IDY43Ra|kKL=oL}=O^IkbemfWT@F)UL?QfCd==_0Mgpe9r+|^jZ&g{MK zDC=!Fg2xS71T(_B!)hq45R3W2-9rsK3R~Cj?R`Ml)1$nZ4Qs<(is4}X{cC7g4uCHZ z%p#JQhJdEtHu?a_0!f-&NT2QF27PofyX*yUq}J;QP};6FbtML)sQE2LpoXv8dU*+- z!&fmqkvo0V<(~9`;xCH8LNW(xYv_=!ABhmE)Db1r++=}~ga9soP*`su2M6DvZHDm% zh7@=nhx5FXE*x#AuI&)aAMWXiQUC(&D9gmfkP@VEDbXPpT8(K zL8YRce=fZ^u&Ic#d-KPw2cTxYzTS07oZB#*HpkVN;9BIiYt|WR;1r`B(f}C?;J3oQ zus}UwTl%8u--P32Y?kYs>6lDU?T5o%t^C@7?Y>rQ?Z_P~KJIn0t@dEOwI`gw)Z42p@qDReY_N6)- zw3v(-vG#;)4H*5Kvl+yT@B^5vU0Bg7Ns>)h$GZ6bClDLY<{$C;jSm}pYCBC4yJ=WQ}Y=T zP8Kn)pWDCJBH8CS9#K~r4UY7gPOa~mTU zK+8`uYsNP)i@AkFHiq4 zW>VA566ct`*g+Q+Y#5%DQABvX^?^yJ6D0Eg;3@2X9{*#0WM*bciTj2`1?6IA{cpWN zk)~|SJ_lU)OpSx$BT)Eo{D*FCD8tj-lTxk_gKdkZ1s*x_j^vV?o4e9OPYN3K$XMYf zy;qthig+B_>+J&BoPbPzi~HEW>rh^L=NB20q6w*D)5K&Mvhc{Wr>or&b%~m{AMd)p z-4(%>pCNg#^&jy(HM1ndlVX$niX(h+ZL!N2YHmV0r_hTq6;56{+W!=2B*F`u1LLUh7sr-NWC!hfiV2UHzc#V63|Yekwsc&|iX3usT@}iAJ8w%01E{tu_53%UPhb#q`mtOVd-#^R&n`Q& zoE8^8P0A$Y_1c;YtjdMxT378=74o(CZiE|9{)(*F_6W&GmDu>%Yzk~wy!sBLpZSW9 zDv&kKnufGBy+io8?`i)oUY^+ivyxElwia;6QUdr~T*AOFnnS5Hzqu+2Ut5==NO5-U zWyqX7b;2>klj3J2Ypn)vU>tsF>`-xoK%0%IvhD&Qa;JtDw;hNw_1iN7n^?Eg_||InfF4ok0 z1CPzt?cH{MN`U>9Ni=-}RQmLlO?Zoav!Z{m%6>rvXKT1lw)FPO4If zergYLq}5y3qRay&6jsom!x{;r=iJ0RPWOfD0n|^sva3p3{h-umB719z!1E5HG!KGH zK@#<7z%=u~Lbj_u9as{Ikbet#dBQ~&K}r2p0To?vF5-U?@KIg{LXes-vL49?7=rzYah?GB9eLW>H4>a>HYw1d7PEMdWeWSczM%HU*|>cRdnH2@=v>vLUz{nzyK6A~P#}Dr{sDrlDP-ERQa)ArQy4a{&V0SV zY2(K4b8OpPuMrU0a5ksjmzp~beN0h^AGi7fivqX=(ZlcnQrPEjD!vwMT>z!)SzyqU zl(H%JYQxs&{vR@BWAhh0!pa(!Nqe8~bsfhI0TrR@GCY$*Zq=m7U&=5^_qLy223B$c zL@L{c0fo-TiSTpzV;*er)<-Jdm%8WgXu z=#ku#D=m&=fJmUrJ3ny_+~LtJLCB@otQ&s_kxV)M}5r3KVT9$=UgmeRl zU0z~JZ5D+WAr`43gh1zg^Y&)$9c4<6p}Hf(wIq4dTIlj~^@jv86KqL9ZI!Jf>^6(& z+7RnuJ)lSzgcRaXgPc+ZBf zh}t5hMxOwMf+$+MbfYhnA0+*d?gcB5;aYhUnyS-nkLYi-bwYQDEnxz~M8cil8kNH; z#{&0B7soEVVgkd`L(IR8rR^mZI2eFY#@4xE$5#^uWdvRY0!TwRF1S@&$|1m|;G6|S zV|65?F?PMZdJ7vku!1^MLfH7YDfc3K$va2osnF&*duF(VaGK-X0}wBO5D~QZBQvHQ z4h%qDN-k+`@TpQPDWvkiGj4YPxwkBMypq@ zK4*9rw`HhRs3-8~1N_J1!=U??y5M>c{LHL5-!sDa@~V=*4;R&|PuH9;zeDTGzC~6S z)9R&1AC5EW-$Be(5mS#wSHHl0(SK<{bQ)*TSa^_^(J{JpZf$(RLyJ=R{=aVff152g zQ;HxAEK|y>$4`{J0wfe12s1~Zhh>Y0C9o1AFvou#w)`9hgM+Xz{nz?5MZgo77SN=r zW53ag{=*ifwv77GN3>h}`;L-~QTS3REuP4#f;EP)l8T{F=9`RYKJ+|qluoRyHfz$s zd3bP>A;M{7J4m)nrSXzG+B@Fd(2G5#QJn)r)X>BiWe+hW zt_56gud=${a6ewB?2KC0J%3s?pyo=T)JG#&jo`TBeSiU`f<^B~QCs^XwD4_tH^0f_ zRtRkCmXD_i1GIO`H*k%{Y=%p2aLGM#!7tMwVZSq-6?1xD;X|xD$^xQUhNkk|5REP= z34Wq4p`R?*x(q%SBPn^K>iwOga9s^1gdgjtu{D>|y=>kJmSquTuYG(0xGDa&Z`Lo^ zHQ%n^DVyWdmTDntx;L(_isqEHfwoqOg6z$#`!|sjW}KJ5u*Ao#VmhJJ3O!;J`R$M) z_sbkN%9~g@|IF5M?+d+$u*P`AW4qAaxiGEFy^4DsEJ=5KMKuzhXb+D%{w%jZJ$VU- zqo(68wi@$8Hc+QS*7ZsOtXOQC$XNX8zEBK`2|L&77|mur`=dcWZl`fDEsSzDb5yRt zt6p@|+*RoJBj-di2hUrOsn6;h4_C&%Sq?qYJfdY9#*1Gny*6Srccfw?OG%~}8>5X$ zC-M6{pIxU)Zf7Y`zjj#t@U_8Zrd!MteUoPeJE5xBpw-BYjM!)aeYfD30lz{wMRnbu zft?(Z5;}C$g4E(KuAXaOINd>_8n|F53+w=2H}+SV#61_;>n%}GKFgbJ>{kjn274Xp zHZJZ&GPIco9s9m2%u>RQ7vn*@Jaf%`>0W}T4QUI6oO7}@QN z-yXQx&xoCV2tD!u*tKJJ`ado_=8RgdfiAJl&ZFf_ivJj>6PVPnJ5I9mDOCHcP)|mH zgY4GEb#1*LHmQstN%3rmo>15h3~^$4iuC`|FtQA@!BwylB3+FWaGt(C9FZvDtK zakw_I5z}|KI&EW#x0tVUSoU60m-%*>s;BID_xSq(_9u$LQL;0?4;n5>L1_BK$@s>Ews zNGD=tn27QKu`Dc@1ZnqDK!}`%Or%BvOuO=?hkXdYU>%p0!RWsk)JJDwTzN0+M6N%3 zB|;DKc`99Z4$vEGFbri?(9MFCg7$mkh`ID9)p`=GN=7q~y6v$iBih(7JV@z~gYr$s zYa`G33aCJbNt25=Xpc7r8Xky7ZLtd@BFdGdu+U@ytD096s4BmS^oBW80UL2y5uWB$ z8cKn&R=tzp^OTP0M=~Hm6W2777qfWKr)|9hLiS(}`&O`T;V=^-Lf`y>ICRtrL_;}f zD1s67TinKiH6R-@6|GYJg-BHbFt=eVM~mvl-$i>ga{@3M;hW?97iiu4;uMs;(R9!# zEY3>+P?Hl22zO61nX=w;Zg_-~ws*M(Qj}>nkgHlILBcGLGic*|6CDa_<(ugU zJql(`63EdAazpATnQ7l#keFpHi0m}D7?{-n=tvHzP_vn3Kjeom{|wS~P-s}hlR39~ zM|cb93Z~Qw6{2KeIEv5HNa)sCQCjlnY;`pN9DZqZ=0LG{Rf;o#C3q4xUsIF9Rqvf@ z$MKboBN#Zd!Y_Ix4WPjxOK@>vWAT@h2Cho(>{e4RRyGR}2aBVi=#JaoupwMDVC1YbE%!~| zvJj{L1is?yLm+pfxi{lFL{h^`&v2q*PK2Hr^R%VoH*PRW{%P$UWu(p2Od^>ZB|}S} zJZQuxM!+2zo{Ujar$sam!L$%>_@_4mk_*T&?d)?+{OggJ-l$}mMQHxUTA>oRA=30D z!_e{w$u#LKr8IW?v~`OLn68a9J%JE@n;u$IC`+flOKxdLXj;i#Sj*`${UgHsmCZy% zsM$c3SP#l+#cHU$j}Y52>0?ij-b~!UXFt=*rR@Qt;9wyVAw4G*e$r_MX3bRw_$&x! z!HKv$;@a;KI26&K8*f1K>%Vx{hqto-; z6slDsef#8`QFmQ?oiG3tan4LQb;ApR(=|r^Bs_}h9*7jxk{e*}soY?V@Okz(r@V8p z7swH&L2Hc*AB5F^8O5`IY?K@U{EN<{-`v!NIGp46p^V5JT9JU33H&h()yae)Hiy{q zRu+v0VX5HOqi{kHNQsX1iHm8(zYaXhJlj{51i30t!Lo7k5=o79$WL}yow~#tG0InX z{CNsNeEG4M+7tM`UvL_|gl;c>&A?~#q?V+iK1;R?Ev&h1o&`eF8PCQ8YK#rznSK0! z2~GLGm2&y(qN^?Nw%p;?Cta9}2tS zGY(La8PYd(|0NRN(5`M!cnZwO0?#$A&yc2a|bzQ zEN$z@d(L6yH?Hds@mSRWc+S7MN3M4x9WU?W zCFej=+R?PiMNAcBiAl9Mqee^7e zNa}Ys*={JtQ6VjMG~)dQ;Dtt_jxJiIqqV?bM;+T?3b`q4DsS!yDfAZfFIJ#`&10JD zPe=rC|IPWE!!*m@l@(37sz}iJSJ}gv;}~ocx_%V~$3wx-22=GsV{1oUm>P4v&#TAXQ&`VKy0xxq&nhqb#tZ zF6@g)c7X1ncd)0&5O_o`N-?ZIoko-%h&v*vq5h2r)nDtf#7IkZ8Dj++CTxsA-b_p~ z?|w?DDzGzRf140MCRmK==-3W?67g?Gh1ULM>^rn!5fi1G=~&fXohcn6NK{c^duPkw zPna`-Wrbakr{GC)IHz2cw%&F0>OkKiT`1EqsYrqT>(>7100#*^-8LQ%npny*H%*)( z5`oy1od*z4=y5}*NvLjP-w;9}Zko)hSe>0rE8?4Q8kT;*z$!Y!6S)f#d9sEBct7~D z*I>|XVi)*W|KG}ZAapn{5K@Og51|fIAUi7}AF=$ZkW@6EP`-r752I#WtbwpNi&|W| zB{*;?IqEGo|G?K+?23wUnz@7)wd6{hiW_;UNIUzOH`Fr&xDwFEnNvDrZ#-21mrnM= zLW#%bV@VF6x_Ggt(}VZ&Ezqbtb2?=?5pMZ$WZ`_*j5$sJH)|!qhk!3@#|q(B0c#8- zHeFN5J;DU0W~i=qpMY%l!t>dq!I8H78H|egpZl-RS5Bo^DC!Q1V^Mk6$LM zwc9Y@ccI;5`}uR|P{ye-`B=_#&y(aQ@E;=58Tm)BL~BPp_0LsLwhb(pX0|i8yK0OD zZzt!rpwX|E)sLL7)vYZV;cuJzO!)h+!teW+(Qd&m9fq1}!0CHS!(m5;EGM?WLnvlS zL&R&s{_^M8*IYtMPV#lgYeL1)4JX<+T(bedmS?p(V660`wt~IcaMheLSf@6qDm$Hs zdUR_oDEO?=7U(C7u*lMH@npQ1hc|%wVp84?m!@HqUAxx!li5~P>8gDY(ZP7W4pr4l*HOA&)v*tN zcDho$n5x&$hKc@@m#1M%veMn`^YJpLS2Z4ri($Vk3u$&{+usS%Fz^4Qq5_-D&l6w) zn=B)=KT3~vBz;WEX~fODU1$c5)8K>}5;sulT_Jc$6Jy(7DBh`1>XRZHQTJuBf5Wkn4b}C(k3w$RZ4{#V?Q7G9kmz zs$SF@7Q)i4)~c7}NeX%`D6j8r?ZC{^wxkP!&Tu8Mm?<+7A0e3@tD=3<77InJz5F>i?pBjFq%YmNEo z)l<5DxxlmLCp7`jj9;MCMs5Y0Z0uqr+JQBI6^db*2R6l4L@^FAzh=@56U5PlOnQ{w z%$tv0yDO%7Z9kt#$YWV;8=(hDVJ5f*`vf@~$T1Kx2A�R-VR3IR-C*8weS93$m9ta$<0_Np^@GW5>`1;x;!!-!BDDzD** zdQ!&u^hnwAVSUMw&{{lDR!fvt(@+-N^lg((&$qJH?!=l`(Yqn5xfAkv;X|Jjyx0`n zy;i@&5VL1A`Thi)UhOhh6fm5gbw+2qDXXVQ>$fN;9ZKtQ%q<{&^8R6acb&_yRYr5X zO_^$@t7uK%q=>*rnOJ!>z5EOx5%4xOIimrvlBDJX$%yhvJrrqLKwgq#DIY#rEkD!hDSPh0%%+7TgG z$7d~_yEv8}Q1fdF@7g~EX3vdt%B`2c&sVkuFtLGrH`*`ZFNn{T1xW%EErg&z4;EauP#r z)bsGi^Ve~JzVOd>@YshaoC0Me__rtD6KzGj06KK>VxnXjNBRqk>$5!%HL-Mgy6}20Q~G0*v%x<&q4LYydB6+^DNuy&u0fVXD@cFg)tA z$$0gyO~Ak0swTN=)vOtNt(BHrsQ728Pe7Nj@iE4FxZFr}7+C`mO-AAk0rA;RmoA#d z0EJ-7$dJnEVC2Yzr6#|2K9t9b``6`z=A)5$r4|&on|Urn+J3I;Tueg*rxYQO;i_ps zHy0n!-JO55ObO@B6?pg>>?71)nJJ~biy2a(*^DKoE~NjvkkHx>T53x$jb6z>Schh!6tlBTkO z+pZl73Tp8n??w^u$za>i;+EPYJsBmC)cOEmS6u|RcKd!=#+hglk1_KBPgcjj_w)M8 z-mChX&?zHUcSdE?l@W`h*(RCT4vWajAbXO(VnA33DN#kx3z)d&0Q(c{Lt#@ymjCn2 zmR2}|=ELmL8BylSB5;k_E2RPgE`aEXQ0JH7hT_iTw=l;r>n@f0h1~wq@J)6I!xaNS zYd)}z#y~Foh#K7op~D+MQ92Fwk0_RMqRY_rJxcQhMu&HwssvH}RmA`&O}*ak@CX(!2a`l9cMyViiusn`9ADghpa&WxKHB? z`#rPG9vKTCWSSDay^7Ely7byFn+QaF`d$hk#$%0AM)7_Ly9rqh@%2hz4$qAaV zio0*(@Mgc0@s2A1>=^Fk-!!VwdIYSPLw;zS!N>{FQqm3IM9Ls;*=^=Cz0n-jhU~j# z>_p{s_WjPI@``jnPBU>i{{GQx%Ohl^}gMh#5O zBqp&wpmZ#Bc26pDA2|#}zO6cGuYeY&7S`(rKS3`{1kSLn*hz9!3;tl{77g7&e~Tjr zwBSd1GE=?e#3YYDdpHe%a^_c{XEUX;3cf`!sC7J|KHLDPj=RChK;qE6;Mm+3QNJU- z*dtR}E*~_rdUTX1^ZwpgbwgeiBvW5dMt@i53j>hXjN71$dy=o^;ws~*0I!Z{P((%h zCVqMry6k7KR)HVa)};9B>a>oh@ut?h4yy=tP3}JBxqS}L+ahzDBk_+1%Un_L00fz%G+!RQ-GIVzkZSS} z17ZFOUqpSq)S1ZH-ga%rN$wCRqC5yZcB*@rsbl!!2(Mcv1+PK^SISqxJ5>sCT00F& z-UoY;LF92NZtxI*WuQ<$puzuaDLCP+3C+$gj0{F@BMP~dFKfZ=_#=2jCd}Rg1nV8Y z@~|=;m;{!&Pb$_KH1Cw+&qkwpO=g6;Un2(?f;(pb6}oG9eCbt-PBHnNhXclhd^;b2fmky6RjG)Juf8bZW6L6=Ll3P;Lh$}RGfH61-4M;)kTg91 z&FZtX=)FL)K!UJx|F2nqoIWqQ{d|Av-%CE}mh~yI1V?GSQoN18*M`n0Da>j@=-qqcaifjl@(ZQ)}%jwEr zgYE2$lo*tkY`Hr_F}E)Ww8abD6~-#JdQ+S}toMa$sF53sY@=F&3IN@&ufPQX-|1eq z<%vI>1k8C&h;-yp^39dz4wpfyJNXT9jlnl3np|S*>~<zwD1|thJj~cbH3;d|7@RtkveO|jBX#Sakb(Zy|9_u?^+k^bet zCoAD3jBbQw?AffDRrdFWN^iZe_-DA>?0coSWiJ~L8tGXr{q&lPj3y0t&D`CXG9t)m zVcBq>CmPr>xB+C2hNM0mURn7r-$F)^Iu+Zh(?!Yq)U^B9hc!gY)hbTi)Genf#_gJ) z<=4sr3XPSg?_+kapS88j5(w&_uy&b7Hg4^YY_jYkUI$1Y6Y^RmJ9vj@L*)E|FGPWF z<5+J%x_lLG{&>zAH#}FBmf&G#AAhoJ3dL}a_`~!>0s(0F>b#+=(obTP%?P3hPY@cp zjJx3UE!I`5^W0#A1oA?DZK|q{TI`&!FP1$7`d8XGxc)TR&p$2TZfP0vTnh$C7#k`5 z*Xst&z$VRNxCmOSO&xj~Yq=EC7O`MAEvcj3Ld7{3m}G^z3mlr(_KUC!$U%7a6`L2S zX>rzv7NC|Zu8D5rD*2hsl_ug`DudF>KA{;k0at91(U5)NM_k1)b%FBF^h689O@`NQ$2jV?% z#b2?5?qSNw$Zv+kU|hU^%oNq&+?dmF{g#rL_khU#h5TD_*%QAVGo9CJY~}Vig&ydG zecvf3mb@*wQeg466X^?7l2q(BkocTNTpfjK_(~_lv)8^m6*AB9{*CNNov(RW#Pb)M z%Dip5cDZuV*#~4=+l7E)dvAo8o=AdAc-OE{ou-`;DsqP~^UMMX#Dj4UF#F4kmLr*n zK7hKsWFL3hxfYYLcGXpb#^O`N^28Dg`?LPk7(S`y?aM2E;*2=DsW`+8-lyNIdb`gA z%$fF$uiCbcTmP~d+GL-fpCLR{e|gs3U&VBl!BSkFsy=%UCbwpdYV%zungPflv z(S7YdX8Io+R@lq}?yiV<74abH>6$U`^=>&i!(K65^&npfEKxHI&{#8Y0o#-R_%k`q-2i%6 zZy6y9M3aX}x-oF$8JV4PJZ$SHgx~zep2-xZ>A{KU7t6`;QTsTlTJRmw$ZT0rJORJb z)TA8pY&CJLWG!@RS02-bZq=OHwj{l(ep%_&8>AVEZ}P1)=h?)OZP&J~HQj*P&dk$L zgU`8z4SgZ?IZjokxHW4KRc>Fx83DYr=)v!@Nja%5O5Yg}+jz0ZxqxVQ}4z zDZgbovJQ>(e=j_T;=3!9gIt}xKf|1JuZ)pExTUZUZJ^xtAR9CS3n4AcbVsRfcMAKz z2=qu)t%=i9hH6bZErIZ?L3x^vDJe6+dFk)XvhiYmZW%u{k}@vSUdSUy7R8{Ws@#1>yJj(Fg4xlnvbp$c z*+jylPUj}$U|$%%4f|>ksxGd0>vqoA>lX%4?njsu|JZRC&35Zm9Zs>Zu)Fzj!MXN1 z%Y>bO-xD7R&={3?IWR=>`YHXNi< zv%4EgRasa0Nq|S{#oBX6wTv}~XD4^)EF!M=BfP9?{(ACrFb#T00hQ8pUeftIc#@3aW5$T-B zM))Dqi>0V3lP4CxKL*GJcWXo^GokFH-(L^QmAk0NUgTo0I>U?C_Sb4V3?xgp*qgQJ zq;lO|WclJ0z~hC{Dx;)WO;(MTj^32vZSoP<2-whL*sfSeB&RE(0(m7&=dN@o9P~pD zenx#Wy5K)*_|j&_F6Zu;#VAok-37qyIILDPg5oqzJciMbGQG0cnlWK|7Zow|3iht19OGyLPGr%^iRRG=t9Eo zfdF&0{G1oTro?_jqJgq=v;VgRTVuj)odap}n6@!!y3CLdny6z%dV|6>efTWjvfYnd zQ$#e5hS2Ww8J~|ToF&1F6l4I~kn4F`)TB8rZuJp1J||@Y6Vo0$P5EzBqD^yW4mF2r zqGA6(o1ece`ZIj$ou148eqQG;IdS|tzS|tq(Vm3k0H+Or?vogBbdalM05lGQV_o;Y zSXoz0A>M;if^-cy_RSsJZkn%N2VQ7&G{|FY?YstH`zPO0&)MgqF`~P?sSLxbyZAJm z?=TI=(ceDvn*|Sc=92EmMCrV%9iP*9s%%p#KFM!#8~75-T^*wp>3M-GnkgOQ;9jng zb;0(JG`b=FxhzkTm8?G71Y}ye9Mc-u288a6GY9=0fE+uRoOWHuiDc7m@gPAGF(g4x zm1zA_-Wxg&L1+U_9E;Oc@qSbbozq_=!s&W?6rZnVutw3CWps2vUgscwYOB~N60_&jyqygo>)C5RV7vhSSVa`l0N!T79gU&52V>b-I; zqrhVXQg;F^%yq;VWlb6EngG@x&qrbb%3kT4l8y3WIyuE00P)>((?Q3DR<|3U>ke<7 zWLuVoe{a(Is?qSAr|awO7yF(G()4vqLw;jLHNXE40Chl$zpifG&c_5DviY_Szym3| z$0S(Jl|oHasoKHciCh*MNI{45g?Mf*8zUB!5D|$W&xs(|#pUfXWH=#)4arMSl2@7n zvN^N)XCM0jA25f-QFMiWA1ewyg^!Yj-XcjmaueXlU63Iw#e9vUUeLse?A-&xMz%J`-v68xZU=>%)pZZG zu&B-y(nVPT~Z4M>w zpB9xdG5+pt?BmU23V+)*{i^E&r>DEdi3^-EETGDa{CF}KnA5qy z~P{4U`$}9irivt^EHm5ObK9GDnem@eGL)meBQjralz4h>l?fK z8o;cp*r&qloBthIg$j*xe07*vnKsXuSrtUUoLRp)@DgeugAwn^=n{kzb1F!?ZBzG~ zZQ86xbJ9cx1HU+<>DDKyQjpBzg2MdM`b{p%MJQ;{9n*5C6lv$Ht-tdS>aK3uB(fXp ztF}v<4jNy7TBR(6#lu`erEGlN=S$bVb+-M+NOkmI6eM<}(MMeM4fITcV77w8Gz)E1 zqI&OFZQ4WBMnt5T8%Qt${3`M&a}x_NlpBATwXZ_IhRl7O$QlEaz;Tw=&z_|hp!=+X z%%~C9Y_Hq7#oDIvd&PJBQKPkn6Z^S%-BHGK#W`?;l+JZ2QsJvXc5X` zT1P^`HkYmNwodv$p7Yi$ppposPAW7E<0-(UORcz^;t`nf187%Irm;`4Mm~7=&s3e% z6f=HA80jc2rZ9)9pye6nLTj-A(9HAxL;=Y3!me7Jk}`y7*SH5P?+7V$ME>yqe+JpW zl%6dF(}~K_H#SmwvwZO%2fCgxUYVY)x37P(`~R+(G04@~>mR-O@3Kwp zLC#jrF3$go2n3j0*#iHtcmA_xR-nIR3Zja#Qj%)wjPmb;2VzvPfA0>6$;HFvFY7|CO3s zfy~VRblrc<)qz<9Waa1zl$H2D+jkNC-!coJ3xEXxbOZoBOf8xJr21=8{)k!sh~Hb_ z>+N9g05CVUbq4xcnFHT1@ZQeGZa{#GlPl2I`=5&cM)0iM05dC7m-ium-ze~ZMVAGc z+XHz2C4Oh}Z`J=TKh0m8l=gjxn%RSFJppDwb9jGdMSGX`VW9c{XOI0)E-6=ATSa3# zAkBX_^nb04?W}A)|J&|=OtgT1sM7pDoUNRttUQ2b%2qC>mjCMNzvQwm#_vNf3bL>T zzOTq%B8@-$$o73;-wz$DKW7$zk(Gn%e{}D&XlerjIy(cn{^bUI@4^4@eQ*39wg6@Y z6={D}Em8Xa8JE9|BtfS3W>z2z02?P4z}U&j*b|=ReFoS#IRW0R@1tx6^!RH?0L)Av zdzW_?fP<@xFTmX13I5MQa&ZEfMgNHYg?IqW5`QCJ0JG$O5H|~eS?X`Z24I%?8?ghJ zW&cL+Q5641?@^TggLvMfDF2P#{nY+O8~{gV^}o?OGmZZsUTy%h@!#k@gUR3MJ%j0g z5a+wj)ZX^J>;IOpvb-lX`v+tNFa!U#=Xytf4d~wn@5Pz_4cXq~n_KIHQASM2}ZRmy*ECx4xmEPrpa|GUrr zh1Fb~>}`NrR%Y*?G5%qqVC>>#WnqO8>)KOw8WHn~{Tu z3&6<6%L!oRV&Q(Th@IW{zkE&qb<6wfn0-GF{*C|KZva4`2hbFLb;;g;lt08ey)~@T zPqKKX3W}1K>AVJ8NLwxuay4tFn*dj$V3!OiToTwGkWE!$FE7KV@0S_~D$}M6!Lohb z>PV@bzB5x1*)#T2@FPH$6rIf1WYP%AR#*wB>LsH+k;^aADd(8W?o8|?253y4it~2& zJ;kzpdqcZPBh#zu+$e;9axYzAZNqf3MfX@i#qS|lndt#{d3(Z)k1-w;U8S2hDhk7% zl!JB{=tI?|PcKq8Y_3~6HiM4&JU~cbr$z%p)bI%We2@>1|9b8$7m;x#mgrxxcm|WaU8@d&;x|%PB~zJ zjjj}E{AD8M1janltKS&Y3_H41e8gN#tgnPX&@Doc1ogzHTIH^nV`@SkxrW<%H^2#z zea*>OIEFEZuTBep>Zi=Q;NxOITG^+0aAm5{pM2ZgobYTyR02;tv~*ixiehkhgP&O{ zsBOCUr<*M0U&Kdfi-rJP*kAEh1|;rNBPObcn^V$@@|j#>ed0B_S0|Q#Xm!JTO(%AG zz(Feb&=d{zH|4(Y5I0BDpcOYpa(&ENiH172ruTdl?UKxYEOM)uCwOBc+`LDl|1mgl z0344j$y@huE-J`F0B{~3?43|N*`zlo2VIfU&*>r)E?=`XFJ*bogf28SKU4>1E~}3W zIv3*oeyM!I*hSl%{+5bVOE?<@y<7@4d;g=SYiQJpA^WI6p)9;R4cq=z24!=g_Hmmn zW^$(imHNAXj@GH@@Q?m`S<%A+XmNgzSzk2=NJzaYvZoHaK*01qVq4iu{3==UJWsw@ zQNEIK#X@;dq38!4&Y7ik`Z{f0^{EJO8i(ify_Z@76);G>UZ*yupt4;%=0PgW{Vip3 zf&T8X(vSHTuA!z>4#{I)VYR>6jf9&j+{PTn5u|i~)oyMU1JPf0vN}p&@(>e3^;9VG z8OTIh&j}1z5dC>Vp!Y{5Qxf*?zjU^+p0lbFnSR2=;<-+mC#E==i@4)UkAxG?^}@0q zzxDY*rrDL5=8=BDwcuB5{=1tY5vomGLlXT3+;0((oa<-DRc?jp`F(KM4)T`l~@X1v6NZ0`?lBsVY!RCn-IDUP5;TzQ2%vp)@3LKwX~xI5D_!7fLII_DbU~)7Wh1 z=dX_V*b2=>Bi&1f6Gg1w0R)HNmJse4S~WfFLC0oMKs8YnH~0?u(^vEt?aZ6-)rWM5 zxH5GyaXYpJ)aSvx^%H;|H<}@M&s)ox?G2^`vcmLD%6B}h;{4g3<=q-E;ala_(&e^) zv*cv*RuOY41F&kPw7x2asF=tF#k#_hPp#((C9@q?2`mqfO7Vif}6 zo`$eyPCYB?*i42<$*n7e0GDm=rLWw#aUqyFmp}$!a;hZ0i-s&CQ0;d#h&mK<&El5p zx`YSuHp#K(wza5Pc;uE%Ot5KeU1i?)$wK|o=LGq9sP$yne!RPcqm)X~>_gXoQ7{)m zsw!2;Mo(L?LHe0_QoQ78pcJ-*D3zZ&F;ruH#Z*DFsz^?Zn}L#i|BbqWpZho%JEj=S zhN?Y9Ls*|{{I7~uk8K=lS;9)6RTtI*n53X3q-*T(v8IXexzneQ4O>2G)<8n^#-lylptXa7{D@R| zFQU*H+JWBGq1TXMW2S0w&6X>!@>qc9HylDjC5wYfwBk zr(fA<*gv&;pm9S>E*3`ukU@5ZzSO2WJ1p3;VrzyD3VZq)f+Sl}{Q6{BqD9bTc0~j> z$hc5VTH&jgjKML8$-ho*oE{Rqd8;6PmOchfx$QTyas5=Js)`7;|BT9DZUs8A~X0xcN!SE2E2!03vFRV&VA3K6;)~ z<--bjyNx5mV*J>fz!Br;L6}E|sUYRMPRK>1Y&);UpryyU>O;mj;s^zoWSo&N`_6&* zUu7D0Y?W<)dOtsgnc9HYY8_989oktox%=-qRAVr1u9@}U>JW*RbOY4YB1j4TRm>Mgdn-XZe?GNY^lve zFIdyCw2KUXe%g>GUXqC_3!M=Y?~?*kS<_iGI%oUZYIq_Wgq;QtR~&9*C$tXisEK4D zjo{$deD*>DOSU5JAL25hva8T(r1&s!T^|X4c4XuQA2vch1Ud|0I3f5%{hGeBn0gd` zVybhQOFp*aPy{a~DL^WjXxZ>g#RUg72>62DK>SH5sx1>Ta_<)Msm3-_AM=;k9i`h(Q7kAYl1~TpR~#iip};wPpj1E+P*b8I0;I z4n>`$pBi49ep7*4$C==8}%8yCiD?5$xx9lUNIX z1HfQ_x4?JAenZ>B6)(bIHKZWq*(Wx4phknyDg9$y(^nNeRDsL3dKK0^l+Gi>=_aZ? zd7i?R5Y8SBGk#PipDXk8)jw$QI=f+*-_gu zz%|q@6}_>vlq-N*u-+K=jJ*!R;HttSd+sRfOgYC6U~IpL!Ae7r^t^_yV)ev>u*Zwu zY{Qjq^S|qi958rno14b(Z`U!SG<25Qh`AVpm|ygUyNdt$lPT^~0zzowcXkkWpMpZp z*$}J2ngGLsvi_`V1y_H6OQxuvX^7+AkLGSp0F_>93=~g#b!nHU^2@~cFrT}A<`*NW z@uUO6+uF3m1u{-8W{Vm<1i(RVT&J}v2;Ln;6MeZ1{!74bIH4Ic&no!3jBcV$M$PI= z@NNNgL2#KGP*5GQRf6Ed_TYt%uu*61{%qr+D0O9asvUt;QsZHNy&w6shj(73YSc7z zSTj1_F8Iq+i|3b*=8D&3UvZ_fd2@qI32_1yyiLh~W!N1cYfiRr?Zk^=K}-pBzJABrwr-4;ZR3sv28q$Ni>{hdshCZl* z$Y78?MLDL|6GVp7>It)3f^~3m!R~2pJS%+Dd;xWYW|@$RG^6bP@BpQ!q)*ea8#j5f z05Y!@+CL9A8RPx_T|zHNU&abckK^;r-a8{L>Q`JJ)ZPYv+4m4OJE6yU`Lo3{Ri?Nq zNFE!Co~t0OC%T=m6qeB~tsi7v{JJ_e(iVBK@YfX``e-o(c2k70Bp1CO zjLsfXw(I1CUXdxRYo1qXu1rQm0%A~9PX%341oE~}LHFB`F<{&VIEWRbpLQ%@S>397 zHZjU7T+F-lVDrp4{X+OkNao5nAS9~mGqIlc*MNn8Mb`)D7*$$P(aA&L6|8lUtWIrn zZ=|ysw3o}1O;0ymB0{pBbHzY=ln^(qb8a;L3@DQ?(7Ls=79^34+EeNcp+M3s;S%1wyk%BObmW3 zlaw%j!ASz)(FwB+WBTe)JRw&R+P0HWZzGY!r};1%1{l;lGmWXel*Z}NcnWo@=`)WH z1tnB&29!s(lse{nqnB8@>No-!UZV-TRg%O;{laHxQ3t4{CBXug9vA`z;cZUgLTS!z zQq|{6`J6PoJ-IZHqQ-Vf$-hmz+wEbPQ6m$7N;At%&gALRV2Ez#NBFj$lW1$$K-|sR zhmR;Usq8y_Rdc~{=Cdmf(_Bt*Opde>_2Foo9ap;(WBt8Mu7tDpYBwv`wZf8 zW2Tu;wHGwMG#NoIYM_XHq3n^iy5LTXe^>w)_pg#9lSSR1 zgrHWxg51Y$hO#~|x>nJ4I9r0T6F_)o+Y~gjl5F!?(o1ok!XiHokLbr|&{%GNEmjpI zwWo0D)z-|QUrmiWRu<95mc>xytYoC6RxWZ(@3nMSf`y}nf!O^Xm4$mRslOnutC zi%{8l(pEz%A*8c}bVl%ORPf_}qcjOPk?_KpK`CB3G;gkhn>RE4nvwP*DRBFl9Ht@h zAxF<12FWj&a=dEn%IvlIhgJ&wo5E zhte~`Q9-zdGV)F6;SSe9kMzEWC_f_W1kQhlSizb{l5#dsbd+Dcl;tV_Cxt?D4Gu_p{I! z7|c_gv!RXYiQ&*^VBx1%^5VpVa{`T3QJ<{D&H7`N?<{a1`)_3m#&!HU+oWX8*X$Go zEDC=vq2$VYV8(6WWLDm-+CM4@sPdG3ZFY_0n$AW2d?R1PSHkXpERa@mx=U;}8PDtG zHnJiM^ZiX#*u)f*p7EmLPPj`iu*;66P91*1oH;}*G&{Ca1*lB<3!|xs zQ_yfM%)E5Z&0MNmmRn59Fo<>{LNitN#Z?2>GPcG3!PMj>kC1k!SlqtBhJGNVbBYAP2IJtWv!c3`8%vXD{!L zEjv=7=+OxfzLlVFDpLrllz%Y2jBQtU375x*x&d*E{<@5RJ0V;>{BW}7K|Vzrqu8(x zy~mto6?h?F3o7~Ejc!cQIPzTiR3QRuO^zuVpdBiWk+;D1%)>t4yWM0fRcpPok1ug&QhI zF2fg2pS6L1b^7JvG^P-zL>}~KLZ8^WZkpvSy%sN)8$KGH1PZt*c{_{UfDHr0d%Pnl z5k|_%%vU_g-y7LqMlfnpt-i6wkq+P=YgZ#IIuGu*251?DkR!N*vJ3i%=hSr}rGE8X z``BIPSA;+X>U~mAe}Mz55GkU^I$Rqc8@6O|GPb9GJp#ml1&x4*N+R0aPl&0{u@TAY z!T46~``N_TW6FgQt2J$pCdUC2m{`DbL1qR;Z35E>{e4W-1vnoCiV&;OHP6nFvh}M8 zebXQUGq#syVw0oLvWmnS+-JksWELNFekVx9J=D}TVyeuQ_c_Lz zQp@@NFh1_Z4@??W;*f!!mjsS!6gQu6K<_P-+M#^*CVUf&uThKI1j zJxiMf%nHduLNMK>GHP8j-9*Z8hnS0B~$CZbb#Yl}M(=rHs%6x54O3)Kt7{fZKCV+3D(@3>{DnO@OL+3=I z>%@^t515+z*a}XGY*@XB7%lhY7uS1t0ahfoj&gXe+$?d_8y|OCrn2#^pKa_nv(S0P zYZZc&(92Eoq$A{B^Mf#E6Wb3_Z1{1@f-BM`0@l7&gH&%tFe;Rlg{ert`y3+?M9T`-5_2Fjn#7q-jIWCM~r@_Z6 z9-vw|DhkQ@CWxkFaXKY#X~6BkEHAa;jX>YwM)t9Ed|_KZ?PEkzftTt`_M_Mv`S*Ct z-27>{J%(~Kg*tG5D6aaYWk#VM0ri&wXhIq(S~Nzx#eDe|6_QSpf=SH=WpW=u$Em2; z->P{#l=MTj9DXQ~O?d2caAw*tS|ZSPEt8)eM0ZKGi+jp3SO-@xy@BmyzF)3jqUIx% z?Z)Q(&XffMnQsXQkgrFk#Gqttj*PQotRC8g3RNg3u&M8V@`9<4G^j&a@jq7g?AAEE zs=t4qd(&I26e7N7I#p=|pKthh=dQ&-E#~EKGDp&H9#NglCBk56%;xt?x7@kDQb3o? zlyMH~GfY@{*2_9Mna@-%NbDC;IvO(HVkhf|$y>C6UXGclsKOaUnVhz$BUP?+@IW)A zOkK#l%k~6+i74w1M@#Wren;_DQE`5>%GWb}xvb4VCH+e^)eMrwg^5D6DugP-i(WG} zCw_7cv0tIg<~zHPa+h8iV{`@Ej4*AMS#2cOGz3T%3-eRldY(o?zd*1o1R^P9EhJz@ zsl>?!5Vn*dKF-;o(`6Fk%ASawtFhX}=n8$MR@zyAIcBO4E6W+lnTZMd1#CitOJtnz zr$q?m(Sx~>mO;L@-cP9?)S*X3)VEf;dC<9r{ltw76)kPo+?=a&tH+w_cf=8iXwf)<1?(J*Kw$kI zz`AaKX*X+87@12ET@6gGt_=|N<~A`O32nV?@Fpf7v=Q?h@^z?#wGB0#Q`0F#38}>& zO>iDn+hgx60ZzLOEZb|R0Y@D!fQPaH75Cc}N9FV)zel^MN`aDeJmHpg|jWfr=ChNXS;D4P3YN)Q?S;OwIh1)pZxA=#G>ydf z5X5l91r)th;%t`RZ)xOVEJ-_+_T@6IKKV}q-dCrKOe%ZCpd5@&9Bsg%Eb1DeZMkcvQt*~}81n4|cPRz%*Yp-+9K z>&wyWb|j<-^jPt%wKY<$0{sY=^O`(G$Ji(XG6>HQ&yqI{6o|B<><=bhSN%MHfZ#6* zqBe;{M}e9`RQ;hm-+4CUo+wy>qEJQ(HpVB|jb2~9fhNB-V_pUA-xA@Gex&K!GIM=? zjL+|8wKUO*)1@TG@#>5b>MkoOueLa{%^}ScBo;KAk@+^}_`Sutgj94M?`Wq$?es^Y z99E78s+(tCiZX^&j-WMZR_%a)&M5l>3c$1BS>TXF=d(DEym^T+Bzl;Y=af^!Y+m4ZJiRd8+SA?Z+%(QsiZ zv^LZYz-yyE3lC%zvO=)YnCRp5e^G72w&@L0|7woS_6_UHl3#gL+wli~>_#gx!CGz; zoiRM$83LH(4P4Tvr@8kS;q zc06FF|3)yh|Nc8{gqRiMAsT1+Tj|-DZOYPmK*s0G^^C;Q(42-3N8SOIW}KpZr4HLG zcf%wu0tF?t_lL@=S(@>G=H<>8RvHn@iE?oL#70ZOct%5;-Y_nkHMIhVv26U6SOZ8& zyq}J_1$T+>ohx7B&l9$$14`L}Qci)x!SMo<8?Q!Tn;BJfBoJ*_25vLs$gE8{-%k(O zC@4XqVW;=PkWZre{)E$kix_CuOnJ6Tb!Ac0iZXR2`wDY>i{g%dsH%MKG9IN9=?iYK zJ4{&n1y%tNG@W~g3y|L}5?6h9 z@&grLFv3y30W8IT=X%tQfb`&M)=Bw|?`HisrsA(CtlI7cwwi7nO>*cO$Y|8hYdbQ47MN|%iY-KaP4lPe+Z!HnlM6#x<^?TaYsd&s56k_$?VqsWyr z4@&^j@A_0!3~Tb^!pnSiZL=}w5>pk&eh-pKt-s(Y4`5q=0+;V%u&DMd$Q?(wI*c8< zMI7+)1K=34deufGwjkE}HUJ43(mtf=j(hUCb%(hQGpJ)wBK(}$Yo#5v^QVO5H>-3( zm1aRRLD<`B7tnztGPJ4*yTANpj^?>v$AYb!_PA#g)L5TY!kt$5jF=4$c?TAf&y_ey zyN4-jQbJ09ZN7r)=G~-HzWaYvSeXnAWoGxh9)lw2DgD||x%L^|z;m(Iw;C!n&q_q3 z?Ri!Mk_mJOuj1a^r!eJS;c@0pb~Dity!5IMjhV`3DPl1eOKiP!O)$I7Nt@8yAO%um zjswIR_#2XHH*MwN>{799k|7)Qu_&1dfZ9mu#E>t4t)|Q}(IeqOc}GyAf$4-fWrkao zetpm2&GmcX5oXxzr^YkVMGyxwvqEO2K@5BDwq!%8SxBp;%mGmr?pd7}p~Wktw|L}Z zj`;jZtOC6L@PN3f=iU>p8Cn5wNR?0|M4{*z1k`{2FbP=Y)f>QpbZhXQD$z4T)E35G z131us!2eoK;*cchaB9)1QNDEMFxjx(Zkt&GLn#Fx=>j9U^`<2Uzot;=(`E4NXgajI zyPT*$CJra$?J%;h{{316e>JU{K%Z^?%kewLAS*$3kof|!JQaaX`zN|va`Oe!+bVLT z&hanV=)&=8@>ni;R=v^CRB6iyXh&4t4H`s$!^Ir&d9*Hp++Sx0yu!cP<|G4^~hXz>XQW@H%Ql5431{koSP7RJdWo2#mpw_Y?BLanra+#B9a2vd4f zW-(=yq#rL5M6bf9HTd}x!DRn%GPxHl(+$Q&L-&gy$0|QIw(##gZlsX2tz?HsMS$;r zMDG}Dt&zw}c7Rf$RG!j}4P- zo`F(4)H87bo<*0=EOPG|cxRykePed;rIN*21jD>X_L7DcVohk>zAIF&%j!7D8>hz# zflsU53|if%{#lk`6W2l$ACAKVwrQt-mFmW2Sn-dsTi6;-dzVucz0=8y<%}L?z9H5+ zyY2|N5ykrDp6$A;A(d7*s~+ZwJ4i6vR@lJ`;jTW$JW8&U!KEY(|8Qit%g{sQLSWX_ zc7FkD#uuzS<=y@u9AAmlDaeX$$^u^>qCgpIhlOzdT%9$h8poZ~KJNX*WX*ejFtu%Z z3A<>$;6jplVOW?!nkGVpw-A7?icMyhW^IwthY(=D{&{ee#d>k1Z4Si>5iGB3~Ky|PGH-4F7eV4lPWNxv1U z)naI1nNVxftGp^>M<>owHqjEcHXwN2x?Ffv9qda}a^@Y2ipOw?<&5uRGjf0~ z#;Cx+y6(&Qzh(5Ws>8(^&c#W|ARw*gOL@A}06G-LUSs64$&&qKCX@qz7{(eEPm-j) z*N#IJQfRnUWkajTo+-ML3IlMRBu`A_Tah_1DIiUoyhCcA9EGaI2idKz=&tZ6Q&q&C zUaTsq=fAYFR2I#X`TwY`!zi**fbU4q_;gqfH53Z`2GP%nBESn<7U7&#O%#|~Q5Gj= zgo`n7+yyP|ITj0h`)hT7OGOCD4zoA|4D%OiP7-JNqNWXf9Y4JO$C9#`_!NuN1y|^~ z`xa_7VzKSflk%2}BI7WvAo3C9W{(MgiJa2{j{f3BcUt&sL3ZFE~c- zEV8s}NekU2sdsLnZ=|;-Djn-xK)lOnK6ZZ71;v|Ob~&?AJhN|q_t3@CfqCKAFsEvn zxKXb)64F`{3e@eZc_+Q==-cN*94ecBPz*m%HsC`QA>{dRvZ(rQIx17}d1~aAG~A+^ ze^#N990k^WRKj8n<|PbXbN`HFi_03CIFOW-<5Hk~`YUXeS*75Y3Z#t6E>-?3h|piK zW4;nXSpMma*p63!Cvr%yasHrHc?E?NN2pzRj{Mx)WUvZo8iD!A{}AIRiinBBbU0hW zNf!SibcI&1+e1{b!#ehAZ!;yM#Tf2jiH@@CKw!KzNw%%2)6edqv3p|Fw*83(&#Ch& z6tRPDOU%6(u2fppq_3A+#Xw6086gug^NfhzQj*H#j4)(>LP`dlOD!9}CGF}*v4l5D zen)lrVheQd`}87E3~TUYP*&S(dw+I!gy{^EPuI{>Lp2h*Huo3CZs}^69~T+7(JzGw z-myPrA=$3L)LkZrBOKilG~_=nAn;=#XWvvx-^gmeEt{(aU(u`J`J6Am-IO0wW@wIn+eJ>c@nbV#ge={iyB@yQiFvg6K@pc;Nn(luw z(-z*sWX8V?5H2-B9-{#%YC}wJ6aHcA7JG-1GeGs&t-3IN5{?_ra9_qrvq(ddHu0W2Rw)%W?J329qffwU&e?P5Y+|Xm;2IIxqwY!% zFIw~sNJYnHiRiCP{1x{{d1&|!Wi~pL4CA_0J*&+HT2BdsSI~}kr07@FEh0H+fkUVX zipj#3f-o)CfykAr`hlejWnHMiBpCbTB;T)pNcXGpnv;ZS+9Qt?uyqVC!>$3VS4c=A zOVE5yPL=bbz5N3u-%DrX$(oKw zUFILD`3T1W+62}O-XLxiKK6vnn}yNGqlo!JqbCwrTc>uJpMWtMMI|K|Yj`tP`G_ii zqj7vwfocqKY|XpiiuuKi5O{eBB07BXghma}#}k2fHoX|g=FhHw{`_)R3I~3R^MNPi4R-?fA=N(6DexZemaN$H z6FG0=wbqWqA+e4BM@PgTLfo4SKU(0?z1S?SV+;~or%Yt!C|(GM99jr*_J6{}rW?Du za+So7)gGv4fCuTP3q*nP4jUY#a@alQRUo9)mdMS85e1G*6W3Dq*+^W6>B7c;5XblH z;_XA8fkJikH3D-DI?Ms2e$P`nyLZC)WV)L1QSNxOMw|CvOlaa1Jy60h$6Rt!pxN}o z&C6rStCNl^eHeDT=dhb~{n1y-cNg32G+zS3W7Ux;h0!|^jZNss1ReN6EkRa^W_K~wh}YGe8J4T4skl+__DYD z@DqAiirvOBMa>vOszib?8(xaPZh;9e@0Mvn$@C15N)0~8>zF8M_%&2SA1QAkyF?4w8W2gkx z<0zv}rfH2xzeFyqM+aEDnm>=1HBOu|UtT|fvB)!Dmb1Jre(AJ`>B`??Tq+rrOlr$l zAs(slK%jP>v)^7poxDz8)H@VLPX)I~*0wrF_UvFH1!sngmNqIIJY<5|>B7_;6gn3S z6Ka%_4?iI4hYNl(>&;Mqp-^(`#|psL0i1E{Kg9H%a6mSu^^-@Uli6tzzNYIPZX_-| zwgtdLZB4QuE>{f2y359qN}Bf5=T>8uphFLFBIG_3ofDQiXq!bew^7xIrXug&1!;VG z5V@Mo4ppb!R4{z)6gHmMJt|-koBJ`Y*dbLd7N=s;XqE)_tmh7YCP+3^cuXI@!N8z) zW$wc~Pn?ePYA{f$e3i$%Re#5+J+O;t+K$c7!!i-ME%J1MPQNoFm`u~Gd)*9|ae!+H zU(2xIT!`E$__zbHj&w=^;o==iZJ{s8y@|`ET}x^bb)aVz6|&7aBYZM)Rv? zertvtR(-?79Lx_>V)yN&NBXsIPiDT6T|0<22tUe1Zq{ntSL=Wum-gBD8nLxS^&TKN zOrehhtO|r1F)vtrui#?y_J`1}GDQo+(J8|~=)3KX30sMODn~Wa#1WP$VwO!>;!0aC zvd70hu13&PQZbno?s(e$uxqqCAxV9BAK0z%AsPkr8^;U*D_2j$Bt-#J_9hSXHD4z& z`h&ja8s_brS@N07uAKe!iDc2I6$U5VdS3Z?NS)e|ux<$Mg&w4bWCn;udale8)2ouw z6I=)p%c;SCPvkrMsTx6)=P~OhMLJmrhRvT9+vU5$ie|>=-E^SCy^XlezVBJk>(Vv0 znY1dp32d{*?J@_jOe|eXknsqp7A&n*i5xwsI_-`!o4;Em5p>?o*J&=?;1EuN|sm#AFaNDsT zm8oJhl)>|75Fk@$wv;Siq)I|^69%pcL;)T?GFVWF4T{)k;={-Uw8$|0QP93R2Iza} zuuo$rjncn{is5K1Ai0O(MLh`{Wdr_g`tN(&IYKy3K3rcPYB z;Da5cf~kjN0y-pE4Pw><%H~7l(iM@^M0r7q>u!ON$$PBE&`FY%Xxz_M`*kE#q5~@U zxYpZU5GI?e+^(AO>Y`G(Xr7FX#HNig8-H%lNL8dKDssaa%3y#=hSQCP&yTJr`qW< zY03M02P2+dP9OD-y9j&dAZ^|WpQ>;bYy+qhV64i{DB7EUehCg z#40o_ni`Qgxw}Dl2r|__-5@6I zYMqAGEHsw-K|n7s2F*b*LSU>^VN>gWt5tEZQeLA*kRQ2tR4T|`J(ngKNegb1e-?6= z3^qL|RaS^7CS*F8mLLrfib|9}IyN&y=TUc1hjjyO&q7%R^hiFf~tMv5-*Fy;#whnH0XuFLEgCF92zp*)j@ytFKT>>tp6 zNU~kU)?7Q0TrFGNi(gmp`y|Of7i#)oh1=lD%QH5z5STj!`fI78B7bL#07nHl$J+^= zMxZDPm67XA0=yZr#V(rJ-R0ukCGMIuT81sA)ELKUnCc$3(=uGnSEq>K=*NgO=g#Cz z=cxj+X+>ej_teFQqZZ)!Y&Va8u*veweM~z>_s#U8AwG52p8lQwN%u0*Hb988S)$@KV5)BLMoFd@5`C^ zseM!8N8T6@S(H)l#W!?^BN!!5CN(I`fzeW?Fx20u+qx9?Forx!5_RI!VcH&@DPbi_ z5tV!%y-j#57Wg{X1{SY>XZ}5Gq4$v$pTd-KDVaVS8aVH(P3mNBjeIDjF0s+%)Ts}b z+Y$O9TS~eRsaUZueN=W^NOCg1X62F-ayS;5)WD|ao6c8y&Cfb&267K1;JsU8-Q=M9 zJY1qa80LPWi&5lFq-QhDCfvxXAyXcZET1FQ94M3jF!qK)w7!LZjQZ)L(^OWWHV;w? ztZ4rYLTEU}m1`dMpbnr{NP36}K^J8MeFTBR+II{wvu*v?8Zme~j_jpu@wAxTWEIi6 zQ(H>#mL+H}MoGZ7Tie4!C)Ll`w46-DIaRt}&VWrYb(OGD4$qbo>2Wc#$e>-(3?XJN zS94Cb!!ioP`w{Gaa0U*tIz1EWm--69)>n?*$lPY7c=hFB`^AE#2>VUyTtnJ3!N&^s zZ?kDg=7tJB0Lh;iL2o*Vt%rCmFwI(FlLCe4o7q1-3`D6Edw|rKPsJk$a?;Yc2xHl$ z(K8x@tXVdEMT=GE6>&TQV1Xk^W_Sz(LZwtvq;0HV>)jK7CZmXrzr}q|H6+cfP9&-o z065Gf8`j@;#}u6r2_qkyqVw60*1+>|EG5B9emA^^kr%dNPTyP27$-v>9U`v7k~Juh zRA`Op(pAR~Wc68m?6tXJY@!o*^OL+b%ofmNbik5>2ux@*NWo?ekH*D0qVyOXWdSe2 zT1t-_N9@{vRqli`skUQOS#xZw-q;b$=9%u=j4^Eme$kPWJ5@%~<9cos@mgP>_dllW z=27c(#>f|=s#-7l7@iIBotH`X{R@NUeJ~k+L>u@-W zSYq7xJ-F8As`QU(rv$wAFGwEqWbV53;Q9iUfkz%aw@^kKDuK+U*@zh?81F5r9(F)_ z-DA_eiGa)xtYn!Bq&kg3=UQ*w<#dG_23Rpg=z1NJVYy&J7YMEnxg@Uk%ZVVh7UKv& zK4XXHIf{@a*?=D zMqt9((m49{P2SaKI=J3elk!gpdJ}H`-1*SIN^{|Q)ad=6U71S7T}6|5=EJGuazN}I z!d=Q%21})ZUpA!U0_V`}^QJ$mL%Mzy)C-?jA$FJkyhSZbJ?cfFXplHm|L>^`8Fcli zMC>N`66Cf~dn5O7S8pF3cE5+ju&0pzCH7xyTP#b8_S@9G&8y@l244=2vSrF=D-z;} zt8uofxKx(B2E89E+ZRIWLsZ!p4B+PvTe|i= zgBj@r?%z$Fv;?D@#Y3zL%NX&xdV0+Q%F~AI1^*nmw^H{?>PBrU6kN<4VM64&R5p|B;+Pjy{Ufl%suId)gM zt8Yc;0$e<=POOgZ!H6o9o4}Nrof8TV=&7QCqp&04MaUCeNX5ut7VI?)*p0B3&KSgFSIQC{QL&`(PglWaN1IPaDVZeTG^V(?1<=@-a%j;0R- z?}@!6_aj|$bHWmNao%eLtXS{lafDQzGS|n`;QbNNL3;X+XQ-eraX?}ddLJ!0v%(X4 zB=MIdx-B7?vYRXh^W2Jthii6z>CASqg^ZC;JXgl?WifyfpVr8kJmy`okOVp$(>^(V zy`lPEy}*B{EeF@@vN06tgeKXO(O3qV{W|*XgWi|zSS$~0fMvEualNN=mOAy!DUn+- zEojkIFXJf`4ByK-8NgaNyO8f>HiY1sF*6rLW?OZm;5%xnbw|EW)eL;6m_YlPfwbTw>=^a!7Gxl zV14-qD}Aki?66C%e2VE|DQSPYd&hhlwZXG6uM&QyLg5#y%$j*ag9i;l^V7cOXF>gVZe#g`? zWuWMuORb?6Y=CUC5{yO3FvF*@E!1q;p`#)S5#oG*4VAuXeJ@Pj`iZ5iTC$9Y#{a5U z-cnwDG9n@i>5A0wcSv+^*t8)EE zN#dWV2$h=TVrg=CN{HSPtg%-xg%cjVr!C9YdS`E6Om~b3a~>DjARX9fdAdePbVm!k zLg$0`=IS|JWU{v?S>H#}Y-}=(T&T7F8D4N6m)`~gl`aW)5XT!_eIH4gNAJ$ZQa>kbaM(Pk!HTXt zETqC87URJB>1HD2=|@C!ioH#H*MbcZ##^3N$uV2z!WN3C>S)e}0S}7BUUIfdmRo{Tb7UlZ7K@;+QNB zWCNIAX<++nqELcx=Cw*Pm(ST0;=!8=6+vQlaqGa#^@Y-@4u5o&l7}B3fALX-NRo;# zzL;-uE~8n%J6*fRr@=%b2bPe9*JgJ&Fg0Bx7p~Ep`mi!a|0C`Dp-^Re>Q@Hh$a(S2 zPnudB1ma+S9Z32Emq4?!nhcQ5Y53c@&XCo=fgrOC!xVu=>d}6#FQzms+im{4J<~YK z6`L=a>H>L|!nczJ2s*HM)pst{Xd@EKw~jkM3}Ejt)Zu6Cr%W}dCx7B2oqQl0Kc}3!1CSo^5 zlk!sMz9=~#{lwEDg6+PR_SUvj14KHQ+L?`|g>r*pxOh$J@5zO0u*kY?no?G(Q-nEY zvK*8y(uT{z7?P37wucktK5tl9$uY@WdonI5Z|eC;SE9dEt@UqK>7S&F20(Tk9Vf56 zdl*g%doGH(ztXbxo{7}MoAillZHuz6=&Nm4ebQZZ=YkCrx@;xkKMGe$Ea(|(3MGua zl4@$)4PflPX)9ySNNS^RF#EP{Zt&g9E>hfUCdtdvFEb&vcq(ahBF|nKVHx7S47`0u z#_UtiP8D)D+R7(2@qtCh10E%Wt&H3YnQV1Yo)^yoq*(N{m)`zH7#iED!WRF9CvHa;w@NQDZ?8J!R*{QY z`VXPD`T2UZ4R%@eurv0I)ca={^k%)@4AS3!kttp#OGIn6;v=I9-UuOts(YL2#&PI= z6vnY?Ws23O0XzFe75~R4%8K(Y6iC3?=ArAW)BUqgxTG;bvELzcD9-XV@k1o}AG17d zeNzIqsO8EKbHIn%ap6c$qK5S|TDvYLzd*nTxv$Y~k4riL++d zQ8JmK_Hgrc&0FuFP99z=_n91goUqj7kC6ZooMzjn)XL(r4+SAZ{#NTnZ$}VC_B}?b z7Z{Ru6tY~pe(tPMIs&=CXMq1wLMCRhIq3(Wsjwt!kZ>KS_JE{N35y*q zOAt!2tLs$vat3)xYkud>#bs}c`*Nb-{R(GYB@1sbL4#Y8OaaMve;-dk4{5-1gZ!ci(x zuM)^*{MbPgd)$otX#2E8q#KL&?aS~WyQ$R&txj=j0kal)Egm?AUpzizK-}uUxq_-@ zcLN($v;mvGt9Q~t>ghISnCGcxYPWt0HeY|?^OHZeQAI&9hi^?TN;;CKXqNqQ=<+KoO#e?~D(XKMG}4iH_nD2?AD z4V3>>m0%gx;lMW%X1+GOo2K)KWUVhR0cPwau(L@(Ov6)bbQ!$3OVZQ)ZT)jxGJ=um z@MFRld>`Xx*QJ^ExxXBZ8e<(UkW1W&c#d>)Lt^-J$udO>pvW%R@))yOeF8nrCAQyO zBk#lFIxEFwAG9QEDuzR|U5W0KhS_*YnPO~Fvh(AGv`$w!0^=%XeC+d=k`^F&Yf^`?c5ET5*@^R~_ z6Lj%6_fQTS*q;EN^gL48AgTKSh@dyHkI3SL54WvKmV8%JYm>fX-L}x98HrBmM}etn zXtodjLGJcPuW|($@9&_(KEb7b?`n#O;vYeJpppVjj`QWHKwr5Mnex4)BNWLYHTcwA zi&yMx{MJx8&LBFe`dWM-Yrfc^MX4kB1kTft<<$P`JOg+%_u%7+mmSQ{gDqY?eq0D+ zxgv_LOK?*^b4ifdUB<(=a*i3TJVfwoJWOJV`r~Ff6($Ut4JP)tGT;jvy^!Y$b+~hddM-73DwJwqmq4G|PB_!X4S7vkGFiN5jr55>Lt1uWFnVEv%N=PyZ zO+`d3S~q}j44=YbmpQE3`Ji1iJQStSEUtS&e1iEw)mWUeksXw{qrZ6EGe)i726c|y zk-&-p_bvc)xiNGUHZwb`j?w(yI9$YD-lb_wEFA*@wQgaxSK6KYD;x*uZ)}8oAQz#J z=+e^YP!I#}gmqQU&hXJFJjFzzvRSgaWkOVb6c%vbV3fP(w4K6Ss6ULM#5&-)q@2Pn z63;;j*$l}(>N@SH`eRq^Ks2go+$?ltV3K_IBOo{%0>SUN`wL?4uYoaP3|D`{X`ehE&DO>HVFpV-96lDBIIbjA=tH@yjZasA zHYz53rk;&om33wS`{3XE)a*oyr-A_}d=^o0YmdJ<%l5gKw;G26ma#%~K$%o;^iR@J zzqQ{yPfi%g#l6O}0S+K!Wjv9E#zQ2Ip9PYaTC+QivPQKKp7#pz<`b1$X`st04Tab( zH(kH+{sOpAGu8W7Gq76uVP{+_>*ACEi&sp{7Q5mk^EDR*W7&I8ejJU;{ux!H3LE#_ zMW^cx%1!_|K}x`@G+I*$vTWH%$gBwtaI455(tV9QdM%g4|IK$AkFp>-*b4@uS7s zSw`bH23sG4rBV65Y1Mjj(#}{cBflxb%>*w`c?K znk6^N|S7qtKfbU&w5w$TNpP@GOt`n?Rq19S>BVu#(tQW?n$;YQ$-jK#!hNg zMkL+wCE;IiS}qJSicksAu2kjE&j$K^Z9*$Q|>w1QB`4U8%Ea=+aBZ?bNa*6dWg6jG&o(WMl$rtepaRZQA`mwiBogB3d zUAu=<1;R$jWEDO~up@lI*zP;KPwC*pL+qNKk5@UCf6awrr9y!0I-0nC1RXUFm0uYK z;K_&I(-|KIBYd48x85|RN4S@6=m>*KKTR%E?rB8`sRl!gMlqI>lU2pyi^A-cc(%mD zS~^bGkPao$7c-Uh*Y?x+3+rj}p;or|$%BJjv0lNqd}W^f4(x4F!Wy00bJ5Y|Abj+b z2Brq!gD?U09Yr9`u?4px%Vt`3G7#pL=~st2dPfk{VRIohg816ac`zgkH*q+~yH}?Q zC%Vt?ziqYaEu1v@8%Qg?8r7HxKeUg=-$xb$5Wp2b6e6pqoF`xk-6mpF4k)5!G5g>5 z)2xFd+!5>dLZ9EFrs;a@hd4WlVr7FCJPpVp?dn;?wXlG*bm)W`3aZLL8^45uB>Ix z9ad)=F%fJxo_KYKA5YnDx#8E#Jeik@)HPUgFjTErt(!aVzcx^-0NWKp|AAGw&N?U# z_Wj`H7p4{SD|LFE87HdIGET`>_gbvJoa>Fp@ocCdM);{N6Z~9rA(Gm|6&lIPxm+_v zraAz>@!zlJyG81$&=3rHpt9f0kf&h%pz2seHJJQX3lmbj#=cw#UsyjG2>bAhXLI8g zUvWq@_gEi0rM~P?e>4j+W=-*rxy{pF_@NsIUdvTP?%7@EbweJQe#nQyd5saD^b71| z=|sjD4e}+V27L?8F?|1d=Th)L-e|HPylKE*z<#bpD8@s?$L)Ra(!Mqr=O6*;^(D1SidA0>az|8Dqoo&Vd_M zByd}$4{aqt$jPx-N}RiNk{g=ino+>Fab|H}4N631!!QPfW!aqKM054Er8A?Ps&T~7 zV0GP@^k|-JOr37y$2hL6*4q1@C^bNNqnjb^OJRfc{amR#i&a<1EZHtJQn^WJh{y#u znlZUeV*kSU^#Z1uLxRjJ*VdKgR#o?Q4+|g zZ70Ihj7#+IUg>1+?zf@CLYO+$b%{585;EGb5pYMFURZbwk&~ke)BPkBgG#_c+KmeF zfV>`?MDTk4oSrD!uLl|ac<(4?!}ZBZI9=5^vxOd{yBC#OD&ndmsFAhQ56|3jH5EOB zBmgI0BTCgMn8y9|!F7y3YcY>00>ls9&3!rbF{)Ix0aOL+99b=PyHQBEiw4pvL`sTx3`V zD>sn~GY!(3{*I=1C;13-KJ{a2N@~*``>dDrPZhxrcZAP`{70D;gFe;|cQ6Gn#I3|b zEgs$Lll0s!xxd_#=XmFArNQeQ1bxkqnU#AlCKE-y7b;i<$SCxIT{pm(utE}{hw z_6;^!f##F{FMfAt2AE2P)Se3Z$mobg`I`J9?ce9|$a}p5lDg9{zjD8iWHAk?Q^5I{ z3ukad323#s55Ehn{c*Tm*fD1c+d6sVV&f=zEB~nDJnJE${fax!oC4uaO79oiflx7j z*DX}rzD>X0>&cazIFbQ`Tw}~@o)y)$7L2U6>!9lopziOnV!oOXK{%>W3u-_1$5g(A zUd9O;R)>lSgiX--Ug>cCWpg+6|g)(ZQnO3L|anhvJq#KceO z!ez$3Qa0#g;85uBrnZO_{A0r>u{!`f^~grW=TvNzuUj#l3ZRx z5)Y_^ZbWsFH>v?G`O;yEMQbrDsped$L?s>`nOTe8x`>lAGq(fs4w`s)7;E+xA|C%{ zf0_j7nah=N^uqKsjF4Ta5J(d4G~JtH+YVO7(g|eT21P{Fws%&7BnkE07BQZM+uvO} zf~7fA`j-C;kUs_{>+uz0er3jQ4L;Ru6F79XUa@t@0U|bngK^SLCRXff;)Ss zPf-5dbOPvY)d$EV!+y}^gr(uct^$+wf$r4_$w@)HBfR6@dSII9EOt{*F9B`O+YWXi}>FALtNJK4;St z1Azi@y$_+PFNw31OsJ!h@@2LJv35cTc5;rP@@e6XeKh)w0be^@YiQOroe}56lhqgY zn3`4Yx!{RwguHO#RSA^K-0W&>k+xe2_E2ibGk5?~I@4&mzDI5w9Q5mBbWC>Jbd&e! zhPQTth=pOvp95OIWgdHm-hOwkb#B3ZlnZz}tn{~mnAyLwYl!lroP+`oScFN`<-nS7 z8t%y(54W}k3&auF`eWv*?C8!DEcfk`7ix&L8@o!Ez9nEGz+u1>yo@x8i}56~R4Q`G z1q(=(>>KCGs2qZ@4c%QZ`v)&NShFTO$oS-R@%Z5>A~vD(AmJ+nxE_)-n28rq!EoI; zx!OZj(=Sm;oe#oyT9L?o7d%UQun9e*x&IC6l3WssT<)8!qZU!?_l6;X z9!uea3{|KBY_@BH(JV8$m&Ca-FyUMV9-`VJ_5D-7LWvizd^SFBN-mBaaGi~VTg%L( z@jtuTvuGRAJ3eGmU5^ylIpNgfk|rq5eExM5lC+X5uGb3UKtgo^g z^gVF%J1C8{J=wQia8=bMARCTE6P*E<^03ccFbniyTtdT@G%T5oR2Xpva`~|OfgEoq z&NDSlGK+r=1$mN?r!SoXp{V-?9;{1qPut3U_BevMrjcS6HG^rGpIxIfTFXSkiAQUp z?FsNTT)E8t3X1*P-y8}p7r8M#g>8p)kfk8cBJ?ScuCb7aPoL08FlL-uXt)6){W&Ic+z?AHPQ7v&khwZGxFm33B zpuwWs5jOtCc+&ft3eXd4#_6#zH0F`}hD)>K%y`F%m0GHQD1?;E*-8SLB+~*Apr!Su z57hV?y4Q7cDrBuXoFe#@0!|y)ueK%9B_}GOi0Li+bOXpY8#P(a_NA5S278LnN!bLL z8@2yN{t!J-D$vOdVV>to|s5Y83#3wkO=j)dxhKY@=i@cyl2+%%Y5@sIxc(bD$`SGLhQkO{Om^ z!n~uwzxAIeDL=frK4|3yaNrn|ZqrdK(*%f1Fu8H0cGmg_M4rD<8q=><7R-}#HKu9aQ{=jA_E8`rM>A297+~xf?CLwK$en08L-I`Y=@$HV2e{aIUPCb38P z6GVn!pOy>dpPa;Q5GEXVh=e3yiJ!tOkGc_bL4_Zjm(zfW0%SBOruTV!SR2Ajx zjP20`8ZMBoKk`hv8F9&YM#=SB!zqYqzvVS`3V|2a=-;X>kya!W-ZIl7urLs`Qzs&# zCcQRf@`akD$rGby2Pmdn?AZIS)&<_&NlPZp5dG1~51(>j0t)9=J1)p;((F=+o|ouf zM8|Snn%{sL3t`@o`2!f7)YBwavQmkGX3k<#qROs1N^cfl&Wt}YYw@pCLNA|%Ilzt{ z7?V3#c^<9;V%fsKf#5xCd4{%}nskiX*&PDPaju^bu2O5c7pOHKPoJXr(tuPEg9u*! zl$Hu|j+#JlMizUQ!A`kHf=&2Seknv0RO_Q!s3)+-u~kQMq3qS}rTc+v)$i86TC5o< z`r;E|KroaFGI`9J8wJLlxXeMHI!MHp|7f2?Oq;1VAL13cG(GK_xwuj8x}m7NdQ$t_ z0N#(mX~2v~QHs*+)3kD`f;mv*eRED=qCQrVPVEbV2O+P5UVL_!T93I~Xx7Ld|Kbe4 zVgMMjE=!MM7MBBkY{RwL2UinIbkn{t*j4Syb3ddsLAXUdE^Phd{UrHUhbUD&nDGS< zcSvX4YO6I$b5G%*+$#E|eD{*-AQXWUUknpk`!@Pd@sx--WbrC;c}0|+dx#akl{Z&o zZemCAAZg(0pGkBlEzhg?YK1yzp&QYeT2UbWw)<%@KS%|>t_D(W5&2rDy*gU~c9C%H z&JYw`lEBS|-Oqe@Xf+fdgg%DwpPHtYZG~JeDyT(3V#S!3WVZvMA^(nLn~%uwdNrHQ zC9P7yoBy**Woz2)^@y@c zy(U(@G0w3-J?8z3v4l?t_U50^uJ}?(%_G{g${HcYAz)DJSNno;=9HH!>nAA#TH>TO zNjXAEjpXb<@UpRm6wy+&V%TL)t49Frgg6NyZi;2%w^scPh#{T2E4=Q=0$A5RP#YS?$9amboq%>D& zPOC5sUd8x~w4Q6?=21c?s6p=4qU9$SUf6Kajf7JFA~aKE#RH&Wtf?cC;ua7r+2gGI z!R{OMYj^9dudQyFrs8QL@HpML{NZ2I6vX!J$Xcd!5h8PZGP4R;eNpM6kHqlNQ;@>MRN&C?L6vgY~9_x=^&$k*5cB| z;E5D2j}vqzwbJxbu(OD!o;{H17)UJ{_Rn@-FVJ2~4=Lz0pN#*mGsxv>fzC;j!?*en z!}Lw>letq8KagE{O-P@XrsmdYnY}(XwYo+53XdLF$1K;A9`RHEcSF^)YKmn($xC~^ z8mV5X(*4j#FMF*aG2=!oQ|g)7kCeHT7*Se{Ba5BIzjD2zuB5rEB`81{^+VPt0+W@p zY_{YyCoFYNj@WPPm~R9XD#WYL-hqo-PXr6t>>@W}D1%UGy$ZJ~=P=$)_|lN!Nb+n9~!r-lBC{3yF zt~hV@J6+pqgCtsuI#qPZ(zx>jB1y`DDcOmoL$Ky9thHpM;$lM?ldWDC<${QJO(s=89 zH@B8vboo}nYuS9>-~R%T``>*hrw6T&Mk;|)4k>0!3INUP#;)Y-2k$eE=K^!MMJ5Zv6bs?&wp8Fwe=+L#N(a&Jq5ucX|f`5u@& zj0U#8=LF(ZOi;nky4PCMxWg0<2YqBjxK{Ad9a6De(X(T)3; zQDeG|_L}`ZO&s@D4MvpJ6halA(_SQ|#qXyLu)(0VGqoLBVS~m%uOLG$EM^!BtGowt zj23L1*FKUV(&YUDty=E$JafuwYs=4Xp82+qKpp|={Yd`0nAqM~YO77D^pFJV`xc7r z^f>B!4)Wna4j%9amiCqvSC!)6UQ9IigrstBb8}|CD+Eo~%@Zcy)!0WFCcf%NW0KLV0b;C4Q>tJfVJh{ST`+0M|YoXjODgeQu=ySu(2z z8R5)co)dy4wxF%cirLYt0R>S($b81uOK;`!%QXmMVy>%F%H@u2%i0+D+U|Itu&H0) zFC)!}CJNJ!!7FLQjQi__mp?>Jx+PjL%`q=&tXeqU3`3(2TfuXw22%S;Tk(uQ1Wrl> zJ@Pumb=GA^Y=QWYtwEcMprSP%nDp<&m}L<8s#!p-%NLA32WvFqP5lpyH|#*KMJ3+b z^km9UR6TFVm>-|*ShnJK7dOMbz1|aXN8Vk6ofUNUiCV)AM2E5Gcuzv@&0DXXc7MrX zELlMh)_XHnHu<3ujc%UIr;XD9l)xPAH~DyR2T!9=jKVba7dt8(c!w9H@*IoSM(Js6 z$fW@ZNmgm~UeO-*xyzbs8seJYUL+B@fBUD@%$VG=SjF|RBId`B7cr9MJ*Xr4i zpV}H_Ks+Y3G1y zuVA!naoZ)k>7P5+f3Z8m#nR6(`QebVjp^GrZZHN5QA2U?I$r`|^%>z%+dC-Z5`+&; z`+s<#J@dYb=h!Qs9NnIURpU@<8_N`bcachzn7>dxj;amlJ3#|FO^ zy(;w$fG_82o|u%4xmMVWQ1c46#OzU4O*AZQ{mA~ijXw|*OCbX zPp|*3`lq#lZ+}|=nfI~bvPr8nMd%*8cfOr{Vpo=5!@14B@b|%{T;cC*ZeBC;2|JCi z*v7n{F}e!g4M|?2rS(d7(Y;uG4~5Xx!&gOa%OmXk30SIB*&L~dp`x-qiPlfagZy}$ z2ShGRCEmtxp7dnb;ur7zTipi9Aqq~Zr~)0-Ql%n+P)T+dB&f!_R82PJw_((|p zrSDP(f58WV{9nV_d};}ORn)nt!u$BlTIbF#*!ttoLg+9}NfUF1AVs9_SkIMk-RWm| zsJ(lG-|0oPPrWZ0;`n6Q?9i3xj#+-D?0iQ^vhJmZ33PaGj~dX_zBdfuy^k8YQzhf} z$fkCd9q>_;-X@tBe+ICwPe5?s=mdYyn>3~4cn+rnc)w(T3b6o8a{Q=@rc@hT}(^2 z$i5^6GA(rqR$IT!5RQkaxi7YduTl?leMd~75v&t_Xo6@Q#SRJ)b3(%gEXxk!xbZ*W zazUXF@}yPMdd-+mIfHp?ZyNCi+AYLS<5SB}w!gw6_QxEUqM#DZTDw%Xk*{87Kw<6_ zLRYj@0E>AglzE6sLA8M5iKMWR?>XeUD#2f1wq7=(eShj`D|fcU`7^@^ zTz`Y_lnWwj-ZD>KYnI}ORMGwUuKx7*7{nHKcd5@(ZoPp!xzjPnqQ6e2k;W7BSVXQR zp=NAs`Cl%Mr9lWB%Obv7e8qng9S?8NY$WGUSVr;(`Vp9SeK@6RvWse&nmAd^QR2ro zZJm4)A&~Q4J5%wOZsvj$7&hYZ zt~cIjfZ^DYPk6y_S-zFPZhV@A#N2EW)Eoh#m>AM>fTR1!`(${Ofx{z!((! zYVvZT(1&TkdS=_JMJw)=M9mLFZ@6oXrR`1Aijd%Ydg2nbw#@@}L95NA_*}3I+&wa_ zCpfC=2M{bdP`==JC@cCCaZea+U&Sf8A zU`I!XzSHZ4V5zqca8>q`l{n{3)45upBJ#hi#X$lSxMT>PR@MgA`fJbwOcjwwDx@l# z=mHWxXApdJcz9?q%Hnjk*})~e0}{JExC3y94lU5+Nu6!<98ep8p1u<%8H6SeeD-D2 z5{dZ%>>G&gkHs#6fGQBYq#l3P(V>*N#lgc(M3=fIH<0=6((nL+uIy@tYMrk|19Ovi z5X7F{(6xv@n z15l705Mc#2x+9_LPo;5Uwlm_pw|{^ffk-R}JUSnC-mhlJ>kq`fk)c6C7zbB3@Yk33 zg_~`dfE0WKJKLjI#@nZuz-x6^TxfU{|0n$;cd9S9kL}xq_huQ(ZTB*s(Y|ST1EeNl z12q7mtvf2<=N9tzWQnV4wN2cVE)K-4dD} zAu{Q-J2EvttSi16CR#H)qx1&*=8>Me$$$*@&vWt)_r7jo-eq8#9)z5Z>-g(=tJ%Wr z48+9|)WPxN5UhJ<2YBvA{zTRWY0M5xOaubyKfXo}Ye`_C+2Y%pgDCxDlHd*Y&TaNW zdT(n$IpyX3k$Z!h@h|a04+@vR=DGbK zxB=D|{tZ|<XXYWZt)dB-t(2B z0t($3OaAjP`x2?5zmk~x0@Vba?Cl}_<6Rf>AhdC1!te#U&7<)F?gWMz-WS}snk*yHqO2?tczTe?KQ$zD$SSg1Z{pqUVq)ds^&#Ls-2lz8pxI+w;<2{WdZK)>)jjqVKmRsAk1Z(jUUTK55K><6i_1OBSEC_X_)`c z|B$xrS!n-;V~CQ{3PAq$R99!Y(-cTRb@Oay4Z*7FU-fG6TJHC${H+c0wXdYxYM0ts zP38^>=DwlVUe(KlOmEV~Mbb6RiIX8bdi9`y!Q3aC0Qgk+MqQf2Yq3QM#obrV!ivEu zuph5jQV_uY!be)%zAxQ<>m-Su)~8|oR==z`(;o+#5~0$4oKI!PZ?vLq3O4`9>0KZ1 zh&xvsT$qXW58!ZHbwg%j-A-nRr1cr~x`}Hk;^Vf2M#Pl~6~p-o`i`wT$zQ@8k9M)8SC6SZmeMZr8;KhX~Ngrl$ut48BnA8!u{$_{_~QHWmdXVrlNaO7GRp zIN}p8i=G&SHWAGf32CD!Gw9glub5PE(+_Dp@^FWc7&C;~Go~Kqi~0U35--gJsc~=j zrhu#*?V_-$8uJUQ8pJln&?nI2cBJ$BoDYJm%CD7|ByEq;b*n^}pz3;&Dm1uW0eOOwbH<^i)? zkVuDyLG?!^$rn)^UoTspm-3+lS^8BK@m>Za3L z+vd$QTk&oC93><~feylZ(lGuk5O`FPSm?n^AT~vo{3~xUslMr#EpFE-dG2qR+7K>8 zLm7IxM9OnwrKoz+x&nskdv&ovdRS~M$q6^}nryFyK4w0aW+mC^*q#2JnESl1OLnYF zz0@R}{5K?3+m$zF4%N@Sy`u%w6?jC*FyOe0*M;BN5-3QZZvxkC*+6zaiJcuRo@mW6T!tAd~T-(!N1DiG5h8WQby- zry~o!T7HZ#SKLR#Z~~vTq?E0-qOun{uXLMFpV8@PfJ<^@ zUz!>1hjel=)0^`J`n=Vgy$mC}WI3_pkI5GkBN@IckRJ7~p~^H{$odH_NGJ z-MnAAwM_}c;k`~pr46(owsFryD9xTC_n_Qy^uXg~5vUmuFK(!LjDuUMb9W3l}R$uOZ0vQM_vx@ z_)p$N#>;&7*dj9}1<>Sl(rq`iYT;RHtC;u7k1{U}Npd@hiRH9)_%<5M6G|HVJ6{~8 zjt{_l0%H4Ol6YCOhsj&E3srXzO~spHNc;@GD4Igg!$ED?O5rjOZIfsZKn-wGL$|cRjI*+#8otxBmcDK&ro(wB5d6822_Z z4!$0^d`OFkBa&7EBuz7NUVQ(&WW+wT@A#VAb`pZ;TubdqLo= zmuy_~r0@#MhsPvNspIphea+M!me(jBL_1`!#?Qd$k4>)Ka59P7WY(FiH|T$cp6a>GlF0-rml9NW z_=tEln9JIwOqFH?(m%G6EWUt<4l2r@G|@p%JUx*QCSuJI(A*=Rg^U>A<(7ytVkV{6 z^2I1kf;4(vUlBx!>Y%P*jDBl)9T+VLt_|$zKIqN;82`Gv!J)Pggi z0|RU4j0oLMq1JyV(2&{K1Kn!eYaDHvvG}q*DLk#fagDAZj1;q%=sB!w8yja#grOF# z4Z@UllkWS(t@)fikA4>}lzWxWXuiHVlD06;BEXRbNm~mc#Vn*W%XcTyO;yc$$VMam z39o=X{tt?fk!+3<-ScX> zKwPw%uL}nyDm-CnL2>A!e02%L`-iT~mn0Y1mAw z9BDZ`d;#+B;qeJw1j@D+^Bp*hGeZ??Z_ikMiVw;qeW3=!0t;_#ZW|3 zDZ=JWZzF%O=Z0!+*?7Cp)ExgVbPF970^i>=h5Yn_wq}%v`Z3MDB9Iiyp!0k3V1=CG z9cO`v!dD}Z7aEGAL=y{D0Lo$dK460_9+U0>9dFwwl&y_mu4FWxnf@&#_m`tC)pJWg zjKnbWLMc;83=aW#v%Dg!s}uPp0)jKH#lu3Od)9wF;zvZ1rROrUF>D$k=VNi7@|ia6 zpK3XTbB?^3tWW%cVW8Gqa=r2o$UK{NoOrM1=$8dA((s*OWsJs;ti*%q@nBsA-_PVD z9My0Lb*$)bF6WIHiw;qDA%A<^?js+rUn&O?j+onj*-WRrg=&{CnZN$q_&E2!`$HadT2h2vt&~1}VJ_Re&^V8hW#lsM>*La~4 z#I$8U2z~mPJgW@1Ql2_8Qj1?wjZAK;tI$q_JR!VtC9@%2Ci#In*^>OYS8Mx4{E`Ql zO&N4Nd(VBA1&wI+K$T2?Eai2?rU~j~Zg;*RBTQ`NeGVw7O}M7cpE-EfkZf%&Vg!Fh z@Iiat1v`5nu|1ZB=?STcnq!MIV(xgeuTNhNG&UGNMU@{BY@Ce8w|Pd#QsJA{@;J^U zMB_l<<)*ewx92ip9Q?xfL*?P@Oro2W4$w~r3>Kuu!-SQzZx0t+O{JDKBcj%_=rjP5Xc3aI4l2TJ-1e`UT>^L{O+&Ph9#(RO8tv5}S$u z8u2Z6K6Mz(+H9qKB$HM7eC1g;O65*WG8zx+qbxz$FIoo+WI;uMo&u_W3!bsS3Q|OS zqFhviMWSyJM_(|GRXVTUa0*6pk(+}2rmO@h_bLzA%fqDdWY->#R?7cE(1?F4^61!# zf*{w1ChxD?slWZ*%xyml(+yPnw6q@< z#ZYkLh;cTyCCpb}KjbFNi)AmEWMxL^ z3SE=5)D<6Jjhi3Cm-@&VltX{3)p8ijLIxJFPHuRni4x(^@Rzx+v4l_j-liKy=Q~27 zA_Fl+CrdbGD^&2p)aBH6{_UCSIN^(iRQW@PJ*n8Mon>rP_bD**XS$;rvp&%2xBFKp zTRMm4I15CZBnxn}_v>qsypGc{i(z*%5H=0-!-M?atC(JBxw`6wrkZ~k5l$-rMg83# zEe~7a(?-c=4&BM!DR0%=Sq3L<&Z$Bt(+guawPrT(Oux`o6gTOgLi(Ldu#7slpRu*r zFQCGjxhKuiO>BayH$r53jlATNO&K1-F2)?i50t|B){VQ=LQDBmNs>ofFt}uW3+bOAg3Cs%i6XI@olZ5hT#wc}|V!Wm|?g{A7XpMx#I-kl5 zMb(pWc7N>(YZ{w}WHB&TQ_|)IsqM3f8@*W&jtsf`hC_WvURem!<$o~NPgS1TpEzZ9 zbOF0?L(1*jI0k_Y+nF}2P6qTjz?6-7pg}CPO+>=^u?u4(UiN=~K>oCmZnLA?qvKny zK2NTEKlAD*TR%~W*gN*v+4BWl@HH6{0WpQGJTSp``MdR8?8Or(iVO-Us{*yirK9;X zrE5*)puu)yx~AT-3nEsW!*>yhiSI}4M1^UXreNUm+WtO>Xhw%;dk|(>GHg$q^vff7 z65`{o+sEkV{#}3HBp~bBZHo2yZ;g@Cg5h%x!Sb|iNMq~9Rwdmp>t^8VnP_jmUrE=3 z8#5UvBK1o&RWtK5eB#;ta2&vMr>{3QN6a_&V;H~ds)#SmN>r~n5SAZ{J?){xQ6KJ3 ztQDD))1Ha=Dz2W|XgTMnu_GJ8BE5>?j)(N4@al9;3?F~}F6QzJJ`0kY;8xY(txyVE zSdX!$!?>A3eo0APN<2ptDp8`d<2_m_D_%6p`Rr$5<{ChXR;H~T976B%4g0}F)01mq z2majbzW(j0HMOliCH`?GMOY(gt^@psv~EuA)5 za5;0MJnJy^Vrk$Mwh(|JCE!VocmsaKsw=jhgSCIwrRFz>qa~c2@&TgdcwRl0ux^TB zYE-5KCj7Y@6Wt7@K7(fn0M(Mm6?(X=<)@GK39{YnK1UkmRcH>V*Wm(UceNwZ<&{hV zj+yZggU|9-&ui8)Ve~kg&4#_iW|N`6Fw?=&J0f@V6;;o9ZjLbc!OdGCrkYc^vgqnq zlN5h3DMEWLyUilsPH*92Ds`oC)V;ajx_)+Vd3`dCOk>(dS~^06YctFAm3LE@Gn)#U z{k)Vi!I3=xHFi89^z=*udqZfo^VifPC#}9w%aU_+y z7z&%m5Cmk2Pt9WEj!L*~U606#Irjsi$Q(Z@H4F?Paw$Q}p1fu(Bv_L}2V*#17)ZZn zCC4AX`r2KXcWcT`Dl@Q%K8GXIY0WI%!s;_ZsP& zf2<%Uw5X{}XV<`zgaf&?;_rV1dvd6bPAz|OEePWMh`)2n7i>7e-|{`Lqn=a(i@{XN zcKWc5I8Z`$)Xb$sq~g3gF{gB`_yaslk{2gy#%MCWVuEm6pR1L3q2J{}_X&wqFO%n4 zmPLrUfEhTPptq?1_27$7!KUGc>WhT1|8zIwOzCQgZzdN*sT;G8P%nS%@KTwy5*b;? zbf`u%>?Zn+}J7!}v=0V+^+6$`}*=R=b^$FPA8=$16K}sK5#sF zq{t%pV({a7m(i&acUym!u-@&{PX~^(mQ8eYNs4l$TYjAn@(lE~c; zQ0rMOIZV+==J$j@2G=Dk_QQ>3xZ~K={dS%&oL5k%b&GO`6HBRNK0+o|lo8ya{>c{^}f+VHrA_?{Qqv9pmsj?Bl>1 z&Ef@!_N087-lJH$1W)uhU~=REE{fAuD1Q6LZoZrKDPG9!WA|6~5Jvyj}?E_%!(1X8u7%j6-7dvN1mLIM0S_uBLHJ-xjvCD&8pT7gjSQ!cW%ZfdPeLt zGRM21AP62|h`WzO96@!oTf4!c4$6}OQ*cU$Vi8q$HpBM=O@8OLJhRlwcd>cFV4TK2 zOI_bxzCA$0-HhhkCH*vh`*Fp#;z;-F3Z_CW&a8hVDd_5xM6aPT4DZpBL76~5PY9o) zlkCP1s_bEn1W!awWN-6p9Y}rB4jqa6k_v=T(wplpurbJu}05t)cR{G1QpK4rfVI|TYm)`vNEY9+J#eUh-bahNUfEQ^(y z=Djn!DH6d)T$?9dfS%RdVdB-%YA|0~B42;bpNwFh#Y%k1Mc0&|5kxa(?$rB%0_FC_ zKe8~Oc@TisDv`wlc+6&L22z5e7)4ytci(f`uAMmra#fIh<6Qr>V4c zoi4)Y8FU|z`@&iuyuMJXse2!7_WPqz2bPKRB>&I2I84-bV8dU8M!ID=8NW~RLHqw4F zH}!rEEJVRdXJki!Jm~3AJzUrJn;oXkIR!4Hr^0XpR3brY%CFgy2#S?!P*#$t++eDu1vpk{>Grm4Nr}R>yx|1`c3doweIblo_Jv0pnlRNhg4W??PVkl2JW-y z0CH9lnCM}^{Gjpd&k~=vWJ!xSyDDX@!F=nITe;-Ir~KBar9p1sR)rKfW`Zi=Da{;f zTI*VVE}_8=SB;LCDl{dfCz%8aLFL{B-sH&X;VN;3p2$&D0i#fguXf28Xm zgM~q`M=k3{iauGzCog@g#d%_(Eg#>#^nW_~c&0U~u=c#y1Zl{y&FyY3B(h98FU3F} zv1CB`!L^Wc2j5+Q-eiBO>X2@Zh5KOi48F^T0kumYf^9}pIKaptb9IVDqn6%Y_QYWm zo`n6Ck9_y@>E4`!8#eG*5n-Tc8;|a8wKxS3@PxK)VnV$+(oIg~nxJSwPhDC+;|qUj z*Z+yoWTDbJ-|vdgGwUuvA51PW7uS(m#0hdOP=e9)|HjHU5ybtJLid9kYV(y?3=Fg$Qc4v#S08TMA|rn(Y(GL z$twZV_t3}qF|GB0Y)YliOA3F{=k&^x&|g_9N~)G6fH@2v zni9$JmCVN*lQH&CIKjEA$>&&6|wXWH1J-{uw-NaJz)r|my? z^)SMd?3aH}Czj?ZA$5_6Xt&V%z3$}_0l##K zjB#gLLfr1KB_~y5K+WlS$jIVTOS>Q52lg}#IYFZ-t4r~GuxKnf};N$32KH`2mM!Gn=OP{wA)bD38VOXrxr z%tn9j6Z+}NYJ9uShZAy)(fOyfCu(EPe3~8B-$d%vL-eO}B*}@>XwLbisrY#D;iKKI zx6kTRej^ZCU|`7DOrlFfBs1`0U&J1_pCA+^Nc%OW2MZd&Nqd0%rFQmfXa&*tGOet6TR_|8|)@&mlY%;n? z7U1>QH%M7&te88~dJ>aJ?UILH3r10f>G;-z^hbEx&h!^x1n*;vkGJs40`d7`mb|s0 z$KB>(1ajIe%(HYn0W&)3>M^zVFI%Kd;Bodkui2EGoCAd`jIh^D`q@8;Ao0f++|_@F zle5yQM;hH2IB=jk#6~1HC#$|8YG6l>TUWHH}ZRgLSc}C*0c8cr1_~(L2 z*5(vt3T%@z$zpO|q}cCu6JM#o2N-{WU_Q&5-5)dqJH*8NsNp)9JMEx$MEZ2K<=X^+ z?o5_3>cK*49s>+;V++b@H;Kc;K8RHKjA)jNXEq=-yd~qAdv3q#5V^7sQviIm(ypk1~>dq4rYp&;$m+Pcs}c*g$AOL+0(-Yc3@)`gKy^=KO+XE!_NnJzM)v0 zRcP?lz<&#B|9t!~q=G1A+uG5x?*Dc&OnYEzh{iFJ?^0MKq7-NPG7%1Z>o6Rhw1Rl+ zgu}W^oo=+xcdZs^1R^<2N)Lakm%toO+pE1Ce|PKv53>O_St>wK2YM+MxeiSo=W7$r zd|I-pQeWIloqos}Z^i&20#sP>i^X?~pLAk(1R=pIl>7IiqNp~(2lu6BF!3_1&JTB| zi$V%ChFm22!Q=sh%jXO$W0a`QQ&?cp)}~>;2dc?6sj2AQ0mot!~y**qz(uWD=8T-O~DSt;Vds9hh}-XytuzgZams5|uu-){*AKFbch+HZi{R|~*wmJzt=MeM!JLv=yTX7OHm z6w9E8_eA-RAikXVV2|5VW8Zi$^~L~dp6yKh4+C-r?LBTXG*^Eu4eao}EHBCm$%7m$ zP?UQgNftcI?p&ldc50+f66ctMhIPch{j-uI^#)n=>D--C>4V3)#ALbc&fmlVg}4{u~gBEmt}{)QuLB z%~9Ry;rY=*RjsEg7$JQ4lu;;40&X#MKJBuj_fNP?h0P$`f( zBY*D}7n>Want2;(kpK-9%zW3D%-&tX9_B`DO}Joh*>3 zyyLD_{!Dp02%f>;U&Jk2z#4iuji0*^7D4yikJdcg5@^q+sfU>HpL`j<}UU2VyXlC%ydqi ziQRwFIs9V4B^BKoUGv7aLlsKI*3ZdVTQmwABQ0#A9GGh_*Qb4}bYNdgnemT+K>UUN zF}crc&AW-PEJRF$e&tUI{OYM^WF1@*ze~3aF&T*piE}+m)<#Q5i#kX$Cepw9t)`hD zwEVz4MmWKowAPcx=23hF^vl&V$F|&R-zR%DGx$h|4uPQpsB%teAC6Zu%VKW>3 zLmbuMA&x1K^6V#*p=~fcCk}y=5-en-8j)U8jitlBT7lEU?IBtl!05-eq~UftQPqF7 zQx;>b0s4Y%DRJP1CVftZcnQ>CIsp379~E|tA<3n%*m!CL1=5 zc^xf_yd4F0AVsM78&M?5(Ni*Nw5QzoCgfGoQX>HZfL=31+b7o9JbAK)h~H}jBZoH1 z%p7X6g3T$!326_=`Iu~VAN7986w#asDD>av z>L49Hqt^*?xO5l%FLfZE=5b~5g{anwp*dfo-dvh#?-Po`1_waLLSQ_SC|yl~wYd$L zTlCE=TwQnk|YX)WKh+~PU=J1$yxQV5c$E9?<&4)O!3|fG9E~4W0cm#s!SB zjTdnR_sLL%l_WhF#xvw(pmk+!*e2o^nBHqOq?a>-nFvaqxlj1bt3g`bdDpM zh}5+_T-IGP@j`GL)1rt}>RP%ermPx&0;R=^0_E)m5K@1^JEPA=1d!1yugb~CayNfh zQ1R$?*bUHuU&LrRa=vjS;~@lktKgPM5+$rU71tbiLOP--$01)<@iG@HVW@}t){!5E z$0HuB^2~y{w(xTnT>rZ2~|aT~#|JTP1l)3x*5hC7LALChG;|1Gr=H zhZ1>FhjrS2-?0~FDo`CeU!*;|CphzXTFao`r1^i;-h1grmV;WE9FwhfG=~+?a3a%_ zys%2S@PQiZ#Pc#@>6t`8WOr5LXzu$1k3)anB~J31 zSmdMWSme4k@>*rrltEpsIXed4JiF_}kDOqdU1@llP1R?jFQ=*o`4FjAul|gfXIOvr ztGUG4mjL1K-69XZ%9sO}<3pU`GuSna3KH~$P9t9 za6n8F{SIm$XjiU$jvw6dWWbnuI^{{8yVoGz!gu@@O=D9(sOQX0g8?O#`O0OVc-a@HvcVk>Sf0t93Ju~75py3babEr?z+a|(cr)RF!vAFW4*mhd|WXw;*c4!BP zcJ$*mJ{OWT=29DOpL>RCrFnn8W3_+=$L$YUIOT3xVYIP4*B71}H=n=7uQ18dzw2x( zl|XN;&Dadv8!0}dZ)3v~|ClwX=GnrsJ9Ky&!c^*~XIr;0HD&g*ALF&PVI;oP_oqGf zx%XU#9m-@8Sq@CxA#07kO;&@)X`^z#G@&e~P!n}3s|rIV&&62Uh|7PS9+Y7MkLS*n zAkE;Hxvf=ooI?rqK^-+eg3cs$OxB-oNhvYc+6jjJ1Rv)rrs+&j*JZS@T?LZy%WX&& zXR^x--)b=IcaR)m4Cq9;6b$g1dOeair@zwpi>09i_ltjb1xHOCv(;Km5v2z*Uj#vEZ@Sr>K@V~B8Jwu~=?|Z9S`x6X z^k_zmi>yam{J-uZqh`NFz>I?xFz-@StME@X06v0UY5FOVP{!$E9%jM6coI!3ie42V z&3q)j`$S`spy|#h(;l@pyI_XhOEMxY63YzcbsLr{mXPC}K{J0sCUr&g{U}iH5&Zm3 zb?;&auzr4nMB^QX)OhWBC9i2a#`QUJ#` z?w60P!;~CE^;h$D6zCOff9}E~ zpIQ~EHvjMU=OcfvFnG-dqveu`*BdTo;qs#y6bmi4?#PyP+%WmdXKGJP7Vd)k-ajo2 z@XlPW1G7VQX4gewOj2uw)dD|p+q>5%dd3OP3+X{}(YKd;Mm5yBNWSXGL027RVTpJ9 zy8b(qfQ)To)-1;q&V!*6(p;*}g;ob+Y~qTzsW$51_4t1U2fb;&^AeSGslhltAZm;P zTW4{EkWl~oS=2z5c*R)pv;8pX)-2x?+jI!zV{7MfLQujNJGgHPbsu+3s<>0^@i=qJ zY^}Z8_sVFKTSZQx7<;Y`0vXec*(lVl$WQ_7!WtM_DhEU{ULBG7a^!h84z2S{un@mW zj7}5h3pjsb4Z>4v1O|E)oz~qhtc4!cLzeuG>eq_QSjFP0yx&B0$nPkT^sd@@7?U5; zCD%%!d1(pL1EDUnNUKZc4+D_P$%KrENlQR^J0s%{19~L6Da=`(r3Do0;`v)JeeSEi zRptGk!bz5RxHTx4c>&jyhj$3pK|F6QIK4Ts&L@B5crXRetW>04K-y4hT1qor%*1}y zE33$d<*wQo2?{vEn5mxr6oc}+c##`w=nGSAUtb%n2oY(H;I`w!gm|-QO*OF8H|C12 z1sW`ikk~_5ouW$n7)g0gRkOSyJ1j%ASZ3(q+5x_l3H|69bBgL8m}7I0GKB*Iamqxu z2CRP&qr>Axhfvronl}1G_Jz$Q3N;?1tOrE(rjOfu2yV<0f)^3CJo_PtD~8_qc^3Q_ zIJMJQLu`)-uLH~N$FK5dtI%-p!%G4NBL}p33<1c)dCXcBc4?33$Eji$*A)bVxQXYK zW5tQ9XD$b--iHB-iRZEnh%`CoKSDZl6yty5(-(C8T%~#CrEh<-;WlJOQfuhV1zrT{ zw5&N@EPEY~b{$rDFnR}h)4Ybn%q_bZFBJ+GXN2^&m2e*Wm{o&|X@Jq7c>W~)0u06i zq_% zMu*uA^kkd>)%Zzf^{l*elBSZG+#Mzbe?Z>fiD~dMd77{ez+@m_)n66lYRXk!#E&pKB9W7c@;r#VFoOFYqCB{60$4sk!uylj8Qj*wJd zUa1jtM1h~)BTI<-)WRH0$^GI_GtKG~BJ|H6Us15j$>!LwF9Y8g&dhpYt}IId#VJly z{JsJe17(dkERZnZNdn_;bQ)@(LM`G{Q4@qDsy1@@YB4{rz*$Y9Q~eUEp5v+Na_f&G z*yB}f-n$~zwKMi-5f`$Rd3b;Mz^n4~y+A1~9OQch!xEA9$iFS3y@|J(Z`PRFkTbx| z;R}u#<=8oP79?)M{n|UNQ2yNl`}XO|=C`=J{ZgAYzZ8Ym93$FpZq91$!0)gQs= z=FiBs0w7<}3w((Cyq$k!amd;cA;Y`J;c4%GP)LejH-9?5OcZ#WITz6+bof?{>ylCO z*vOkKqL{%8mH;Y-^ETlT0^gN0iquV5h+~hIxBBoifrZ7$2U%<9G-DpIQL-FcUSGqY z-0-w1e$V$=UaK^1Lu6)6BAn74n4m>Mb7j-(`suU0L_;wLTz1qJu9=xFM=6jEhc#r5e5r306pWlL)l!^j#qx7e?J9)&aX z4YNP{+D5497$birg^D$=wojUgms?cC>9yU-M6s4 zo2=$GzsG3ZU7zMX=*0AyG)Gr#V7tdq*83+E`am@KX$D9$oX!Euq+!LsllSiKGeca( zF~DUD-fs`6OdOM2Y&7U71*jjwJO(q-mXK@-Wb?&LKM?;pc z&a_{pqqBc3uc-~is-Tv5&zY@G{FDmT8&W??*BngOJuugQYO?(0pE<|st`BEf-`s&S z)Kl`&p$%W*7b>og=WWx7zk1G4j2O8T7<2*GwWC)JFEQRrH%bAf$IW-G;j-E|H|0+4 z@kavRz*Ijk)qCf-34f$7^-j44ngd@qwozjstu=pN0>dZt6!atKKOYNCvZ$%Iu_S>4 zXYVV1>@voXL@Kx@;`R=0y95!8ddv7K-c$w$+(nfv2i;Vahs!}PbKW7*=TpDM2nHZS z<@lLMl6220f=&B2Yt@T-T~RJ+o0hrFNz+l6gWinf) zUO5zvS8GHd*&ff(^)w>XzL1c{0-x@m5@mmJgNLZ8HHI4n&`NIIy8VOBEiFq zV;h-LO|4LW16X}1U=UfI8T8}lTaM~}JRs_Yl|f%QJZOQ3=d$(8Z4LF+9f{3Lekp&C zrQRUW2J2Y;!Mp6k@nr0lv=w64aHTL?lm`Px-)Tkj=zRtSOoD}~K`C3DYPfS8gnTw3 zW>?rWt}w|lc)rOr;PAo@JE}#kH_bz0xx^}ymiThIeEaiMt!K!<=h4v>iK<@Q1J!wi zx%_DhsSDc=v4PZd)7>MuiMVBGTrPhOf{!1jfO^nLy_@eId%S;K?;Nu?*0U@Y(*g*4 zdN>MJea!1u{qOJ99DuM@mr)@}erq~8on~)$z5q+j1V^!2KWX7O-Os*8X{-nvOW z{;rs#t&dTqIJP4CE=>5Et`xaUrofgZLKR}D_IFF zmR|(8BmH!z23+50wZA?3%LwAB7wJCc$}hNwu!(jH(X0Sxd@R(*4K+QR5xKChg!1lc zf}asP8d`Vo%C@*+=^1V~VlOTbb@z3g&b*k-@jE|f#{hwd8+bodI@A+qbJWm=EI)oq zsmQJznRP(gaQiZ97_om*_*sf}{z#7Y6D{H%$q% zavY*bff`ln{H_fZ_L8tNR+r|?OqrUku45b`H(7GN1n%AfaFEoU^_$|n_dFEc`0IJv zrmc#E%9JJaUq3x9IgiK-%^!AL=B@0hg@s1eS~A!adI`4iZ&_oDlb^Yt*Hn_St$&eJ z$ekJh@jsKt6qbJzp;(Q++KPhKvf-c;b$DK6(gm`wiV@;Po79Y{XdbcMwtui{yq$-* zBH`hGSnFO+j3n`laudVHUDf5r3NlOUnmJJtqBbVTzR%d*;3K-XN!q*PYSo!jT~1jw zM+{_$dnS!fxLx|{d&Cm1{jvJr|4J4&Pf!i!=*7C-;}?HNXsp9De9htJj}Jc6w5#_? zSewF=RsLKcL8_uvjx+MZpMGW*uV9Io=IE@Y*>zBhPkR5TZXH;&u0VZDC{*9_tjC&F zPvjq4$tz*K9q6C{DMC~UYDwk5&^MGQ#=9l}^EW1$O5G$)%8y%qs{H=~1S0$0ITfF? zraL666|#TO-d$`1e#H}-<@tQVe<(ouHg6sxgUS*KD&79w{y!R`g}yw$FM<(Q85M?WIFTaVx#u;9<9IXy ziR(+>AtJlQKGi%5lp3ZJ=g;lSQmM34{rH5Fn)-ivsSFn$y`#?GyCXmep_2!sSVV&% za5FICt9-{cl#U?E{~0NmRvmGZa^P2uTDG9dj#x_p*_}v@0*(gF%2Yr_qcM4C!7KJ? zJ#bBH=iDV-l?SH1kKTt*Q0tShIB=)HzLqH9%^@o^K9E17B_gB{lF3ABfa&rv11bl6 zLG*tTZrtH3c~@X}N@`JX=8Q2_$-k@dsbFi;uSp|Il*%rV2ZrS() z1?IuR#=7a4(0d=A(laLK{dxw|VmFHUl7o{!yEgF_IN4+T@CiTO)!M%ox#iT!cg}dX zbSL2nR*#Y@5uo<*w^)0u7s@Hn&8#=H`$B(~G;|87&v4Kt6SG%yR)SGB6XR|Hc?7&| z2b$WdlYM3-#q?OWU(-ebohydxA^a=G^@X+}{5kS8KK#4zVDuKiw@2(-rAntm6+nc{ zzXo1QZX6zde9~_UhUTCfZ81`D(GM)_USV8Tcd-YgHSB9X{B1dm4&abq(IVQ-d8L2f zP@`IKb>OEup2vGxR0}|QGr_D8Qc~1t%=dHi)%1c$3TX!nV(SITj(R)7jxKxgI?8K1 zwN6-aY#rz@k69Ngbi=mj%erY@vplk|7+StLZLPK|xpHSh;5%D+;w zVoUVavR*ucPjq672JI4=NqzQ~MR$KeuoYoy_7T9>!KHh2bu_c=g~6c^rt|Zke!xje z=V%MViJ{kwZZ2;ZEB03ctdlXpP9{7R2an5d_EGd)!(h(YEv^1f4Js~}NG2kX*ZAeF zccZH^&YXXTVXxEEZr&iV4(T-)W>_Ym9-j=GL#|d|Qz!*^UMq7iFnPF;#hicMLNzm3 zlsI}$1iQONc2h@`&Z{^J9!4BgGOi9|YdfFe3Jxds1;7ARaQa6F567K?yr{1*+5$9a zV+!zDr`!kc_yiJ|x2CK+oKeyn#ruyGr5rEWY*rMAYmG0t(+!ad`{7wz@q|BC$nL zz_HHZc~_Z#`%7f9v$kQM+;u@!e{c*aekOu<5(4k6(>Pi?A^c{b+eHDYaeGsWs|q`? zLis|)%N4DFfGu-Y&3Z#(`*&tWJUO6|(J{}Bl63y?rvXE9z{I3%_gjBvP*;Okv_G^b z0+?pXgaW;oT&qdDWgTzcdFqmXm7(DH+xN$_F^(4>+=LsG$ODQCNrn=ftXUyBJ$=4erVb zxx@h0SD17coSWU5#n68z%v568|9~}X9P^r^!b69_^7#Jv{;3>Hvez9ap5HUCJt<{{ z{A$N7c+=oxVMM0#*?av?AP(IJ4W8;9fXd)4%iDec-8l;|f|7;3fs&tsiiltFD1!^U z2a{Y)Q0#=I%n}RzW32v=uK22ruO-%el(VBcfiSOKY_f*=#bwoIG- zQ33$M;6I>J02iG7DaQ%&yV%_iZ=cdqCxWm^N;(wl z>1htM{9Bap@%(=a^Apa5IZB~*^9E*hdErc*3Mzh7n&jOdp|JPoMXXpq=Y-8^xfn%% z#yZPiqE11|^QVjH>v0SWL0FZH)^L73=E0vc+bVA*4Kg1*`vQKSRC}Z{{37 z{y1Y!JN5?+t)`9DDdZ%OQ0GrY9MSF`K1f-DV-QDYfp>p%g3wwBpDgYH%R+_(osQb% zDqNN#j>`71u`)R=F?tW$U}~2Dz=2ghAwyxZwSL+eyUg&I<=V45WagE6?)AZagCm2? z5#`oFkyzqykt-l%5ns$K3j~g0wF~x2nuZKJl93vph*rCS7*9f0nM+Uw)>Pz>hI_nq z7HBvrt(SiQvmjbSFcy~@@<{6?axKPVuwk4jG>RT^I4p!{p8AViZPTBiW}0%N zyd_{3ftuv$*;rG$Os+&$TGxP%`j#5c7+t~RBJMN>!(R6*Df~gV$7F^A-D1Nm-$dPx zLk-JzIN%Et5Nyw+57h#!yAz#TtlT<!TXzx`6F9+L9CBV_)UNe=jV!1QxRsG~{h3c@s zyKTby5}QI|doQ@Mz(dx;K?qf& zm&D@QSwd2AVJwKNb`PaP{UAcCt?P*0>C>&C%_!Jb;gUX&-eNZUKdihF3tpp$D5gU} z3ghTcJ_ufn8cdR4=uMhEoOGbF)G~kC45sH1`(fkqNb_*}iM*v?SD$1y&H8A9yED{A z0xD0Nn!$NLap#&?zV_&KwAbJl%~?+!DcS0_zE zFPow&EiW;;rt(o~v3w`YWB!P7FI7uo|A^)mAi#a5n3%++>>+oG!tLX{gSUSWE3%bX zGhA=5>dl0E*t&i`@&62w zZPh-a0L)YSo82|u0ScYU1q6T2NIKeOwk8h_Y@I`RAW@re!;X`VlaAf7&5muSV<#P( z6(=3rwr$(CZQGds{^y@(W--gU@2VDeQTNt4XR9*HPa7tW;a$E`z;ic^S|BB%;mjf= z52+ez0l4aIPfv_vfEUrctb^aM)`xZkmRz$TXso$~iK359PPP_WqJK^w;WaH|*@)2|j5)F^vF%hJin@B)jC=~y1)k%5AF_}Ti1rJv@l`NZm@GvF{@3F$D>0EFAcJlKX(+HpZ>n4 z+z4xT4f$5x(OX8Y)VY?-1&1B)*a* z5skPoWIQijI;%|WV+#W*5bXlNw8)Jc&8G;tSj@58&IjfMS5g3sDuuFEfSE^jv*N5W z#KPF3Z}&eE_cuVpIk)B->aww}ss?BZf7M&WP3|3MlXLhf(*xgm(XI)Qg%22L)TN!x zLCP<47`QF%{pEd2`O?O_*yqJi?|SBPN5rV<@}e+#sq1gY%!dLVBQ|w9xMpW9*jHzEJHo_>T8|vI&KT4WXC9L4{ut3eaOMShYMC*#VuU(NFwSj7*>OgV>`_<@XbVVQS|GiWFd zVV_Rp9O|^m*--W8Kb|3C&2T>`yVI5(Z>WsdP;)sipOWPB7-wc~|G}6Ls))!kh&$Go zFj=yfyWUL))ivuZxXd40Na`LgbD2sX1$F#CDkUQV`TLpr3r^UY&{lasOC7jbyP)K2 z>Apt#a-1W}-i#J>%93gn9@roh5m425+(fUTZ73tdh<{bV?}x^jvSGHpx{H`G*{y3H zfoONKiH>Hxy%en$Il%NXmM%E``Z#fzs3e{!U=gawQe2MRHCwqz598%nWL4=&&yf{C zAhJyO=f_mg#~>zMxS0yTSbJlWsC^K?oxnU=i6VIf7d9l0l(USs0) z<}FoUfQt@>QfAFTu;LY9TUR2f``p9pn^anCu4Hqx<j9PN2PtG%8=^^hPf+2yEALiR_#v{>G- z|K`V`4p_xonYegum-krezS@Ti${x-G1Kn8t7(6=HL=>` znGfTyaR2S%((Fe9ZaeF!Qxo*wM64fF6Fy;skt1SOtO6`jAnTQ51V7DcDwLq18A0GY z{B~=FU0?};+LQ0A${Ge<)OtgCd#Tcfv$KRy$mh|LmK3Puxkq)VXB@jJbK=*A)-QI` zaeIgb`Xcjkw1bEx(@%?pMBCcrB%vGkq`3O!At+*Y0{qLIsQ#R^`si7wPNTxqet<#dy6S(KEu(p0 zlbUW8{xwxM{=m`gi^4&|En4_{7W%H7-0d7(d#&*2VgF_EeqGFG2%pw#lz9fr+s{|C zI#=-?xfkY1@k;#*-O!*AXFw1Y6}z`O>`VlU?JF0Sc)b`RR&kp}J)e7?#AHF@64)U# zR>Ab6wQ5>=K@3tBa)8C#4sTI6e+WyHZE(NUlCTNeatA&jYm~kr|8f%-X1v{{7OD?)3Iil_jqbdGScCU)Q-V5H`MHTFQ?is<#+&7Y3=aj(7#|UvsanK(;Om^2lwxBR&s>YD8@>1ks0_lpMzI20;53P7-CbHi0U8A7H5zhF{io*4<9GYkJR$+c*wrx?kNU?Njm*;A?1?r6qrQKq>k~DQr2jbZd{?~rpSd( z4reg`YTc*^-3l|~u>X4(63~ibSM2N1hs#;24ZGB_bD{Z5w3vZhzOkJ!&)_mm(P%A} zcY+EU#GC&d_zC|Q*k~^%yrLbXf;Qgb#&NfeQYB;7BQQ8g@*REk)K^w07Yu}vJek)! zJ4b(}AEP|5C*;1U61wE}`_WCA6)V3|3l7S2nMV*LTuweBN7Fjl5)esMxJet1)$x)S zJw}vV*%)|#O-tx2x!AVX&n7i9D3?xvIhAonT16Xm)thy^?Hv8M10Q|cq7x@bp?eJ2^*ujkW#G+O>H2Fv zS9?NBj&+6XCz|>1+X)#w6nvop9ORN!`1ni7VO^^w=sX4Db&UlchxG&8gnr~ zcKR6~bChE&NBg%@W)ihxgiN6U(Dy=}i7qZT2sRRO+;|hWIfl?c#WV|NdJ(Vp$`*w} z6y5KXr{OGtAtioR0T%=g{v;b?!0ZOJ9dUiC(G#gMq12}ICJNvNzF&={#m?3m+c5bQ z?e$RoL`PA{i5~o|+0)}qtee81__i>yCFP%U=Ay12@9Q7q5TA_7@(l?mU8 zNp_)G8Y=SxH9U>ZA{%Aw3PS4 zt=`jQxgYMpL#QtNQPDcKY{5&;Kp5ACmMPa3JsmZS6lSg{Gu~R&2ZC^pmgZl-7&A5+ zpZS$thUzFRC=4607ghyQtLE1k#m)`)FcTaP$1e$L9$i?PWeB%?)^HwVJ{!=pAa)&28edi}9ze&~on|xg%OW zSE_S5x4A~9ku%BV|J_|fE9u>pDDrHgEon=P=nix0j^zThDQ!1*_RW#LaoRqoQ~&;R z3(tX9>~wDBN?+z6H^(rKk}vi9+c08-s5wI+o;o$QHrC(q* zjc(rLp({S-glSQ<=urobq7pe&k#^fS%QY3Ak-XH4^S)u&qC6<5i9HaX0b(Byf0#Vy3>k+-a!cYF^^`G z#%79^VKLR@Q6lPw$!TVZOoGKzMU@S^3gAnEfD~$K85Cv@?W!AZC}@n1QS$F{7m*KX=V=f5_P@STF3(15 zl(n{DP2}H?=l^M^K@8eoi~0>P{)&2ceRZ0VV7);676wJbk35Cm4>kuf`|57= zkb_pVP$FIi;g5LjQoIyWOWG5_9P3I#z36b(IRirQ7ff~XY(iu<8P3*Qu>NM%ahAj< zx0kCO6<8piJlFCpIfa=)LDtUlP$F@+yp9GZ)Q#PjL_il3Egrp_`SwTXP5BrghK~ES z@i*ly65{GkuU?lXPVYa8a8oRJ4!C#ZkL^mwIGgGk5f6RBm2b*Auu!$I7-JO<+KI7# z9uYAb8q)1!`dA?6c4vZwn=uK;DygGDG4{Ys)Uv`RW`a&1f;~ft>f01PPp`!F?lVRw893LrJs^IuSR_m;(`FNcb3{E#x|aD%g{M)ForBg$huh-F)snnc(Y9;oaHI&83g+})S;irzEg zpTnr~>vp<}ADEOzc{oAGj5%u@IB3oEV43|DGFcXnG4*ImZ1W}uH*^5?ZoU?J5y;?> znwoX?zTF=rP{j_ge?R;}4~0rl933p~{p{bC3Pl-mctase)FbavIt*7kw#AXrr7eK4Z%jcK1x)C*vNU#JNm4}BrnQr&+va`Ja?kd`3tmY83Ug6YhuC+ff4NNmR-`3f|f0{S7 zbBy5zElRSYcKL2HPD1gJGdF=qz`^=BE44yFuS8g(W~e|WUEV}FoP6MqyHgE+ceIe> zLLi8qt7l=iS0`MzKqIR|$kS5A71MLcN37C00lP|*X1iLaY2iE?ho>@x;{Uv_Is zWSU-&9bq|QwXlPfpikupNN^o3voc}+u^~iRlPu@%rZVZoxok~JODW@yLW_VV`f@#I zlEJ)iNarFR{e5IP#?pE)g~-HX zRJ>}EcNVB=F5%6kv!a2&^C6Bpvbm0?DjbNg?|igl&Wc71oMD-XT6ZO_(YXocCS=$N zxMdi<`!$FW6=rP?cLF)L0@{S#QhHhh{#=_zlhqa5 ztAN-NY2qCiWfl}hi+HwL1?2pWG(#Mc;feBJLjk_9CaQ$pPX94)`6izs9Dl@zK4og2 zjIg$Og|tDxZgd72)rr5tGRTTg5nn^4=w4YSu;S@axSrnR!Z}l0?o?zk+}S^Do!2$nv;m)*Gb0ssPj<0}hMg zSwJ?CF;Qw6_~+z`D>BYUR$PtbhJ^QignbV+jrgBkJ2Om2x;cp26e~A*b{q>l1oXdf zv!__t&u;m`>(4V>kO?IjhL#VtJt(wzp=jBfo_G@(;fHe4So_TpryqtdB z&$M3i#gO?(Gq!Ie^9wk&H<`n$YRw7x4eK>&3_DqISAH4S z8=tM(uV;_&7=LcQooUlh)Z*gy-Toi_`?e)jiXW72S_-S^CyCeG#UTTBycf(#_c{z@ z#wnd>Kc`j=zEsRq<6V-cF0@8OFZbeC@Tq2P${`Fe*h_uGva0{euntW*zk*M%X#y%e z;1kPUj=Od;xRhnV}K0msJH{sEr8U)5gi`wmE-B&6qejq??4XF&*lT7B@i9b|)z z&Zw!#&>Mtez)U}fhoI{Nol5jickmMIH%wOY;z+r*iOHoHF56dK?)NBRrR$gBm9L|~ zM@#jUK@tM8a3%K}MA)OhZBZ4OKY)Sc9C!VF`9U4j$%tJtNemP$nvm<7-%9`BV>O%u zda_t3QeRM>1q#JPo{Qt65hjFaM~=0W`&MJ-Fn&8ir8seEKx?{`KHTTY&gBFc76k zt(T+s$ z379A2pWuB%Ls6BK6Yx*FJ3XX#zQX7_BWNwRu+G&{qDIhVF(l6@rjoZN-sk#QqtRe=z>)r<6~5i8ma@UUBVij0 zoD71u2PHC|)o7Re1pKxfC@nh%-CKu0+qiEF$|!X8#(jAEnrhEl4Cct+l;=M7;UYZ2 z|EQK3UgXsvXZR0Z=e)ec@e%y=kw$LAR#tqR$(gq_bMLi7wP0L{VY^OIeyhn@u#&aX zRjbwKUyYs+@)Hq2snt~Y4fH`4^M0)n8I`2;z=l1Z9C(WVyD46YlWnE%HPl%{_+XY9 zyTnDdxb>+Es1e`;#P~$^OZZc#I_jJPsGQF@3!nAqX}_tPF1w*9P=~d0dP(16Q7qN} zzZeN9M^e@u96A*{%YWxam zU#V3{z#tBXZ3AHzVRnO(h-_naL)_nyg!0;I%i(Iq!TV{;hCz=x>!JU*126SA`M9!T z=CBlB8J}4_Zc`Iy3^q_7zuX_oi_4K25gG_)J77%C5VqCTHFiWwN>UtP5g8TT&as1& z0-zvObz}m-$~-m-+>M^$o><|xVUP=-$mn0pK7XMEkdObA@8}~LK}};sJjsb^BVTw_m2N}kDOyK<^ zQB@WMV6V=P0_!z_6hJg4K!YfMzdIH0yFC9ihZcK`)uKJfL^EAJU^z!ud zjL!sQXoe;*P@vA`wZlZ!`D+<%Abvl;V4gJ@oiM_5s2n<|*l=j~{PI2O>B`d}78VSI zj}0G$n;+8$VK!gU*#t#TOAFePVt?yaLi3lS{Wos+>d&qSD?=#fy1Q5YZ{ro`({WRcdZ}uniU;_fFsv0`n3z0d@?p2M&Se|>*w7d;z?0B}6T1xzbTX1ofS{@i>Jycy?hb6cTsniO0&#_I_{%7%ju{I4`O7$n z;bkl5%N(o~Xr$sb%1ZUU5Wojv&gKM|%21ZEXHgu#zRv5vO35z3z$mfTz)n61WWLsB zmKS07Z?Y=C4ssFRClY00&dn_0U#l3YDa@%|vQt76>Pil}@;{QaCs(j$zYLr)2nmfe zJ{iqFHx&uRuLLIv%C?3ueNr;Qj=Ba3t;{V9p&aeOY(J>sp^}Z$;-qJ!WcdI1ngyr0(UJnPu=4gjqP;^>q+>XK}+c(2Xy13=rxtyk%?@ zuxb~(N6=({ycmG~Clb4l%6Zd?ZU|I2)hE;qm^#@fAs2%!R!TNbj~)4OucvINi5qw5c|`4PMNdgvkn zvVZ{fcSs`Ump&i3+{S|74zlS>dZdtvQ+nu>&7jGpmW-B(_CJjd zg9286vF+#rd*5fefnfo^o-aoDl<-uZ66En(fOCaFL9NqEcOdh@q(%wTlP24RVHAqF}(S%`~Efq3Iev_ zTR1Frr>E4TcNRYw&7@o6H)8K<;hnfA+1|u;dt)ng5@#h05!<4LCJ@u>WAX)`Dp>7T z?&Et^1kz@Jce=flebi7Tp@O<(QkuIq*a?Jo3Lh&PkZ@|!r21lqhJ7C3q1{8>a=x^Z z{hq_7y`jwqnTQdB`D~ngGmE^zEFUupeLA4sy4&6eHCais#rw5<8I~MyAsiNWF+EBD z)Lf%5$ZH@4&rEPDbS=7%eBQ=w=bW7vD(wBQvuFgpz9T-f!Cg*kiQhhpCaWkAJN7dkhw{QTiIOd=Z)5(2wE>HsU?!N(zV#G9`~kVScG-p2aL|9HWMyI=&4^UAYzjeWllL%67~pq$X$@v`&8BDQ z@@{yO43#ZsPEzjM7R5NHNBF!Pryzr!Dd@}JQZjZQF;2L|U4bm6d7jV@lL9GEGOf%3L_)1Tu{|DreVo z6|(zfUR1Y_#6dF!<--g&=Cx}4*!j2i(@4!yOj$jZ?A}pwCU2Q=A9%buq~Bb(4sgaCD?Ii00 z$kj>+9&Yob{ed#5gyCwHIMbVRCubB~T%%Eklv0GM1qDQyE)I|1$i{1A7o&uEzk7Q4 zSfd?&jOq`zuGI1#W97YN^io`98Z~J`T z20^fa#*(Ul5~n0)b_2Thj{d+E_j6E?D`c?!L$0)*Ew22yIDrzsPQ_mv)Lu3pzI6Y@ z5G~JEGi8}2Q-!&>_Ti855}`^@a>ZIrOBjLqaDiCCL=_G{OFMmB zQA}bvcqueOY6Fqd$GOI1=+)wsY}lQ5^&^`OOYFmBZF07q&;fWx9m-FZfCZW(+~d~t zw~-Idnb?Coz8R`xfw_LhW?Ejp3*z&?B{?JsK*E-@+8-*ARp-y&Qa*je7jLZ=K{caCo!gLRo3T1H)c^@vlWOnq@+Vm!&U*Fm#BHu+a)tYi^ZO;WQ2 z*e&fD&&&0mnF0M7t-Sgt`x~@lZwmo*{`9|jNb(`LK%J<8S_jddX8VAty}+dKH?mZJ z^!F6*AMZN;h6R7j-*_!p4i$s)32)Tg4@4k;d#t1xjgz1scD!GcKT&TlyHSH=~0PCN7<_XmI!1#h;%?Z9_9%@Q7734j~K z9aTOb%OooeDyl&LjvKumwKtc>7|c}aqGK`rBzA@$R+K@IVeD6WYr)=eM|TnUU; zLtuZdDg@xpA8%Z!Ns{TCNLNJ~ht@8$rH!C7B5BO!K1MG=lhpy!0qt4d7ymuu`@9MT zwG?p!p;<5g&1+jVrWnF-G&yUbLZs$MW$c@h!H#Dn1u_2EE{V9>U+91pyBBSeY3>XP z=pMYl4WVW=3NQj~B>CXoi?aisBZ(1E&Gp@U+YTTXIm?b%-WfbqsVkwME!%Sdw78{{39_SE^xKj|JJLTKqtCwPg|Y)~eR=SgFf2xov~ z3MwWkuKzP>zNlGEuEsh}#f%n}R7luvBcM^>9G`x+p{TUUhPD}Nha)H!_^?IU|5xGc z9snx38k(Nce@!-cEzVdeDf{p5VI3XjGm?#l&1I2I*Py=$v?6b-uQ}+6Djn{SGq*Lg z4Nc(qzWluZ^Y9N_qHT3fQ0O~;wKQNE1C<`+i8di6fJ}@XtX4H*Xc?C%=A~fyy1_|+ z(QqjY`J3!2pye;0p^TLNs^V76?EJsbjOzc%Xw>Da?4NNJM12*8a z{=?PSWZB%r#Vm_QEAs(Y+y?vAEJ78@Z8JwK3iFEYBoEf4ZZ8puIj=E{z)e3l-wsyj zuX&Kz&xuE`<#u=!?IT(j4Mx7$yebA5*OP??Tlv^m+P6MY8pSY{;d(Jzx@C?Bp|Z+%;J+l~TENjoo7!X`_%v{US85PO2sU zI4$Dts6?B1pzUQa+l#<^>`ejFPVMA)Pc;=y91@Hl!4sz-N7(ih(PpAZBO@lR97)c5 z>79UY(t+Y7=wPlE8bw;3UUULOw0M!w#%>-nYm>=~Kw<3+)62YSSV(kKdfqf<0`68@F{MMYq zD8-00+g5zTCc>mT-guRI_is|{;eD2 zMWXod11|q;ymC>LbWW#5SN_(dkDY`;27{`Yp&Q#bxVL^JGp!a7igEu@V~Y*1K#!!3 zM9~uYrn&H}G2n}>5D)hgFT-?koQ8DPTVW=+r*8!HB{ytx`E~BTiA=>Ca z1;@q06MiPk}FDy1DhVh#|h2=&EQwfP3<_R@WLB#0!VTgx;PPm|KM~jqE?%5rDjIo+n z?F?({Bc2vsZu%_MXNEmTh>T7uSTGXfR@3G=OT9=Mc(En)E%!?JE)lNAE0X9cGW)Sf z`;|IpbH!0OFU(Nty$L0fSUrv(o+j0ydiv?ti*LR2Et3uqg#EEnElRE{VQ7s+C>(ox zDtK2|v(9qlsXfKU8ZtNzV4`xBQNB`UaTzFgL-Lv%RGPLLWO}SzVRvmP(3;_D{=}hF z(z_HolZWviO~S?Z_?!6iV_)ysuf_*W%YJ6m7b*rRXwCmN*;uA@H_*kozAa;Q`l&9A z`+2xYE<+Fyccnod1vg~cpeopM<(}1XA;!4j*&t+ztJOH#95g|K91$}$xn4CXHcjWb zbS%_%8*@N6HJ>r4kF@LTbt>*33CDQ~^CtUF#(ee}PM~`6!{eG}d%S^KaEa_NjA^4} z?da!h?`npc_MifF*%Md%QBsbN_RimS;wrwn=TT(hv`-m)4j@9H*EqV%H!|CPpe+PIv9CSN7`+E~-2cB$bivK;BFaLqT-e$NFftc-xCVrnrp*g-cCSF}(j! zPYoQP9}aNC5GUO28eTm-3=F+{IkeIY{~pQCSnCd!}a2cW2M37b^7>ta*>1 zi6P8`!l{d`qj3uuIneBrJo2h4%=noQUu3AeM2cJ5;f0+2+_drs0zS8YGiUP(_Vv1) zc-O`;xgeG)7lKD8K^`!NWnfKKYYDRKx!?toQ=}Ya7BD=qT$ch@ACztoMSU-WqIb%# zN|q!tZ73FCyOs(bE1zrJg*fODxjBEtP?;M|dqSZ$FTxbKkKSK~+(MY1eneyra>x;o zh!mXNWENy7-YdN35j7t*ub)<5j;~(y7Cy`ehfKsK;!vXMaz0fQgO&ccE-ZT@5>EkQ zNq9MfrW}p^Z4j^Y{boSoV z=kH6WkbuIJfR#9Gv4)ew2O;TMm>^dwVfv+na???`t!72mcUyI{hIY~croaf1J+uUG z`X!y7mW4r=A3xk5FLo+zVK3-LDGNV-OH+Tdk{r>n%*krj2$M^$-o=g9k17E;jvW7; znx9woaW9>pP{$6N<+Uhm_V8G=~knQv@)vMlfUY#hH1Abo#~?&T$2 zd7$I#E33h>a(+n4m+pk*NO*UuSH-GWaBjeaCV9+rsvx2d^uNzve+$B4Bdp`QB$)b4_HsLb2KXZG^ytKF_3OCew?W+B5TrVBa7Bi;M8ZZ2IYfNB@FGs5tA2+Ok--M!ZEy8nR{(6bj9yMh&&-@Z+Ogp6HfK ziWSyx>auF}nR^eA6Wtcu=|Crq!t{unGn$*PL22u+1@pprUTsnUA9I%!+?5`vbD)Am)~dL|?@R{G;JvwkhwuC$@O{#XL zS?nx%H8dKvZ`kp4AqKItm%iU!Oc%{fO1g0b9#Mcwo}OeQuk>%NL$jL=1i9FDCk1d? z#S_EN$?qyl_Ti3x!O@1bBO{p_!mECKr5PBu9|%xKc`Qc`{@QXg^j-sVI}|iLzkkrn zNl5Let$FjK-C+TI`nHzw#&)0QRo$#h6>f}{tm5NFIQ9o;ie0f@mJ@^-jSkIa`Ed+B znp%UxwOhRfAJT#8r*2hN`CEJ!w+c_u4K+$V;3BALs2TJVphxOj(N7;0za|5GoEnAc zxM?=62vt=$^;akYf7Y~nG8N(3`R(G?5R>F|?^7n0iueE|DeLL7C__p%EvtPpZ-4xR2SchISC>tq-52KV*GBZ>5&J9pX6KN-=k{?n$<~UaH?h9#|r( zgcJ1|@R5P^rYFdO$#eA4bO+n{p~k{8$b*Q)P%xuMfAbz))p{lW{d}-L@*V+4{*;BP zq|2X!8;`meYS6lC#tJDQ)nnH}F=ri)+M7rrYQYtg>J6X=7!)OOOR)jpc^253_6^Kb z9PzZ>q@Eq9*s*Gbt81C-7UZVBmyPtfnUJlCbuj>EOoDj33XF^hj*-QfqjgtW`Kuex z9o_B6DA8jSfiI3*eYq4uQ%AJ%{!xk<7hmRdNt5keeR-jyzng8VIcB#`>V z>>b|n*H%s;Xi%U(j)}=zcxXNh?StBNX9}J|>x6;hUE6 z<~#xx>6($aA?YuAWBlE)8`OmBZam!w3rgLkp1#|Hu=UTT(j~i9o$aq)@g+C!_=|9Q zx44CUQT_Xk4YIZ)J_-)bqt(Uea`Qe&{qt_)n3Z6&JtUKtsqI<1raBni-nwWPnlQj` z{&#=m>LdOcx@9wG^*$U1#(HO#Dijr_3?RiEg#!>!+08w{& zSJ+NXvk;H}D|WGkBydWVadx0$umDk&K2DIqfykmoc0d-%Vd9g;cGHbCcMCXJi(Y7Y zL!Xr9A&T6CK`?W}C4MjotqQ(#b1sDU4Ne;02+x$4?-zNdQCUx>jmi_E6%SMT!9nm~ zfGg~_uqM*B@uJ9dM0_4$B;Hmdjiu@UBSF}!%FJn=PUXTW3$Gsbzyn@FzD zWxlGt?W3`DKGUP=#@_KmJRU%>@WrBvTT2USuH~Ag(zks&C>rXdyDl;VQRdxL>pqI$ z5$zvbaT}3c_SMjmd1!a;rJypUJNy@XetSqSJ+c%MfaQC+Ev12rrcm&4EHBH?JBIVL z()vVb0;_?nEYeOr_?*fPg*oP~Dtpuq5IV?t7(1XJWVy^xeo@47HU}(s^tzgu467?n z(+WhQ=Bm(VyA1as8Q#@n*n3zh2t~DW%c)}3G)UU~5i>Z=pDLlFIhM635JAngIb_nIMM%5J)ypu|O%Ep8tb>L?OSdF_jSiHJvH}-#@9mxv-cJ+&By< z+nV^|6X?6nOP%o@xj%&V^qolWN0=gABO!U6Q;Z=|rtG-ed4Y$JuTgUC z9mte@?xG8sPo{(V=0j}YW7a$gJiwY$X(XKa+mP(zdpR-$Wo?SCBVL3IVGKnijMSA} zs!>S|qmQap)V+Mmngd2s+#j^XQtKUh04F*6{e>!2eF^TIunN+g}=2 z_o#b)q33xufVB61Uj=K|9KOVvovDFIz~6c0489ImuN0^Z=+a;z*t$~bslsU0M!uhz zeDXAnuG<5{e(qo77*`<8^tKEv8*u4a7_00$S3QZ|+w-XyE3f5*mL193zCn_F8s&Wy zdo_Y`vM7k?-2kQ4=ANcE0P7%3<4+o~73zI;#bX@~sJWTI*7@^%jf@a`>&;pjZhSeM zecUQDdod{{Ew^c&916+QJw784wETjVxNKW~mc{iD<2=oSObP$xvVStXf29sa3#;e| zHKx0&edQ3TF+Qmcf*Qzy;}Vp=J>%L|B7f#U>#L#_ATQ{Ki&in2bzgw(_L3{icZk zD~`V-R59^YHVBCYI+-Ngrxy%a1*G}=Lb%H#<sYV zYCj}xDcm~N7FqA&qLj1j;lE>zOXcp2prcOC+Au9vX$K2zlwro@h-wuTBoPg@79u=} zhMbMZv?o;mn+l9nAChZ23j&_LMC<(hDb|ZDk#qLUhT-_KW6fyotCn;YWqgy!kLD#o zK%M3WmrecRkDKt5QOK%y7n!s(BuA@w!dPAvt$5P{^F#g^bzXh+pT4@k!HZYUqOjFV zfB1I2$UY@Z2`R2h6{O%_L{-$+3(4MBNKiDAE))06))Xl~zfq3u^*2i~B`M;WfW#m_V0P~BHbl}{fp|2%rxziuKn;(|mDw6s>z6@eE zav^>v>{6iZfp>I&Ycvy4@p3^(M=40{zAov79p~O7ucU`+97SdR=c$QT2W9=rlr-9Z9SrPmqg6ZYrox>T zp^+gM{nPZmyooz%~=pV)brI$#RvgK`YVUZ*Q@F zKehYR9`OnJ9|OT~IveE5eZX^ZYcWlwD^d~mULzC%rbu&7SZ-mnzq9}7yJ{_#aF}jlhnqA@TI!*{r4q&l+Y-Osre2qQJ~b=H#RD!;QqY zZBMh!njM5*acfUO4>vY n(kbJbeTi0X$~fwo%`1eW$PH-QB5Pb4fkyk-KMS54c~Dsf zJ6{R_3?q&DyPxmK=7%E5^~E|Gxfw1?-k;EI4F`lqM99wS%F{{qKr#Ouv^#V3Kkzjw zPtAdI_tYxLf`4NMxnV?1^q%S)Ln!X!N2B5R>+u`+V&L}0Ds@KoGG6@JS*6sDdQK37JJj>?+)*)6 zBq3jKCG+)m@o&4Y3)14(A`osH|JLYpc$ZO1PH=9tV_voauNo=xrY&ctC*1Z`BjgnU zjbrZqkTfgDq%C)}-}*$ElqE%%E5%p86WXj)$MjkP#jIGU`VA?%C?~|(c@PpowKsOM zb25|mLibI+r~dvik!$GIrBcjCoxY3bkr<@z496#z>ScDz4_8x5$fRf9$uI4(cQmyi zMBhX`zNB%1EOo_;`V8_zH4-myt4nCiBrz3(3j*>-T!y&fH#~l1t4X_a|CpSWs#O}# z?+c?N^?;-Qy`FH)C#fFfW?1#pzQ1_t%{j>48nzXabyv`Ujr$mJ0GIk3DJ`cImB@`N zq&dX`ZiVJ~8a~~PVVuPPuZPlM7J9S%LM&z|24BA#`f&)U@L2e<RJc^`gX)^DA!{D`6o!);{Gfg)caZ_2U(G4wrs1gU&_JsM*HLg4(eZ>?$6K2UC zBLl#h<|w4Dct!MU3EL|}g7S$rLW5<3$u%3c$i9-r`u3)O%)G@Gad7ZRYX^zw%WdUZ zeK+q|&Clu}5_begewNsmsZEqI1^pOX@9(oPaE@)x-^I5`RuMJM5k0=xUQW$H<5bGP>?`JyKQ*M?$Mjb!h7 zmc)|rxBnxLB<|IES2|2LYHTCpO-TlTsk;6n*Qqscdh(q>jZws9j`F)<_zgTcwad+8 z_9X17{vYz-^h0~Y0nL>M-xcOh#Kgi_yk=fS*f7w`?>@XZ`2baz~34XMx(Dl!I@M#Ux{?|z^-4^ zC?^q7my;Q(btCH~SjoHF&Zm(rm?E*~-`}ZwZ3AFPyD=Hw*ld09K&NZ@&QOM(4))gxCx$q>KFbFqe1%EHEdT$}V#3jRowQRk6VjeW+skGeNlsHh(9JpZAqW zl`9kSl|MNLk@-KokGD#y1FEy^i+twd3#@vTJU7}>d}(&>>f4^!edoEqb$0wguA_=3 zuD#w6Z*CkvQh1f^()7|iAJSH8 zzMe=DZhEZJT_||hl{YXZRDX7UV*hkA|9?>9s&4!J%X1Q+ zKEs{ZQgL~N*4o$fz5$eat4b!O@1$bd((t-icNXpDFzO{NrZQQb8;|~CrB2~N9A6{D zJ}=?wt^O?ZMrqF!V-cfgFF@Lg844gua`HO&6hKC+V>iP=5Eg5lyplz)ebgS|p;YqW zFFSsE+|z>1&BO1mn}2xMiN=R*T~@(D#dSrhK1ZCG_Nf^3O>!y8tG^WLe%Z#WI)vCl zBtwdhN6fY#!--QxKXU4u77x+wt+SeN_07&3XQm70)5+JwssAx`Rl+h_T z+ga})Dh5Pbdba~LARERw?4Lc0ge@M;x_JF#sy7s%@O(TO1DDfgJ_3+?xH!B^*q4=2 ziGU;=&uLnB&+86qm3gr1&F;=rLQw$SBJPeUceMSQ*Fekj7_XPCsW3GBj4Tatvs{gz zwp)!UmG|Vuf`2N^LPw2X{|{4&(>Nbuk?UeD7KT2gyJ8E!!LwUuxeTf`6}4-^n0_eX z=PoA(=(qEE8R~8#?JqrO4=~?z{j?h{@RTbk^7x`qf{QqQ#`iaV>Va9L$Wl#(C2SGF zda14Qd>a1wrhX!ai7(9d3xOZBr#rOWevVow?&oeAB!5Y!e1Xkm6W(v}jg$uf$RQB) zg(GWqQnFo0i07d#HMQqBh%sW9;qon(?xsAsE@waVc{PK@gQoePh zc?$8F2Nyz&&r?9E#FfC02t;5>8FiPw)HiNM)D&L<5wuG~$l2`%^10iZ76R6u^1*|P z<(XTDP=5;-#9@)}!k{*;cRiHQ3m7=@utl)?jURica>dubze_RyWvzc5vI z6J{x~;<#SH2&Dj;0s~P#lzKj!9^hJick65zZ-3GgwxuUHUL?=a(8$Ffs2DJO_l@(M z`zbf7JGegZNbDHT*k>eTkfMqIV~u}FrKd=Qjqx-;QuPOLJ^GWW=H~+G9s2$=BlTrr%~Jgn~ZZwP-A{H2Y+kXPM}ES*4(?SeP(rqJ~HjiGB;}B?|20) zaf}+=)CwL5T1p64TJkd+0RSjSo$I2bfvU2nwC6KWze^3n|e2|h&hFQG+rM&TyV z(PJ(%->5D5#KIZI^4Lf52{fXkM^(dIK%3_FozTZ{yk;11y|2AgyPG_L$u>rPYJZ7Z zl`Re*O*^2BB;{{WQ2tCbvf(_1_4yc6<(CS=mVSu}FH=Y^4}tzK^3Gs1)Sk;PODMRK zKQ!bBQi~t7lB!xP#QlfYwwC6XfaW|NFQl7exMtNt&~!CybRFnjW49Q=?#9)6hjSpE zQglH9W9_+IsL#usIlMmOv+qYTY$@Iy`$udCf0DqmegW1UQ z{?v?Ifqk>U?}hAVy%F>pXtTOU+QE16)Ar#Be68cV(?-#ZL}^VT&!g4eF6_aKpPKUU zNm9CJRc^d6p0u2IKE747%S*9c`RUzk$mQmDF&rj+M%f#4S&?D{S3F|4Y;>{es)zhB zp)W@-<&32(OZ>@Y+~8Ln!++?fd+0Hrg4l9NFZN=CFKbe1nB4RCbqHy!Ekw5Q(laGu z?V}^U-Rqj!iD~h12Duz>d*&&)kq`4xN+Nr*V&Srx2)USgY$cU7hF`vPM;3##kc0$> zRXI%gH9aotX5J#>fW*_Y{6S22Raa-e+nY%yvPNplWr|;=2c`QD4u52{79V6g?vwcP z&%)!1-Ymf7hla&-w}2iefy2{Y*>y0eaWV1xM6b#i<|Lk@1n)-0LApfw%vV9-Q$h1e#Yd5yl(r_yVDBn%@y{UBNtyD&vp^iBK7!+K zlF*oY6K-eO;d>Kv-+z8|2k*7I@;d9O4%4W}mdMT%zw?GeK$thj;c}I&j#=IA_$Cab z=qufYRez@i{^EoAwzF-A2sER?P{>Q+9MiHmNV+Eo`zZC`YJDYrcTP#mG_{@+q;pp| zmq`V0I5uWKU>zfTxK!1+RMC^Tdh1SVDB%r=y7n#zsDs6EXn!Fqc(eF)jzdxEjku6fH!c}WHHQqjZEW(<;?yz0Z0r&Jfn}RXDlNDj27Tu;W+!2J^`PWC ztLV$8)`+?^qSl8e=^_vx%I67s`FM^JSEJ0YDvL~%wN>&aXA)JmYW+2Inz zx6jzByC!YsR6})U!0vCq*+Nr|_l$K*84?tz5OY@JUQOKyELsLp?+^Bkuojdv-TE)D zOh0=MQLk4yK7=&&;lkBzHeW>bhIsfeq>HZlK7=4mZ+~N0+#b)yye8@J`DzX)YzL%W z#%R-Qch#Kq$Wk^#&w4B~G36|x(SaVl*X-U^3u4ipYmWRn&DpTydq(3RuP~S(kQmq4 zxw_=di!(j#LfW=sqnM}??2^5v7C!XGn+prRMslw_0M97zt0yM6`tE09Txm-OZu=;k z4TlD@HGgqE|#b}TF#}x4iF6fnRvyC?k%R#UeTB6YYO?sGL zNI6k<1>UQQD_vrcva1jQ0*DE8mmhVNFh)%nYs?F*MyZ_V?8+u%dj`Su z6o)AdnIurk5uNx{whB>V)84ar_1p6VJbMdSKYtiU&SS%b`}ev&8&Y01bUYS#w|&c9 z1KgyTc)qV={1OtJGVAHZHF+^%ohw{*ju7O@WJsx>eI!?pK*FP3NaW%sCqKrVNQ^an44_Cp{zoy61eYGxM;h-eK~kX3ei+iBQB7<>e8MmO>i! zMgdovOH)b7lXq~AwL>b=Eg|`p%Rj!dJw{gtTe0(ET=cv18BO!<+?bXvewYcC$fZ%} zp<0iNJUM6lxCmA|nhhViSktLY<$v_vV(X^!;L~EHfYmaZcIy_L=3>Gr4jpU& zo+d3ZJ%&a$QP=E~Vr2-C^oDi)D%t3R|G>c}XgqTp<3KNa7>(FT%V(}5=+j=vP5gz> zA3=)VV<)1@wxrepRy&mH(TP~t40gL4Wyq~!)ba!@K+#5;ZRRw~5Ge>90&QCo|9^@i z`h-Znk2+IfmK22^L&joM8~kzlgr6u^eFqNNsCnB^aJuh#A-BGEZ%ouIzM-#J@y9t% zjpXhOPcKpKMR+ebOLn-UDdgA8LtDrjVYehhsw}1}Q||mE+#O!VuJ5mf@)NbqTo)_i zRE|$auT%p!6%p59IM9x-_jF^p1t7Fi z*|!m;8_uy7F&fF)X%u#FT8zU#gJ`+4)sSPpp;N`h4roFR>0Pc_if7_!Tn`1^y+KSY z|tq4T+C}7B{?0=WCSx0|7kY_LgrV*6J?65N>6=EKZ;qb-EY7r<-cLBv43_dpH@j8i8uN&)vmtxEP1VtS zRwKZJQg&6xH;VRpyfcdG6Z1iX>7;Vg1Yp`5&N3h-7doOQx1{p7#(&X?8#$b5ybgtq zVmtX&Nt^kzPDf$%N-WAQD}os}ARhIT0~C8Ruz9drpZ}0eXvH!5G3a;)4wIgIPC--@gQ?U%593gc}x{(P9mk z@Gcn-v=~J7?6v+O`F~Ki>1XLIPD9@nEOA#S--Xw_{qqO1xd%fPnr@)dRBqxD$X^*2 z1#VTH3qLilm;o6a6No-yv{PAJm`H$*qE`6YYIDe5h106*qnRQ}(?daa!`=Jj)<@p~ zqiy{X5{KG1quxY;r$MOvIyRdOHhnc;aXyX>wQKrDpe&S?EPpkI7Y}4B;C7 zj)*_D2p4mJb|AC(6<4!|2+pHLO8+iZhRn6`z8(@AvE+NNkD@gYPF_S&8X+Kg&-0|u-ylpzit_4<<3~aiDy42}}f6kl)N`Z~e zaaYUyRuJJ*5OID)QjQTBFrNsqe~#~np?li{=ivN(Q@1-}T6b*hfJdV>v`m&k2QJ4U zDV>yy{CyJg>dLQb-WF;oat{6r=5HG)Kr?|;EMzVyS%3WIkdhSFI8PZ}0$2c|i9+xb zKwYG1E@3M|l6|umSxtCEuB;0txe<9W$R_n}g5@H6Hf0RD_wx}x!}H3}&(MeTd5nuy z=tY2RunMOM^*7fsvKKzt8_8GSXr0dwYR^xVbMQj1d3$}oss##D7?)>=IMG{u8iP&k zHB|drlz*YXdNQ+}<3MJ*sTX)XTdCREz?tdH=$fk^wn{F0C4BSK7!Mig@Ch{2Qh=jC ztTmyi)o3_Y(OoW@uB<|M#Tfb!2-*Sf)rngp%|t=@(QVx)obz#WxvQx@FhybBG*0-3 z$5yidEI{1IUR^{;Y2kei!iyE5w|sH5Yv%d5TE7Hkwc?%QQz{+pmQ_6s z7j;Y(LbB4lV#Q}mE58x@nU)H}E&0m}0j5r#iui}{V|xc&5moyjQK=_$_EV?Bs}3sX zOn>|iC3{7@{?b3sU~AoxMhT%iUd-J zQ~}g3bp(=_%r#1law_jI69>MI^9gzz)xQ12DH+Q;3lb8t$A3DE*jnb?OS8X6G)o|O zxUxC7mzrl>?yc}&k(Ipcl1E_H!}UXUSARy;Q+mVon6Eml*(B_OFK0RyikVU&AGa1=w?TT|!z;hB9}>8IAOC(~$$hvW0mD7<%s8AABhNXR*s zsxLVgsLybh=~EG+QCYjnKUzuEhMzfBIN{qg^+pJ-6=_I?@Ii#HVd^zFyIQKjo#K26 z__TH}JzjRKQix1Ki{pYrO5aCt#D5wjrm5R=X^AlCqv<^ z@{|6g^1G2N16=#0%$Q|--cK}`)@=H!^}%>=oZ`!D3v*B%D(ri7n%3SrmVe!mVP4sD z#O^g4buTTy<{E-!KDBpV7!%nBPnjTY_&_5iCURN@=gR5pJw4#Mb3da@^y^gO(AIH4 ze@>5wQ=GiaR6EVbboXc)>c1nP+BSmx=yGozD@J!o%j^n7=K9TtD5=Y*$UcT1WhiSg#4?%Nud!ZzQoUtGIP;!xWpzF zV_4AJ*2{14hUl&8x3>Rxu2&NmYs&uFhUq z8Isl7Dr$FIcZ%tM^!MD9$Eg!YM zT*{pWunUi7w1Jp&Fys!TYV&W$30>XNzg|nd4hv73jGa|hJQC^>EE<`@&$$i#(FW%6 z6OnAne|q=?l;=T-?ksq+$(!`~y6&Rcw{sNT;jZswEW1LNRv0~G6PwdpvW>tBe9R?h z7(ZF>$y|yi(R>6L&VM2wHGF4_%vpdFdoO2K)0Gwr`#8V|q~5#l-))S@{Yr|L^4WbK z+m-yu0_V!uiC4WiEslr~?44y57~cQhT?o;BU7UZbWV}_6Hq+p3PnCn%)nKxCuwz#1 zRoD$T!h`49o!n}5y5z9i%+t)VPv*~~d%Cu=6}l9>;+13T&VNIo;0*NA5Jyaon;P55n+rXMc~)kaBJn zX!r$FDYKHN@YbM_2#3b>BQfE{subRDAW4|K! zRX3sLB#z^@#4o1=Bx`;3RZ~~JpC)tRHy^$b=E=+%_WL8f4q3Bt{lw>a}x_IR@Ck?YzY0iFll7B z_cHzw)|Dm9^Mve#_r8`W;V~xFs$1l;i0(vu<@?#@jj<5Is8+bagsD`b+TnbCezEze z!N-qUVt*s-+Wk7}4Go`HhBo!0!fVa4^72L_&iN3ur{gyZwODQNWz)qz#(ejwEpD;p z(1{A`*cP7N*4Kh-ltd?iw?&pzbb1VJf;t5=V=L-b$8?7VAP*__Mz&n7_vUn`O`Xh( zdD66#&R)l#uMFBh=Mzk~(lut93)cC@BGCPs+<#qk4@w`p^#~VrJ5^|Uy>G7SQL+Fx zl|btJ)z81 zp@&G(!FS9tEN#;-_a0sYjrZNh4~on7k{h_zqq>gg4n$OLm8-bGFU_y-hFg1tiXJbD z<$s+kEF{<8fXH@L?97)e!~({ejZ=Z}ri1T$B&_>JA3{bJcl%sN(?dDgJkJ@6arF%M zz(ew;Z968$;WK-Fm3-1khjmnT*XY+jsNQ-7MR#oFcT`z9Rt%|I=*z60GECu`b#9nQ z{va1m?n-biCC#&L;$Wl^Q>aEbAa^fJAAfho?IWxTzc;2Agfq#f6%s=uN*6(onWf9o zl1;?_oUyF|y6?n>TW=l1lVGW;?1W2USwb3RtXW)d6TD(+g>DBw(tjnfx#uWVpO_C` zxxGjV-S4UO)a>YK7TC!C)E|+uHZ(Eze8t+HoD75AKKiPjVn8D7|6}7EwnPE69DiE2 zZQHhO+qP}nwr$(C&0DtZs(x?KJ?c-$Bw5KiXDiz&G_e$4fsP8OZT-|;Ykg6%hqI+Z zA~H9Pr1yYqatiXu=anK2IP6BOp0-^)EKb@9RGhefpkTWsOvt@l7ySDbN>P|K`_h^^ zvOjTo+}_xva3kJIkkCmR+3QX*4S(Edq~=;}q2cguIptZqtVJfq3Z(!SQj8Ja$sI0C zvlNiyi`GyPq0#Sb9|}MGMA)!u9c%Dvt;u+WK?6s2>yCcz?FT>wpLBC+0?!YV-GV1^#M?`uuP>BpQuejB{lVd%9 z{-cdIDUoUuux`*Np)x{TGk;kQKv~J-OBL%!A}6V%-H&tOolh1%kv5Dpt+)Yt57Xt( z@tYD1le2*wL1hP8<$XojX!kX3?KC^Z0S}BjjrP}+q5}zY%r(xYQvl38h3$;-UOy#zPg zD%yB+gsMgqEzy)1@xku!y@yhRIzBj=c^Z4bL#!11sO`sel7GF~!=1&2t*f`$&Oz=T zwn})3WD=C*D}Gb@Vmo=ztPLVR{!Pjn!2>4WmuagV4q)l}sMl`vX+^gLf{jNc~ zNET;=SIbE`#eb|RvnqEJ$cSjUX@L$Y<}(1wO~gN~JS%xV?TF;x75(!*GvD-|@G8m1 zD4U9`U!UWjP#zdfQ12K8V!k$B3wrPU?bPg|^~QYGe~fhKFJnGHp)O7n(TwBp$)MC| zXkbAZug4@jK^^;EKRYj8`v|kf)*>rA7Gl;nJ`Qafcz;J#0d%C?7<5f*4b`R~ptrPs zxI z<@r2njemj6FJF`XbU+>v0{eOB7oL3J!#RTke?fy4!SXYg@Rzlkll9DjjF>P3K$fIU z5A|6vzZm$X^f;Di%eUkqzLbzC6(zovB|5s|`IhFt%({C;*4VyeJ>K5XF8)DL6|ffZ zt&d((SvSQb2M|w#ai&5c^-fs<1Kke;?YC2bA%FOR;-wQ3pnhvwE23|<3o!)IF9%gY zuNwg}kH%M!D1(9hc87<^&Maa~!yRTd+7Ubr^XD&~CUA2Ta?TsY++B_3fFzI|Q&;4s5wpVEw5g&H^uH|MAHFB!K9^lZ_a zRYB0S`D}M+)Ok6ahnT2tyTdu{F)|c_1*#WeHX{Q5>4C(-LarX+9^US>lBKuFnXuCN zvEbP%QaKgtW#p5G1Ivb23dK{iIR+QBw0~($d%na)X69otg{6~^$TFCsI*rRVZ468B zv1Mk82#3q5+BIZJ1i&0;eW@fS;o%{{H8edR)#K{dJ)yix=Iyc0FEONlK!qSY))~M}LWYhY zvdG=2tC)SOZW^D$qOn#P0)i0$+B@j5NyHt}iRT;Iy*qS)Y>soHbhFc;Mv6|zNjYC@ z1BM>ZN|+TMfTeN%+8od{a+-XJEq_FiN-d#Ti*U8%z{UIyYxb6Uii z6f$+%Ubiebl>@IXJ8A_&DJiCl6tVVPf&pOEfy z&JP|;mKJWrmrcdPgf&xLLey=i9h9uDj&YzCo{f8xs{tg2nAJtjGnh=1w^t~qB@#Ap=r zsP}uYyVqjF-|EDm49l=_Y=5ojl}a|s1dQj7V@+pow3`5?0K$~1ln|{^CRPqfc2U}j zux=SU8d3b>@LI$np*HcV66oaS?sMY6OwZ-R{qJ1yYp8gNSiW$inGUaHYw@uh{P+ZW zXGWK2==g`gqeaw=d4F^2cO=dCI)}gi(;YIxFby~0$pjDh7R;h%?u9bNG6WBE{s~o; z+KGLTd6Y2ul})i+Ygt9(P_-D}f`Y!~U#Mu7Y20%atNJ_Bzlku5^@ZK;XZ<)Nifgbw zr4BXCl2V01bPX@fJxPqrz9BDuA3(}hS~4I{sSYMtKw6@{f`9q99{Ab|YF_hYG+7lH zOc$*_nb(hai=2Otlw~J7kqa9vrCO?v`CPT zu-KlrsIylFOxg?XD0|7rW`mEJC3`ix!+pg8u8__$=+nh?WsVNhqK`Ew^A|RXBlMkZ zl_P8@NllUSgTSi(|Vm{Wr!YG?}IBMtYzg0I;5d(2ZjD)|T1C(&b+ z_ex4*hEZ${?@+n>4)QeEkfO+cEu9WfQw2kt=#UPEV-;h{joPI@Ab#~;E`QF`Gk4)1TqnU-y5qPUB;hDU zF)m2b7wFo1ze>l3w2403P`gO_JkRL}r`obZtE0H{=M>oNL}>+xP$h}BHj4?qofw8P ztGaOy4U3;%4$a`=tx2Jg{-e6Kq{sQj>l82l2;EefwjKwPclOJ2h}4S`O;URyvl1#E z^Ot1tD1Yq-rrY|1R<`Mx2E>9UA4w&DEv`!P*e9*oH7wrH_Lh=)>uKbVn`GqmYs#w? zK&>U+U_f7!3#VdegYyE{@`lC1*Bh9<1m6LcDXVvhjcS5p$L~eWHpt~yqih=?B(!Hv zR@6kn77vJ9V41DI8T_ChieL(o52`g zz851*WVFJo_rsWxUT;Z+PBfyG$>4}YV$b0sC5Jc|7k(Z_4y1D>JWuu`g zr6;$Dc*Ef*?80M^T&_xINrA9R+a_sLf@|*q@NpxegF`1Q$8sECF%+)pqmML!7vGU( zNPmQ~189BLxgW#}GrD3k;wihj(`xmwsqzAtTpT9qM=B-gMCrJb(r?&IJ7k#^^NnMn z`oE}Yly6Z3I*9A%jlkhZzkrfJ%qoT4obFjLPZ?;S2FIs_Oml8J#WIrOjK`yM?b|Ug z18XiW<0y6gh%gSBEA9RxaPn%<+J!fL>VM?cdwcIgHmnHvlqJNK60hsu$bG!`yMJY~ zA}sK^dbRczMhkUn#bdw6*@v+VZ`P!x!>xJ&Jl#)ZYC5s2)Uc+!+F&ZBn&V{y#7l2o zzKv5nzo(kji(8>-oko*&?rru;^k7@KKs-2oaiR$esIV5~AgRyaJ|nkqPUAx{8-FjE z{8^q-aqnPGYzQiu)~hn1^+>nGc0Q}?*T^;w$QhqNp%%= zB~*%$im)@Bj;q3ogSuj3pyyMA)s!;;H>&2%VU6Fmv@ak2R?EPS!gBvjjEi z(=RP-GDl-Xh3h06Mrc>do(A?Zon~~`niUUz_~q|N;cJ8@7w_z*1(?iJ*doT&1QQTz zVAzE*9?2D&nUYNG_|Fl_i%TCc9Ordi>d&NVuw(1G<_%e#)jo)AV2`cDQh%y(0MODp(&iep1o|%B}(r6W}fqH_d*Cjk&D*BO``Q!)&qmvMm z`#c5#GXvnSGFitm_Ta%_9S((@*23#O9uLea7jYUhI{I7bgZdAN27mHQ1Hs*eb1F_0 zPx#w(igWKE#Cx2F|2XWA`+@-zFgMegIj_5+e^!!>lq;G2#n}$dFASd%}mR+#b zNq)FOQ74w!{3`lNlXs8rGFAz(=*7;h)e&))=Mx8GZwOA&qJJ>i0oeRyBO#i)XIjEX zV;}Rwh~ZOkk^#3b25V|77u^htQr0Lj(S~DG7YRU-=Qr1X_kX-l15*Wee)$~SN^yuAfr}(lY{yyp~<<=9MO^esW6l@s8mV|8c9b#-4CS7IsO_ps+lexA#7M}a`QdjX0 zZ2-$t;qf8W4(h=(zN`mK!@-}mZR>zhOcc%3_g=F%v;t8gB%m8 z6dmm;c7N*wnF3s|i4B!a-q{A(B)YkAPZ~!d)@$r#H+$iL(;&U5zCJE`YAev~db#I> zjFA+bEKi0ZQ&xu24I8z_IGvdVKSV16e2}uHfsH+by@_@5m&haUvP2o|DRhUlPz$PJ z*-30w%`#J-J0mTRubBQT4GTlYj>?faRWY-slz-eX#SKJU#$5=|O)ZX6ovK^l!!9q` zJBA~>ZH&;GfnL30MdZX=P&Q$zW$c)4bZP*`E_;FwQqA;Nb5SZfUjn>MXlPr}GaXk3qgpNk(= z4S(S&9%3Rdnq2B2pfPT?Ov^L%8n#V9p3K+W13P+HQa>1>wEa3e1+yFbm7}{hyR(cr z_5ZZ3n#tmkxjiZRn(=Kk^n}brj8P8{N?Id!)kZ(1&zrIh+WWA!h&P`$w~+BP%=g&X zMqY-ULR+kHEdY-)8GoM<=m}^o^Nb6SuNo~W@;jNit*&lyPsDJB zdF7vlo}pf4+AR%(@${(|qEA^-dU)F`Cyv=;+}$lNC+{G?K#)Wo*1A=t+;N`Hbl5JY zN@`2|!gbaD{?RqC+Rchk_pdA7Q;1kH0F!Awc;%X4F3To6lQTx5@Q^2~@0L^-;(z;? zDRc=FCW7G0c>k?Y4lXKUU{`jpQ&FR4ne$pjR(nx}#zAHxvIn3?rbZv3MZq3#CS$J? zvM78SHELPr(81AbBnmsH3HP!Aq)wf(`+ztWtZ!j4-Z64*zQpzmjr%J?GBDD@q@sYf zJ7Z)a5o|Wi6pMVlaNmTu$PmSYk$=Tf-1Y@Q>Z5XtHh1w`JFNMG0nC4rJzwM|+m|^O zu(j_vlN>U#pc5~GG8JhxDN0%VI=@*7?Z`@h#OXmyj~;i*SfS0N4JXBu(jHqflPT)iaLO`U#}hoqp62`Ptx41YIeVKg3o zh&GiJ9Z}Bj>h>}o%y=XMVsB35mR<9pr6f8u<>Mde zGJUM#u3_c0S|D`dL>U{j0P?J}M#URtw!V3*eV-iqx%<|wgEJ>49?}@(E;+$PH(;iv zyh}9b@qLaiA6>ogG=2?F(|^&^9a+l0wmY+!SNrS!Rx{Q-Vw-gVxh&;`CVP7P)PRHK zBLg$lX0xuaUxieSh_w)&O%hJ{fSvTWIlyB9Af=it30^`x6rnFF(IrFI_=h8OpC}?$ z#%JRlmc72hx~JENu?sTnpN*ILW0|*+7@9BWsm}!EcIp|=mi45$btq|=woKfp za))3!QS~N`7m709KA9g2V16&s8%0w0f4IaW-im@*&w6RjpBl|Bpn@qkrGLL^6izIN5wJF@c?kqR4IIN0F?HJyGJDgcrL;5107A3d+VopfEb_!?SF+|obKH>CN z-3yV=`^9sE1pctKP~ylW>Or(|(YfBN;81+~&^;PJ56&wxPg0oZ;ZziL)@-iBmO zilaOnXl%+K@hnQhR)|O*?escktrqs1s5iT}5r4;ebx9jCvG z%gZR3YRrr<_~hyqX`V$8k5v^xK>WxNAc~AoMxTJ`04Pc?myO2H1Z9D-Qp0^J&HuOd z+?0_-aO|SQPDbm_Nh{Q3Bdy%8fXru8-BSQo@oQNGUQ@fd{yLOGTtPf~^rK#bYT2_2 zsDHPuQ1G~i-BJk7hjg^b)PiTLZTS9oEEjm>9WT>M`k|xlMgf(WD z5n=#0%h8&EMA*2)U{?2=boAAJjBZR62U<*`k|wO-!eS=fHqhN`QA)b&DRe;I>z-et$rA z&qwvn6rM=gYkoC2wF2~jR~%1t1uwvxod!(D{>O)aZI@3wzKb9coS#6euJehU4;7w| zHFCe|OCQ-rmEctO=J~A3pu<(mQ1oE%Rym=;jo;8?I8HP(#CxbC8sx+sKx=scF{93| z*|3kFyKjvCys0S(wGulJ5JV#T_J0JTgU;KEA^TgHLn*}?SBaB#+=ZP9q%>okaz;9s zdk^p1A{n^tKz}PAs({xq zVUva3`*bk*X$u;C;D6iDX9Vj8dnXWg7!FEW56|$4d^Mc0 zNbmr~qPx(#SW-$Z;15w7<$serc;_;qhJ#N;%*B2}FB><|C*J^>bZ(){uQSk6z@m7J zeE7ENVh19Cn#M-UmPqrI6P`8a!G9S3js6;6(>3$&Yl&ogM$u|N_+7&Up$~SC!SSib z<1c`>AK^pGTB-}>sjsjH`HQjPj#?i}b&Lq-D2X1Y{hz8E=vR(*Ab*4tAm6$UVthh! z+Kr0Xs~p9%Cf(lz-JFW9Sqdx3OIseo{HZF-XOqzj#BlCN;k)E>dQWF~P7$X}b8j`p z`IH(Sb#cTNOQK;3$jE@`oRToio#XCSS$u*$qR8IUC~m-@0U?fYtyH4wX>NXy7y!`> z#&L?%MF~Ao^+6JAi+|50Ma9NXTfgg#Cg*8Fsj7d=1y?s%OPc|D%1O%v&eI!*6r}?_ zbWwcMlO~2tT!07I_$GV|3Ca|jrAr?2sY^7q*!thDf$H#0UUYx_%x}zq9Q=+7PY7?W zv;{R@hWk3I1oVzr>Lg zUTRJS_hj>v2IqNMl31}qaL#yvJ=GEzk9WOyg_#nji=I}ut*=0fUuFG*#{B%cZa&@u z8`;Y2sc7P70$@cU{(ZNOF47-vo}wPpU^B_s-Xn#RLDuj&tFsbTcOkZY9>jm&z)MW^ zVon7t6Vj)ix_>TWM7uO^f*sTIgowMQm}jKRo^!4ohD5G#OR?|?E4`W@iYMRUVRpQI z+(psEWbAtdB_L&Abynz){S&|12VM+@#eK|}`hNLMawN;szWUly&XI>*1v-mnYm9cB zvu>G3B7GI>(L2@19E@4Xwov8i$NDa8<|$w@F1s+(xqn;JZd-8tt6CW(u%$Y5xt%~F zLAqCb$!*2;=v(n!D*EX3ctst&y^VrXWSx@NzVsz5U6#%o7l(;<0sbtoI2~3n)?YM! zoi%2^8mlUGNjq@a?>|d(3Ly(8802@m0BJliv*qJXRiiy!wRnKU;_ZbmsaXXi+qIjc zWA_Z)1An-#xzKJICSuzzRWQ{$LADcd0rH^fD_!jte!_sy#>4pRfx)-Ay;!4 zk}2zm-B9WyrBy=>gLD6u^TTWCJIck?)dJvVEc@B9&0NVdprLM%fQ*p<*$J536M9S~g#7)G#3uxkP^y7F9T(JX z6n|4~A=R%4Tt;7IoSYf)^`24X!XM^JpB(9$f$}Wi6L(Ow=lM=7lNyTUL|KZ`H;a1Bx{nKg5P&Wmr?Oi);3!?R1{(qHi ztytZ~2&LP{C)w*r6U zIBQw+rbP%-T6h|*k6aV~mqdUcy?-VNeqCXrZBCv``D1Cv90jV;z78DZR0H{z4e@Q? zqejBc>I^et?9SI119r;AY}AaFv+pCABE;UxOZR5c>uz_85j6v% zYpKK4BL`G=d89-jUbk@Y;(Y^wQE0^4GMjVW7feCxvJf`;6r3ktKo@!mMNjXEb=9V+ zHD99pM-LlpfCV$TCcUC11}x?h`})WdtDG*+cCXG0pdz20iBm^c6mOOYu$5xLD`M+cX23&V_`9-+D8m{ zfZESN=-=6lW`F1>-T3obUIFeF!Wyz`YP}04H-FCWSI8_+g%7dGWq&ZGB}*}!Ua4Zr zi?c#B2I>M8DRaaPGLtCIsSTcQo{+ev6y>VN3sKF`-DW&7ERpzwEFm#)m<3=v%-HRFD8c`mnoxUe||+buwa2ld#edmYQ9b_WQ+FlHOy zZ+;s8S{Eio=s;?{n$~75_jg3c~wBTq8b`b+76NP8@}g+{ePn$d^9ke_B3AtO9znd zYx8awQ7h=V$a$NN!8zp88pUFRv%o761|_RGb^-f}`3lM3fzN7^;gF{Q zD;qb@36aJ6#>ES>ko-fKj=XT?6xmi7|_cAEGeeFv?t#1&2FfpkQA z85%`8WPjB?GNb{Bu$%Z{d0=*0%JbRI$@fRm-w}GSF>i|IyFuMI{1fpRKZ@t*eYY16 z`!<2PNb_5`ke4*ED24z{Tl|UesT0a?8#2nUq~D&&fsr3cDuM3JnATOKj_YH(zbr9> zCVcBZHgS~r>#{ws{Bd^jd!`nlDjIERMuJl*3RUMK!2XlyFS3iS0W-8VS_Kj)Ic_b3ly0 z!=Dtb+Sq?ugxxzj6TY>o-`4Y*Ryr#`2#!~)lbPytS0qmcVVL)2mOC|3AeK5lnRJl^ z{nvfAS;J~C8br6hz_1WNi7QGqH=_%;jPIOno$m%vZo{Gp47pJ`eU?jYuV9i;@J_lk zNbL@7VpcBy5(A1YQilQgWyBJu4X0H)9NH%b9F%`u_9qVPW@{9G3Gf>ZCH$;WT}j#a z@D^1GAYD>H;wnx>et}bPBgDjpDaP4*O?Z7cHZkL>2-b}g6>4Rq z`cwM@zG@5;h4S0p8qnnHfoo=S%Y(TH`ST(6#xtp`S_aA11vW#a&Y7PLg}S^$-YoKTvp zSjNw{2Qo7hrW#{&u^;(0GiAp_mXr2yErddC`IdQ?dykIM4tTAXS%VrGu+kMya;eaP z&iR~a($g~j)hd&nfc=x#8)1j%cRX3CjnaRFMko5vx49G(dKAc0u-pF4Dh#Kq8KP}G z@={wemwU)!5=*IVZjB|8;08N1>(w=to*2^(NdAt(=XeYdo`*_t?i-?De4G(1YI+EA zrOrHV9B@BTAHS$om{S6ozyVsh0ahU+eRD+%+9#r*Up(dc)N)1sMMM6FndCYI;0}&%CQ*h zAs76Iz2}g$940rjaV4pIA=Z*o>ZgD1J4^Gjb*ITS3X)LIwjBo)tmhJ%KCASM;{jc3 z>?7+M%EIyg&?w6_C$Fo_{3=2Q3mwZT9&tw@5Kn4Fv`;r`u-H^^k;VriNE_iy*e#tt zm3^N82aqIzf6B{!uIb9aL~jV?VP#Nq_N5qkSs=t&43f<~=lEqPB0e1TqbPs9k`&QG zF?B=U3a7ojw+16M*2{^w7$(_`sV{YhO`n>XwN&?(5QpA|qjp+Vynov4fYYss$xnL~yZ`@G3_JA+&Pg#Mzs{ zoG@=y3A}j+v-DXSZ&?GzFb01EgRQ(s|;wvx)}_IP=(-KD;FZid;X1rYIe|6g6sj( zzwX$mhI8&-b^jSUbj`U80Z61%lc!9v#J_kH;X)U5e~E(BT!p`D!U%sU6S2PtT2M)C z$l*by!rPHoUt{^S+{@l%X6&cemHur|;Q``q63=X|09Blf{D+hMC^u zWSMvhpEf?|zexq~d|mb-P6JQ=%>O%9ge{3|Zt#XO;eObd7S*-&kd#T;=A%!#%iw|e z%SmR<4}{K!N7Yl?=Iz31Z~1B6IQ%lCbyk_5#lYo%#otQk;?;jM{!&4v76NMmH{=7? z!g*7unRSv#0^gj#-7s>Ct5%|Ni&+=eAt;vj;1Y5sku5l`MjY_y?BeI#UKpJ#naBR8 zXYYIs-hB2C;O5XFXi3TS95-iQMu!dd75!zXCZm<7E%Tx-rc6*tB^m1AHM4*uTK|?W z{$N9h1K{g#vW0(o>WnH^umqZ^U5vDHHmALD%)G`fESIqK6GzcLI_4~}z4%KD6MYq* zfcBJ9hCmGEE9@0_ef(A*P#e20_$-gR!+K!;okzp^kr|C-n@5pJVa|e*o3@?A0LJ$z zyrev*yR_g!YBmb&G1WP_j3%hAe{Ne5Dxfe5hReV7iR&jQXNoE1wgbJOMUeUBKu4fP zsTNYwDE6T3!@y8_|#8*Snz~Nli00X+Ciu?Of{tcwq<4FC|2o`q2d;eQCEJC zms$zurc!^ze*0aU9-Vhw&+`s3){?%xK(!B-l2> zwEBG)G&^sdt*@!1k1a6J`LyRlcRbQ1U>jL*J^nfUN)R0)Y28N#Q|f$m3Cfe>?(c*P z|3orm8`{}H56UU3v(OVuW#*HV-UDX!0tlx}=?+RjWz0g-+=KKV2*QB7F2$_oQ01CyIrbQ&{{BCz zZrCak<7<}}DJN+I(7wB;PtoFrARoi(&)jX0p$QBjm5_7U&;-yH+w+ynu6eC)a;<+p zAR#4|ioCF707*E$*cJmxLR;ht8)t@X1Zw^Qt%SJc-P`#5aB~JG*lvKC6 zQk87OQE>&FQ*M&?Ts8}=vJ&2bY~-OI;B$Kpwa(}X=k+QF7&)ePw&#${+TBcmGAWKw z{l}Pnmvmj6)iYRF4QWkUBq(vJfjxhzWF;Xg#tQP&gKpvKtu|K~&P$ak%j{Vt#C&1v zDRDO%EHa-5D5X>hAOF|fYByOWyn@Ymi|=g~1+igc3$4XyH;LYLbl861u?rNjUWOcO z7wv;2!Au+aa}-W6PtQ>&P>8Cw);yal3|(uffWiI>^5X|RBd)Yh88J_(N~VACjqJkY zT)#|TSOC!_fRfgA=BU(bJv@wlvPBidS&11NUOza&%U8t25516Y+0DcDi9H+=1ctWw z-Y(;z94;HAVln4j3;o)*W~;iw|MLM35YRnm)^L0EC8|SDzhB6N_U>?5n#V{GTfTVc zn0!w|PWmPq`LFxwB z>{Qy^pOr9bT`=1&uIBs!{j$Ntw+bF6se7X=vZSzLb_jYkJLi$%QGi*`I3^v9Xl=(5 z%!_x?hY`BGnm9ehlyrm}Cm&ruf6g2nx?bt{T?T8jPFB61CB)ZS!^wX*n9&S}t=Ds* z5T5TJ@a1dy_+=EFHx6+R1RR*g?ct+3lPoOI(yf*e=NUEAV$^buWV3dbC~Y~JPXR5>T=$kGhLT5{_AW-g!`8)Pmx%W3py=Jd{aDV zNVOo*MfIkYS5?{o0wKJAush%?L6g=qS1XG9OoES-S}wKGq{V+#71IxUgZPfx$a>>f za)B;GRZPB}21pi56Ye)!pM#SHpV=KU)&y#q{IejPNob{H%0fefjmooMO+^C<${#7*rHE-J|@J2<(n&dK(F)@Kq}lApIt zDb}75HGoriF!Fz-D!umjGM2|SIYiJ=W-@YP$3I8z(sMrJZlE7y#7<6uNY9@An||_! zcurykgI_BGZa7$(v@SawN1UGq<0_AZvyEtEV%y2ZG6ov6g3ciauk`0*tEBflNCoDo zNT-uiT_|+r%obp@_*#0zGPV>w@{A=_nKjW6hl6d78K!?pb?8(@9FU88B|Du}#Lf z0{Q`uO<8~}5;vyF5Zjk<@86fC+o!8Y(H@1+EOPym_kiqU^%?>|d3kaj)u;YuCbnSg zk+p0XM$_D?%^^HYzaF>OL$atf_YN_2w7c?oBBOtt(8XD>aEqVP6+AP4%bkU*AH-b4 z_BB&{YuMqp^S#*R0i>URIzUkWq))fhVTT@1|HVlka{ju;fFEgiWUrjDdt*eTp1B!P z%he&HTYEcR=wbyo#XdLc|C_$JQOjNmI}uLtU?qEK#C?+ouew#$TlVO$U@*c9a_x+j zoo#;$P0gbtZ{KxP^j%)~F9S7ctfR~>KVq=l$Rjuin5xnu4i?sDaqjP5^y-GYco@nB z0ed2aloiw!4ewTyW~FCR?@}&#i~AHqYF^q+h7)Mvp$F5}DX4x=?MwirN@FGWcMB4E z5OC7!ZjXPrG{Jo*JI&5V>V?&?DwI;f@6Lb6wi$W=%MiaP{8w&q&_w_jWnKo(@yZ{y zKn%uj9<7G8Ki!zEbVEh7(jP?*Z&tU>G~JqN)zt^{?#<#AX|9Q!iA96yo{~oruGmPu9mVJUzK* zYB5)O7(6V0m(pM*Zb5@n9sv#u7;gF7hFF$q8zXg3h&3A)Rt2rP^(1@V86LVY2{w0Lg2T4u$;*K~5M?SB2wjOvHdix@*!o~)pdtQ4l62#6tw z`-O*o58*e?==Jj*Rc)Q%s;WA+xGfc;u2(-lUOHd2e8wKo8uOD@1=WY+tiP=m(Sb2C zu;fu3;E5&obeU+w2B@8dhnRmci*$kJaBHWlEFJ{`{55xnCEv=3ygng@%$bfx=Dm7+ z3>%(YczB2e-n*v7D_1D^aG18@qD-+ks2ZHQ$7ujq>X;Q>K9$vI>IV2yO{rHa(Vxc? zx5v90+LBzDghQIn7F6?<=gOB8HRtp49*w{aowR4 zIr_a9(}ueKE6iHYHVLOGw9B<63Pa!6_0Qb1@*i;cEi&G9_w-9bA%zf)QjRM+YdAC| z8-NsD_3teZmM_Cw&iK3AZ_4j-ugsA0OJhN->J(#C4z}9p2&Fb?bv@>ZET(oq-do7W!;~D zPdmXe3rC6@EmFR1Ez2$z;1zsu!TyHd(fi2~`Y0(ykDk_#WxR@vw&z4LsT5`O@+;R< zOgpqDX&U;J_W98mK$lCQZcMCkO-2$jM|#?;HnQ)%T@7>~l0%@o%9*?G+W-Y7?-fn2 zu&y>I7aC6Z+_Qg)huvKl*zT+5Dtpq_hh(kiLR)RBqiYLd@43q?j1|LCCdL+tm~u*@ z*Z0`#Cc~oI68Z(6x&;z%PcstwK-Gg_^*+p;@m`Dn#d#0mZ4bX1WlO&mb#gJkKeW%o zVo~4XVTZRG@@wvfh9WdozPnCx=C@)0_@Bi-a%HD1pND^Q@m%R;Bm(%$A7{U=x-_{0 zfhLY$+f|-p{R9KOcD3DHO5XYttbji)s?43rmY7RYY|0O4c2W+L^q4PDIaEAPgkBRr zyf{u0eJ-FiZ43c06c7HHe>guCg@5US3{~t&!lJ6jvka+vG(lg!N|Gl-`_WLBFZ)PEJ8x<5p)bI3?h~ z9+v36Crv9kTUb2R#Yjr~`3q<_ftH%j&ir!t^%EGY;gia_Qxu4Uipv<7#Twb)a*#2% z14u$c1sLTOK!mnemn)H2LWwV@?H!+1gJZoR>(qa)hSfmmrA#@-tOenc!&0e`f;{64 zy(_4o1O)wN0@GIOS(TJqN!4ysOzW2i3Qz}`vmjUxcXBfR-2d4jZe6YNL3xR>?bz8& zlYg6UEsYV;8xcP04w?%KmmA#dhaN=w1i#^jw7dCse#n!r99&@>+$kh=8m7o6U+ zlcj%{a@l-aavS*u8DSp7HH+{1)FK%o{ALnw^T`}y+Cfc5`k&pBjf)U!qgwS*L zJn3cC9fyAvP>kV`xnGxI(jas=2gQcrN7-KZkzDc$_*xI`(y5bQpuLHl!1n1D$2x%A zomyGWEweuP-g5!M081jXxtN}(72L&HKNf$c^z|P{I3jA-@ihkQ{5#U5TFwT^&NRt- z{}`8JtGorNkDcw6x|F)OJN5nJ*5-`-}D^XDw+)C}HFf(3^_I0aSlF zV4_iyEM>?qlle&lGBXN!6=i2#GFtHu(xVYrCDKN6N~*rLyQ|^ttq9!`*0RCJrL3_E@mgW8>{6rC zNRZxkf=R{eXMArIpm!cJyF}a?AY^}Ecz3#}pnQZwSaG8$H!lbB#aE!OV-|=JN=dFi zH264qDS_J7+%qm8f_$VFgNyPo7A+N#H(|9l86PI*IlsI?7RJ}7QYM}?gNgo|1{yu}(v##r1s}_G1+_K%J~gFd!xCQDT3M#pi?1 ziEZ?$WoK%Muq+9vr1IDd3me^RV!IlujpaDGsTT-~X){QeaysZh&a)b&y7~9^Rit8N zE0gc{64MNWV7R3jCpe$H@hX2j&yov#s*`2x)Whp6gPghp?iYZCY{N2-J>=%cM(GxHJ#n!z!q1JzJ0-%jCk=%>tu^DQ86bG(xRrjhjb}2ei zXKUh=p}*+Ql~YdZU+{l1p;xqCBwc*?^RMtRfrx+5I1#O8xe&Kuv%MIT!6i5YHh$)ec$3}m!Uwq0CG`M?%+g=zmgPtkHeSx{w*ZHWu?~E28H8DY znx^UA9cTY>NfXYqxhFdBpFGN{Y5EQX<~`9?-kCnF>T%5CSonzjZUHj4j73f@z^u)q zBovTRksl|dU_gH$_tK$#TOI^5SS@ghy30(J4iaMotZfoJMJ0;WSW)y3N;Aw1A|F#) zPa7xkbXQ9tT3uLgX!q2_12^BY{K->P1O(vi=r96I>Vyy5;Ix&3xc!2FJ^Xz&vAk|V3B;6Lp<>R+Mm*N-1bRy}Zx7lZ8Z+Cc4H>18w)giu&n4%-w z4Q=Qb%`I-KK^R{IC_!)=jXw)QBP;-`Op|PdP*gv%s#g8rAG$I5Rr}8ZAw`kkr1DJ8 zXvthr{y%>eNFP+$#b|P@N*loB#?*h-3~?(vNj9aLX+x%AiyNW7(x(naCSS(4K|()H z&~BJ&a+rMzUMu0eYgRP}_iL#Iu}GqZU2s_B17;!|tz=;bEx%28#NWIOS3<~vY2jcj$u0oZ)5K3IT{j_QLK3DR5_zzI0JJ0qoSZ_4 z*T;XD@s#@${s?08EVI6?oW9dl3T$xg(Y=cTIw)F(dwVdapR`^pw+ z)Q%ZtwQQ7V>gA5(&hlAiwm%FYM$XTUJBg=F_AI>3NnOTW8D~Nwc#G|I9g5W)_5xrO zr=<=Bz6<`ab5vEioqOollYC)WP=5^6lF)xN!2J|!;bR1Z7Pm87?#+l1^i9MXNN5_q z^bhfogig#3IT6)+K0wG>7aoGxw15rqIVdQ01RZDF*rcZKgQ6Efgj2akln|gOi-F8g zmHDrQs`|(~egIAMl!*j%ebb;(-Ll#npcu)O*4uz#j0%PyWM)^Nqly=-Q7!2iStEa8 zY1`QPnhg9cgn~8q%qpbW&d@ly`i2CeN-EWNQlriGv?VB`gIRtIv_VdPlYb%E#~Z@! zF19%jA!D8N=bi9X)_^MHv%!6}CFNA{fk3gJ#+><5n{OqzV4o594Q>XxmQB}@cA%q_ z&hG%j_26X~Lp$AZS`I=vdU*0ZAd-K8NKs00f(mkM_td~@D%*Kp24K<=WNL=1Jr=D} z(lm^yIJ0?eQFBf|EXG&rztu&i8t*AW|6kYu3(o#U1F;}Y3o-Ml5}(}!`JjI5ZL|qG zO8uDXF!beOJVlbA#bB4)F3~}HxMRX$Ux-#;L*JgbssdtmSUL}}t_ zA->Z6-ga*9tnr2Uu7z+6ThE=;iE}fUOrK)f10oBb?bL9UTqj+l%4z781R#hph*5s9 zD4H~WI-a+8V;Gu|^JwFjwmtL#+ltuuRrXwnVL!RxD>*V&_eutL7DM+Sr#5C}&bn#Q z1EZ;9DaoKgr$e?vyeWV37|(kArXKG@_(*(9i5=@U67hPvv`*eRwFUM({;xP$c#Ex!3 zw0EfG;OS=X9XF9U^iq#4TeEOb zT8vj~7#;)>gHHF)Aj_zLKj=#PpkQUIB1dcHf!89Pyb~de_z;aY_t4Eu+JdkzSF#(l zz9S_E9mO!!z&w8wDBdH68p92ORhRBpIH(*;q5_VYE0>U(`T}XfsvO4q?XrVGl{&c< zllmmO5eBP)X_9bHO|-yje%Rx9Mzc6g$5JkMDLTrZ;Pt1M^p_ogw!jGPgL-PG!&^eJ zcLXs!Wd`z#yturiHh2TB!9(ZXvXa0@{+Am6KiUJKv=M*b2fcmn#;L0;McHm9I})|9 zG*WhXj~a-Os^)AAG%3U8`e2hwNoia@QF;jB4hZZluLx+xTF7|haq$AhKhda$Tb~>+ z=n+%xFw#>N=CZDRreTwin6G zW;fyOZKHpUSdsn@NbP(bBy_!T{g86~WHSQtDXNLL^hY7RyQlZImi0^D;S*gmX+6{BcFSZ`~Y{9IF zUU)iD8GtDNEYKmM+*P({cjkv87IDM3!^sbEQyzbfJ}!kmIeozq@~zkcS%aNj^xNQCB)%&JC>_y!AE+&q9ce1tiG5YqLNyqL#6yxsuEKOzB-{Q7oQC-Kf2Iv{KS@yc$BV6-j5x{wZoX9r;}EDe%O++i2W zCX}5&=AingT|dNx85G-fAr;BX2Gc+vUPDMLD8TqQt$x@rBzAP4%C!gZPnBt>{V#vT z3KpB4V?E3jA!YMXkmox^_lm)Hmg0_2v|dFRo(l*#u$$XgurMF|Miv2QjA*u)TcSu~ zrd%h}P!b}>gY7_z z(3D@buKz&nH}`_P;81mAr6gpJ0(gI(Xo~ot`Rv0frU(cdoyb$Aodd#_=eJ&z>bo@; z-1omBIFef4v;x`~H+jG6q6sAHXt%We&N<=E!u${w#FP>w1~>2B42GHJEK+-E`-E_b zc^P&5I5y{9?>SeTf&;m84!Ks%t*yVDe4>;aGmf{fHW9DvLKDx<%E zlD#vC-q^_b-3_!ex3mLNzo&@WJ9s!*np=SW)L^Bj|5NEtwg>}2#>m9l-p$$C5@2L! z3XoxtX8ohimRvsBvllD)sd-wn8^B(voU0z&GSWR0= zoSE^@Jph;iu0SVe%Rg=ZiyOr|Gr-@}-c!w->}~&Q0HCq}fgE@k8Qt967|dOqK@9dz z<_r!tfALeZuyh80xY;{d1KzJrKpWs+!noL(zRL-+0RF4MpG5)4S(*UtoPmFmB<%m? zw0)QIJ?Y&J`Zuw6AwYj>+WadV;0y%*M;i+x=f860l$7KEwnmnAAfTO*oymJ3$Oz=( z4EX$)?fnfjrTiB`AVAc`$>~oI`Tun}{kzToQWvp*zcW358!sOtxBp%-BRdyo&wpt1 zpSNvdZ|7|3408TgMIgY;(gyg4z4Mq4x;QYZ+gUog0HwwLC;Ht4_qWX)2m&wxfQ|s5yNLzkpLTyu%O5lI zAM?8mK3)#?4gfPF8)u-8r5W)30_WvyL|e zU%iCw%x!=GdS*5TCRPrXe|gmZY%81h`F%fwEdLx~fBYQ(<$E7Z6Kgx5vonB==`R=X zU5@{MBL2Pxe`o_3RW)@)B^7A@&+PmaCT?e9Z)$004q##D02nzr8F|1ly$=KnJ3GLO z`F*xcf$o1z41kfr&K~rh0&s8v`2fu9o#6hgDhCIEQRI*5Ux*XHDEe>2$qHbU{u{9Y z7-jxO?~(HVMqDfaMwP$Od!Fjw=si#E--!EvJhAD$f6g?C-6A zoq@KNCiXV=|Lpl*7W5DJF0IQy;5&WSf57(|ZvRl{ov+7V@IR+c>Cfr=*TKT{_Za@q zW9cuf3Uacy25MTGzCV)x5h8B{a z_=@LEltPhnGn|xT324d0KrW_EwBh53Wp9xH1@i-%{L?7%?d7C+^nDWo>5u#CN-UM7<0mA0(pK)DsnFgId2*`T>Epy0LRFHE$50KGk6M1>o53op`4f6fiY z8kT`}=;}n@`4Ca z%{um73e*E3VQMa3JPXYeS!s%=zTADuzz6ocC*)^d)(9+CT1K2Sv7EM!syApI%;38nTK!bRm21gZD&8==F`9(oxR3b0Qd@a>2hf~;hnZwm`_jl~zn)ld6E}u`V$z~lTiakll5**|I#eIS zx|#qkeQ+gh(OrYU6rO|nJx8|UvnhQxuE1V^hL8zSgys{`_XtY84fGO950EVE+-h%lcp%Imvd0V+$V!O7KERLIe}kN5P#({uMJkl>=f*hkuve3Z40N@7) z@L7i(VO63|XSB<1Wm;xriZvDDwY!?tt9@+w44B0@E1Nz6k`Qc>* zMvAbe&Nr&0Sxj>QdF+KR3`h(I9LwZcCoZ+vS)E0dW1!zmm2~cI7|m|${*8~n+lF(2 zeF{aJqwQ@vP{MBpxN;}RJkjZyk-x@J$%I>fX$;-A0^TUn6;3wTTkRi}oFHx`lj0d; zDPs{zXZZ1@PDsD|Tn3!}$|cL%fsRng1mKIWN^+k)LD5f3YFpJ+P@vPBm2e`I%?nSYZ_5!oOsRY`hPb}X9uyS8jw1r3|Fhe50M`MAl@){KG#qi)+p6Jwruv62R(`JYGBzadMFiIxdGj0CL#Ak!aZQw zX}%Y+oO4x5V4dW-<6H4i_X{3>wJ~2cmE{|+4Ebd4r@`hydZOt-OoBY$e3-WIv}m?t zkN9=GQqEN4(m%Y-T;3_ z_SUWORkW&(SOxR5VuDpjPZaCBnI6}on&rm9+_Lcx=^Xl}ydd7Cp&#piTh!^UJ6ll# zWxG#dMQLFeV(x}beNmZzjvb;JFPRN?4c0@X!hKl(0s9oed^dMfHX-pmxf>j)(2ylk zH&Lv0pVw&OiZqXvD+c!EoBz0up76KQfZmqxv1g1W_nuZsLWYp590_ny{0ySt(fP2F zO)`k4!y$qbEfal~Yk;?ZKpvEx<2drAk1!Lp-|3!q2H1aS&L;2{NdE-Un-N4<=iEg$ zt5mumynF;ZWhf=Qp4gtBCF@5Es@T5j7_-y@R)$#L+I%S0&t&;7+>#0>{iw>Vr5s>xpP%w3`2W(F&92?`YxKNXih{{j0tct7NNZy`UV zADnBT{RxUWU*wk8Hdy;Za&H1?7sB-xenF2@@v{!pvv+5cIZp5MTg>BBDfV`WUwABS zYX^M;56;EW{!lr86gPxYz;w_4w_GFrm9#(@`LM;#MB_dTjTH9_8c((ov^Ki!sis@f z4?WjmMd*Gf*}L9eD~%sa(uU%F=dHU>gB$&AA19f7iQQEMDX^=5Erm@S$VGGQk(jAy zxOGCHeI9$N*Z`r#kGEaiemEaanY>gnqV z28M7RpfZ3WD}E|mL^p8qBmB0OaP<3y6siNoD6cw)rXa{-e%fBiUdyGnD{qy@)a>;n zV^wJ7uQRTH&9`8$jG!a_YNu59emm~^gK-(2?6pU7-H z!FzG_8C38I!zhTUWn2{&F zXJGvNZND#Bl6`WFLJM0k#&!_0b8$`YkJ9zftIhsII-II`L}_F|K9ib9Zo0j<%=04ca@39CXo z;cI~@CKEd$NJ#&TJNr{Q4h|n%eOpwp$2uF6%ax)@w^k|gCy~RR-j3y1)9#^*bqxFo5xfq6$u7w7;7l$X`gF;%Q69$JMgGWMxH?5c zOaZ*Vx~EfL`fFl9IGLC^HimRRZ&^#dpj1lZd9{W|XG!c3eTaB%sO){?ZVJ0%B!O){ z(DG?kPT_2}YkZlhF^a3-hZC1*ROxKNorZ_%BT-CCB2dgjNicWAIERo3SU<)4n&^K;XZBLuK(TBdYA zSd1B9jk@NQX&N@|%iv6SXWF8Vk6TNR%s+OMuk`GptMUpZxFCp)c8(0qze+4J2SS>t zVT9b6nStvC$x^8b@|&(b&R&GGQ%RkFFVw4rMD1$cCV(^?*vn7aC6FepJZ&z82qIyD zKj^tf)(-M-nmS`%ol|K=J99bsiWG||ES_Bn@vxhJ{T!fI&qRhVNFRC1aJ`XqHrTjE zq`;@p^2X!}?>A}uvDbxFJK&o_zlXghJE)LQo$)xBX@mil8yd+@CN~lS*={L+fQePv zb|nZyHwwekr6<*Uel(VqRG*UFq8)FD_{rS!KFnVR{IDhcn5Bm0E>yniUXV6K|D#NdnkAAT!$LdWd1}u=b~~^y*v%8nKJfr|LO1CO8k9gl3C+L) zt>Or$VnQ$;b{1=l1wBUh81CSIOFIp-@~p#DCwape#(A!nuP=+3Sip(iS#`U2fD{2%qt*GTTBMy7TLNZ2SX zqk~(8SD3rmr{Do56#Y9-;tRR!H~VZy^n3TCBIL*MGoqH{>sv|Bx?dH3`0~oSBb_bq z`CzWj%k;?O_pl)IpPl-DLikNzYA@=r^sLkT{O*J`zU34(VZ1i^EUxmi5##q^C#^*0 zP%*ztqSZ0%<`&_IWruvMSx2*}By)R_DIxzaa1vBxKXa&kKS zPovIDdSOMtpr=UEd&#SrcvFaD3UZCq#?-17m-!+yPvuu3&VuWIm*E~|+FrfBd`0CV z2O@?VOWT#{SwTZ6{b0cz8z0G;?wWWFRrIG+?RaX6pcN5L3?&DJkFdB(Ht*uQo%;!=@rF4#m%Fp=3ydyRRAt898F zzfu4N>TaT)R36kLAx!LTDIO??4`^6nj->mS)W7NQjmF=9keohOBD60KGKjz$baWd) zhLiDiV^svkG5Oyq_+8(bHD(<>K7JVAg4c2_-LDF(ha@p=IQR}hkHI&L8%~Wwuy{p( z$%hzy#j~CxB=cbC3vZr|Xt=?E@K7pl{9UX-#Pupq$q}t@Mw@$3L=V;4p(hFP4lxq>i43sk~acm1KOm?Z9H zjcXNpaOwn@n|06Q5$+K; zoIHJ@@S$NQyhGMMU1Te?*TuD_<~}}Nm{}|x^|hgYs_GTK>;*fi%h;zuR(-No=6%p# z*z7%Ve#RllhGZb=>~hl0?Moc>(ONhWfLGYnOVYbjqVD~j2iqT?5+Y3)c&s5h#hh55 zE`!?J@w@EeIeUA{w&7!-oN#ysFXmI`Br9CAKx%}9-8!sr7^3|Syd4aoIm$~wL2m|f zEZGu&MhP+olh99v#b=2^|23=WiPoDlQyCr_wn>VKdb9sU_B4enlNwZ5a1914_ zM!xv<;*+gLPASSttYEfBFu#z1ifkcjf(iExzDZdC;mT3jCVNO1{N7|6?04Yd# z<{SxJ7-s#oA$N#gy`Ch2@~l94R7YvY5b+O!{ZH2%G*??RSWq^SEnz=46S@;z=A7!7l1D~q+GEgE^W*r_NrwIir47; zu$*@=bnk5un6gHyPXJMXzOoAG5gymL)dA1ETVu=7@N!@dw7$+Ijtjm))c#52g0d2S zcihX}$^}AGO;zNxAXk(61hX$u3zpnn-TIOAna2>EIaqHJW_s9+Y^NoNTM11)b2#g0 zmFKLzk%YvLQwbXi1A7v4x(?!Nng+SclnkpU;{_B4ZfH4n81H9yYerHclBD zP+!_%)mj*Kvxxn%A=l!I)2FBV7X-Ie*LFB|6qhX;uZ_Fev2tI{yz9xJ>K(Xe4PEEW zo9A?*_y_jX_y~!>~vm9h_4~dm10Nl4x$e zg5jYQ>#bcJloU9E$vV8(fQHq zREj`G$?n@|(1DyHLVOXDE(`leiOhdU zYb~R7IKJ5h41w6oi4>71Rqg&O7RZy<_S=jgjim^6j1R=xiJXqVl)~H|zOf)HJ@&qg zO%4fhd_^%+E+eJNm2H@R!y8`NU;sUosGBUP3{zF_8lZ&0Z`1}*ykRu7#mihM`#$2H zUV0D2Ql4HeRKjGuann_`*(PJZ_NBp?-^-DK?CW!R+1DohF!AK+-SZiHoZ$r0&1Hld z_eP)_)12RPZgE~_A)Ffu4%eJn*E~Sgk7p=)ulrCXO3gKYh~JohHQT;es!9QC9O;m^ z|IlvOR-xcdQ@bi5XM8N^#iUxl?j9=2+}UJ4SFX}g-09QI{Ntw$uAy9sS`i@}Vdh(> zeC+jW*j0vo1t!A91RZ2P0hO!?5isA2ffVG_J1moaoPEbr!aiFO9IjT_B|L&3202C6 zDeIfMy+7opE}!~;!`%~&JR3eSc}D2BB;aeO@42iihUb~A2yQwp= zf~BZ=Y{0=NVpa*7($EKWu9e9m7}Qnu^#gg`uN+cUFW!CztrmETwjCyh2$Gm7;CJ&+)8{4iQ~4JoCzvM7&!)Vb zxK=H^bNFMLm2`iV#_N#rkA$JRE^1tp@c`n~j;ceCoRyj1nzNfOm^f|9!yqTx@(&_u zO?7Qk@bm%vGtP}{%=Npg>#kHQlUE*3nGe@8XlzbbnvZL-J@HkhU1;+ENqdT1vkbRy z&r5i%Vav0t=nadFU}(&(wmw^&8?E%Z?Ul)d+?3Trq-)3AYvvCsj(AubxLH&U5&=>; z53r8nExO$2NTc7Rj5ZT-3PDv8JC{uj_<^4A6#vAhmXx7snu%z*UGt)B)wfHwj`+n< zNQV_mm6tQD90{#Zh?7eaX%Mz5ORNw%iqfC{0!LXI*(F1aijSAx76leur{1>9e?1U) z`4E~)&w`5NeVX-dTm2MII<-EFsshk=HrA!zF*v%awAO+RrEL^Cy=b&F{3v3n;qi%3 ztZ7}1L{3fhu`z34^-RMKc>|%nC>eZDPaA(JS&**lM~}bE0tzMo9JTBB?#+9~OP>CE zQ^!*3&SY-O3`0ZR9HP+-Ux#+e=#D)<)<}A;ZfM)oJ(?CWRhhKNL6VWR?|>j2dU(lU zMen}qd1OtU62)Z8CFg(6se_`GlUTY~kObssSGVVJ z60)(h|CMMY+TDyW9d#Be*J@MeWH1%t>o{MYFDd)N7t8qkn`H#|>!538<1b}VwX?~U z9DyE%pm&gR243DRR&!ht)__E$e{CELtIrje?5xO)sl*2sXY{9(IWulOXJX)As_d`X zGM^=A-+LJ3StIih=ANFs{Ch_Zvb}D#4BIM^Y&e*A?d~1$II9ZK^zC7#64XyMd93jQ z6?mrhSA6Cu$3vprM^Pi@*Gn_T2K0 zf6P^Qy(%Zh;K9T~3({h4n&J~zPtBM54SWS8lK{1nmW1y!uFiomW6lk(fC-qe4p)^6i+FG!^sbAXGbw363V=ik|BrPPv=X9t*J&s)paG~ zLqoeN{JIkLbbE$NiU4Cj9ZD4--c4(t8}YUFYMRFDhRIPxkTSaLDsh8P28fcU zPV^l+eE+`YXGR>~;Y^QbA0K_zKQcW}<$jEd7aQi~P~C4V3Q7bXTlA=Ff2DDWT04&? zM3fc$?J@jlF4ySgm3ZUO?9ddR%-xy*{JA?@w^z3CAf#l1BLIa<-x);=!F6RnO^WhV z6w1g8{JZ}Y$KwsnV-=O9=S}sgSv&*7D>vH>QTkKap}D_{YeLg7E zk06uI)SfRUxDf8+Wk~`@=<3V>vv%&{sh;ec;uS5DdO#Px-jDkB^1oLZ8Jnsp_OBXx zbFpsJlcfy50*XmB=5;LvAafoI2)`*X1e}8`E$Tva7t)T9+v(`HDbJuuO5$UR{e+VH zmyTfOIVEwySQ8?p;({$fw&yWa1x(itL|keiWgIlWos;#YNAl!cRe0iH8d#xw+H>E# z7DyowctC*Vp+jpB{1dSKa@7p|Ut`aYW8h{U1#s&vgXYC=?`$O9)B`3dv7#uXyjF!= z>d<*3P~1`3Mwj7S2+zK;_Xu`MybHs+1s-i^w8k3Rm6N0|>f?ixgrdIkv)-u>u6(G? z>()DC)9iQ3U6KF(e)E`zW%qDrk%icOK-b3-b^s-1f@YQW03id>Pp2hgVJm$sY4X`usuajWQRqVmPbm6s)YD%Dw0Oz16f+;IiF#YM0B zI00hpGLol8B2O)Ng>V|qGHs{PBqpE7mS;bKYd``SNJ%q+8f}ZuNbXW6m8sMyp%X_H zFd)Z-5Soq6e=>YrcrzTlg6fbfl3JCvqh}=S7)@2QVIxqQ8zifu;~sdpjhkmXy$&BL zKnr2hOHcG75jYacnXbY7G}5|)4-wO{7Xf&ynD8alLs~PT!tzGs??4r_NlJdKd({oa z)+|Y)C5*7?{30L{R-YK_x9NS|f9*XzB1P}P-R*5db^6*;Cnf24iHg|sHXM59wNJ{z zG7F|!FQ6+<4EAUaK4teOfT+?wQO=1-wC z3@{kHTbJ5TA4&b+zT_cT2#cw#8>`LH?9@Sn9i%ArECalD`Vr8iam z*?**}(Q>b-h1ivTzABCC9;EXVy6`yF(ZYdLR!4TJ1*ne_^0(yfUNlZa+da=s@|UeK1SWD}9d?K4K}1 z+9E%f3x2;Sd7O0|7<|83zJ(^`1Dh5EcNrQ43fye?*lTLP= zRI4d>!v6iIJ?pHpJe(Q5E+kt_>0XzrY)kTAKlQ&=1A+C$Dwza<29}qqGR+Jy<7MrZERZnkjm-4jb+l4we=8k$+TCa#yW}Q2@ z?JMQxnEMmN&IrKwC`Mm`#ug(ygG4W4NNkPbWv)&i;(57e+cAjqS7WGfgUkqkJqA$8C*kDzV(PIXAVHN8rs|%ED8*7O8WHH_M3nue)2nd1YBW1DaWBu3#_NT?R(d4aTOtBC*&5Ug% zS;$%~dp4k+R?$?`@FzMh3)tw_ZGf8rwNl!cEY5I~x#r0a38qtlT&U%WXPPQ0@S9a) z;`;kOw0s)9$mPjGm4VA98_BZvHTicb-M%c-S{p74uXY6ziW2uf(fm^p))nB`(i&ywtkL_l)EdEH_e*IgX21!R zLQn-|@*uaiBoxvBmnn26F0hk%*Ace8PWyPk8#3iRu<4LUg(!QkjBy<&!TKOCO$hN? zoY~N3tKF(c+q@5XlQ{Rmj(R_dl;}jDDjHTyfj-%tLP0S~opbwiYw^YxQe(P(;w(|S z$q>K|-?(ZX>l5iB&oo7hOr926s=m%#t3xYsm-4r85#EHaYZhI~WP1f5WCvDiYO0sx zIFhz^{sgcr1l_uyE9k;KZX8%Lqk^by`J@tS@V>|h!{@5SGLlT67isYl7$eIkc)J0v zdljR+d@SeYr$^dQ`DrHJrJZQui5?UbtDNPHf=5+yHV4Y~Ch>?3ZURa%8(Qh32pXbo6SwG+>WYvA;9o9!TPH8ZizDiHuF``NtIia3=g)q~*+YuSUOFqDk{Zb}R3B?X$vfnb7FAFs$%bZJcUOZ!s6=D$u z=jBM*W73ZCU*6t#dR)WwGAm8w(bEZL#VkKb&nP0=9B^|OL(#aHWtweX<2)5)O{2nWqhg@9C z&w+I-!C>m)86CytP6`FjIWJ&Obpc3Uf#x;9O1_c%c%|giceXdYL~tnEXezQADZ{)S z#X3g^e3%8`H@voJrPxUk0t_>$rD813^3MjIyt}_t8`Nl>O*<9CCI8fbk z@In)-j2x;9vZe^b=5nd%8>uHihLbJZoLlY8Z&!$ryO*R8HYZz6-2iZQUqVdG>Coi- zVXq5k)6U#~IYcGVvrkcI8}6)&(fB*|OS;_8TFT4#32?rfh__OdBIGqjIwO+DRoA|B zu<^uaP#3J$-!}@4;BH!#UJ_$r9y-D5|2a|V!PmGQvT+8d>NC94`<%N#HW65z78Iug!WNl;k2+(gaOf_7pW z4dJ@A#1-Tn%Ul$$2aAqfvOUM)t#GYnZ9HJglRrlbf9k=F6~yEjfd?H^qtF^Mwbc&&h8xHd6GBg8YA1ZRcRxA`Ac6G-=JH zan=7RI(o1@Dr}{iRQppx4#DVS;yy66vZqYG{-)Rsznddzi(i2b`Y-II0-yI&Xz{#C zzkE3ESEbZFcK{;?uS#=c0l4Kq37?%WzRs$Wdj7n->geG zYzw({b@A0+MwG_q!+fhifZCq214tCV<+Of57E~NNBImvExo9*~%HvnuDr10#7&W6-N_HmDD7b7BC zE+j^%w6_SbV2-3FK>qmQVG=Nar|uVkS^~FBV@Ts9+8}bY`{@k)$?n#kv9eVzwli+u zvN1 zsL0opz5_sIia~K`^mtJiU0UHEo(9!|19>h&#dZJXzuPr-$zSNb zQS;Tb^!u`-mt91$k=diwtr~DY>-kUTY}M{HOdJI^gN_t-^Eq_p!oFiM0Q3G}Y7Q|t zqggp{bhP46u~p0OBZ57(berbSJ+TC{*<>eOhY0XYP93+05!(0fvWiU1mb|+`&0!(a z+nU7w2;+Uf+(CB@Z}PI=o~v2VjDwnE#mOG`=hNu%O>1$1A<2p$^-lP&P!>*&ng1?e z=E-bWf+h6o!&jz0qSqNzF#^r2EQ5P&)+Dj4mMN3rc{_m;`AqyKifh(FkUp=A2~65R z(gF-XDlZ_tE>6<8emBjT%MQi;tC1`CkbG7}#a2AOgpf$@Dk`J>c!=7r63EZk%^fjT zQ;Lwz)Sc2;&8wkH9SPGVmXt-|A+`H3XnLgKzmlxmFsO)l(f5(L#Q)&zcxf3_L7ns` z`|Jo~Owqkc)8u0dpHv=L?Pe7f4|yLl)&gufi_1v02-!#f=@1XG2roCl`bBh=P{_Jx z_^y>@YN@9=RO;N}lsma(6Cmp>Gk;DD++Q63epto>?Ym0EY3ibgZ#~q2%LE?{a0^|Q zz`K_BBJ}{F#S%|7{W}J_gmVJa$IPTRtu{pA%4Dj1P$r%2Li`<0yl^Ea2c{lO2o1pJ zf?k=<+&vs%iNVrX z*ycS=5}@G6EJKW#TR;h!{3r@fV2 zKXX*7_=7`tr3kIDRAiPJYH{ePnjvjF_g8MW>hZ&Ma)AMT8Wr>verkz8jtcPabFJ~r z3=IA;abe}<&1fp?wK;w0J(vrwCf)TXkADqDvYuTziCSYFkr*!8G$D0Tdu5S3?GeE_ zWJ(mp;Ie;y{en+-wA^U&tp+Adqfjbz_ns$^3vem(;Xh<=nV!n8!MxDPb2y_DlvRIhfvc&XOMH`mpR9@!*ESl$iZnl2{_93$hHsN z$+;1faq>fthcM*3sK6T?bat-m9O#@`I_AQ*V;<>0cTl&2|K@JX=>}lZ3PRmKDs&v% zX#6eFW#&^-KJ4g|fRhZok_VzWrW5~$c^?)!EP>96!1UoOxghF%qR!uy`59c^jH@4L zh%yPmBNrenJbsBE=WFAR?mJN(>acHg>|vSmGtt`##HcJJtkVF%e0bg~=#Kyl_J{*3jc@S714pL;|fm z{c`~SZ9?i=wug1qc5HfRF=`H{F2GjxKeRMTTmQ?oPzOifcL6{^cl1LqtYgpDA7RN3 z8AjamAvCtqGONGo*4!6q!{mTXd~OtQF~owd8bK5~qu#fiM(ki1e@iamlL-8UlYF=c zkc77FU_V(MF~9FO;YtDNM!D?|tNV)8!SAsWs1o?SS-wrVo)^pO@HeuJ3as0F# z043K!!kVpZhX>$5KR=c}fx_ndFU)Kp?OfQH1BhGtW^0*h5C0=1ygaOjs{0(Ebt-H; zm)tOW?hKAZ30;+Nc{a{B`wF~26-sOso z>t5=51q2#G9;ixqWo9WcSiuCctn& zBLOw6WoLv-OFb6%wuoJYT~Zno@rfvHXp3*X_!-3&qu zk7bcg70lca9pbPN26q+mP~7!_=4fb7HlwSA|+Jig!4eLODYuQve}CI7mm#OdYYG!8v~jNX=ABtQ~0K z8#6PpZ1t@sumnuWwi}A8er!YnSCd>4e9(ay2`?S;?`q=6UJ7JVW;!uXhBFx$GwX#~ z=4IN6Z=6&t9OXfF5Xk;f@&cyoLL|J#cshP^n`D1Q@kCfW7Q7hgm4#4biDsW+JU>9% zEW1=$p-OA@oS>gsE=UL5ull@(Dy8IFO(ULEZf;fORcUJYiN4iTEf+B(%p)LQRX=BhlgqnNJ4)KLWhv}$^Bm5Zu-(47&Oog zErhO@Y_pIRXI%^aKhsm5Or9oo_A=F)M?S0>YHJ_@^`SYPZvP*9;_%Lny*p#GEI`bY z&P>o+z#mt&H5NA`=}4NE7W8DWXORiAVRD8AHAWjvN{5KewQT{PhY*GP=Mnrr{!NfL zx38{;{$Hl03BHx8GQJnhKcNA=uUuue?Oc{-J>-@9N9d(26zKIie;JXn-_?4&V@gtQ zVj>~J9q>4Ncs+}CQH+Fz4p0uOmjQM`+)@{khGKEv=>A+9xV%7FnSuH2ik>G)SE?c$ z8Jq8LR!S_SYP2pB!u>llJ4xY2`%U@n#o-^HP$3yh5uuydt`XQJ>z=pv zd@w19P1~Gd-yZQ*uKHg$A~7!_fJCqDwXf@c*T&0JTnurVB?Yjm{T!voUVx`Ml7+Za z#5UyG?92Lqn-4QN4CaetUlpIdRtPLME4RV{3$XpX6YzNJ?8dWZaeumxrI|*G2+H2l zwqg8wLL{78t|A)8IDyG;&eAgWAD`Q4sLK<!~^+fh;Ay0GWRxldKlu81!1CA-CHEsGG$hfn$5_3FHP+?F86YW`_9JuMHPR*MdR} zZLBQ(!@^VRkarKP)B=+tHI;fh_Nc0eTJGe8yNFoys{uO^%{Qo(_DVu<-LXAkqbmq_ z*7bVGhV6&l!jtldTL4*5WOxbdcAS~`DW+gUtvmnbw7ff0mV=`}?b0lrmIH%+s}@5a zoU33*Ombh4jsCzqE`rBQf;G=a-R7omzQU&8-Uo(#5I;EeH6Jop@ zEX&!n!bTUgHM*}LmQzo#AWvU%C=H*{p>3K#*)O2yKZJYT`haF0C-`O zLSyFaoArZ72Y^_5bme4|7e;r_>P;Xab05`WgL&X+rSy~lB;T+G^PJOEt#MkVh*EG` zb<>x&^|W)aem?fJdO+z?O_gRd{l)^^>~cr1Pi4X-nZSI)qO*3P^pQr|X+2J-&m?Q3S%XABfA3jbRRO1(?vPgaS*gX{%*tPn=7tO` zikd)UJL&y7VOK)#hgW8-7iCH}#_Dfju47}!Gq7fdATan^=2M8?!FTFsLjrN{?GzC9 zBJDPV2PDZVMOWWLu8n+xg&Zz1@3Hh$6|pfi-K2oR#K8i(^G9F zY>|%9rT{VrAcq7{J!&z+86^@-cAk3Vp3X|m8xzAkrlCsSp`WvW0!QoONZ;S>W;;f?OY<*?o-|<#`M}=ZPUP( zMdYd%f?5b@F4x5Es0ucPDe?%6$ON}>=NR>fcUK@bs;jp|cfzXq6Y-}#nG2g!1aFf_ zad(v?#D>sz4uH;`b|*7nyMU|Q2#Ltg{&jpto@zKUqk2Szm-$gZ5a=z5JmrNZHUQek~bqH+C8n#&TX&P zF}PIh1qbrigeQmz5IE+I%~ztpub57VeOMmZNluOHOJ?Rd3&N8UL18R}a96jlu4%L2 z)x-u!j`aXAr6E*O#r`Ybhep3$>|8}`v;odhOLX_b4>q?kuifQ#jV>sUSU2qceyxmH}(vBTKOpd*|C~pZab&)hACCToRo9p)8$=bX#sGi zjgQyj5}!uNa>e{shh%s9gwPH-pkCK;=7l+*yvg7oO&8Uye14!1<-^~$%(=`dG3?*6 zBc!3qzbv#HkYanxw)FN;Mu8#4^PEhx3Y3?=*6@9pR-2#zQg{04h$6YE6Z4Aj%61vv zpr>%ygXl34sy#2_lO%%l1gg-Y+yLt?@9**3wrX8H*zGA8@hyhUGO!&1cx~_Zz!RfF zrq3}gqI3nBFckxx9p8-&vWQ3$Of?WNg(rEgV{@&GZ!I~oPaWEVST5&fOKM}jn7#qo zY%o#bA5P77pE`fQ^cd4l3Z|%6Y$=!@DC{YN4DrASA!Qi;%|?R8zYNDTodTZaPRjQi zSsH~bw}wj$#w_NJt{Y%0sCqkS3m5GV-w}P~@o*1ztA|kv_o_6kR^aNgj)07Obk9#; zh65+w8|*qnQZsE}IfGf_Nj5x2`=7+T(hN%WbJLsYluU@QNZYh*h2XKGHjOpJ-da(d zvWn^@IMi6M+NT|}J9rEi69L2Mx1vIF?ygd56Z@H1DVKi+SGO7>>+&olby13fi3%N) zdK+Nn8+P;AT%Dpi`%rjA2&~m&+Zl`rLa69-r;A}99yn`wyOs9 zI*m`HI55BIJ-1bu=^&O4y$dL;)c;LTIs@ThBXa&HPperVP(+1PZwUY&F%qz%6H4&o z7oi=YYH$`{E#}Z`J7!bNU6xu31eSz>{c#I{ePx&;BuFe$oLqj7bZYQxkNu~Sx}Xor zH*k>t8ogxJ?*?h+0#15ZV*1MlQy#Epe`r|y&zUA`bR3w`qGOD?r&@Qh5>KE7D&|d# zwECS%g)HFtW@=QfgBdWE-|h(bO2v>O-LL?YXe88QvNiKbbumwjYY}JFNV$|@oHu6m zJ%+02xXOzIsQOc?U@?4Sdtf-B{U9RlinhA_)3X|5~~V%xk+ z`|Nsr*+}x2zl-t6NP9t2xBg4W5vh9{?Y29w`l|L`;fP=waDjg~H~mo=n|)lUDl`FS z=Qlxd0@o}OGeeaATB2KX8LkhWtuW7;=58Jq=Pn(!DN7q8KK^*&!*&2ojyjWq;PsKD z>ZcP;##UqOQU%mdZU~mX_p`Pr$8lq05=^D=F>dAKX>^Ivr@qV4-hDcCGbNJssR&vW z?;!#z?1vAMF|||PASoEcd|WWKEo=sL?YWD+Ys?gkp%QcI_P=hzrnq6@%rP+|nrS&q zG4S;fQuV|PRGplsIeHVSuh83>1@*rHz;zIc?*^Zh>wtW(|FGc=9+g8Enkj???k`v! zAG_y+`)}P)2*>8ej~%FosH%4pm%=o))9%2deqBag0%G=!Yf`L9R*Ed+gmCf$|IGIX zRqJG-Cso8Pes>S+A{o(wHdsT%8i@O+3{%xs=ah?@5oAuqkS~4}!rf5ib@U3FK-XiV zZH1Va8305gNO%UML(^OKpIj6Kg(Bumk121cLlPDTH+G+=x0Mt&MA!qF0X_ABA4;s7 zQPP$$kK}AXwm-Cql>D4WMFuvtliiAFIRoPi{tOHSBa6Y4f;VXifwndu{s-D1Bx9CH zc&IE&5JY}~`v^X;u8sEP@DsQZ_oofQ$yc5g*8mi)%x&f9&2`oaMi}m{%D)VZ{EFD} zVTsk{JD`mhgcz_Fc;y;i>X*_4P~1W_@Uj9fmfvHf@y6OMw{-99KeIovTyn*R2rrU@w1&XjGz z@&YKCqy0(?E>Pk7jq_AWIaDdA`NmmuPxt*DrHWET8XJunR+0U@RYn3%pvA4)%~ECA zd@blvrPwg(L9cO&9ytY{iNz(Zv4egsn7KuH>YLhlblw~E_vYP)>+cB)bs;bE8AI|h zDsy&^(7C^yu8GPRrnEwf=^VI_RIQM*SqA*W<2($f@T~$qcL;d!;yV%*9_|A67Kt_P z5jw;sunA7>F&8CZN1dR!w+PE~%Hs{r_(62eOsa6y&j+E3(6`rGy0c35zT;>yn|naW zG&8hEpw)#UsgY(pV`W8@nfol6TZ>xJ$+%V5y27K&v=E8_UBdY#H0P7cfK|9?K%|( zLd62tf7H?qxhh%I!|G@UrDObI&F7@xA`)>dsv8$nX*n1gb8$c~*olO62aRD8DpVa? z`D_`dfviS}@|80htn}Umlew-{ZUF4lYFW-rl}>l@*){Nfh}obMhC9&E^vu@lGA{`? z=7l?GW6^AC%13}?v;S+lVXtEAzDAo*lYm{!Rd#h@RTfKV<=Bc0{aWYg)%?&{#-(3a zDbq|}L>hcQt%TH1q}ms!&I5U}cgq#m)@uH#ni0qIDP_k8OOMzNTE`5QU;v;o5gxYV z+LYOxp2BD$pMI&`UMQf6`y&Pz+S9z8S;N%-Wq)L#w-8Ecr_#Hslr4kCq)8{H3R@2A zwqQ%Jves`T{rAy%q?^mF&kmrz-G6!hs>*HzWz^J^wr=Nj5PiYI(^b)*Fbm(YdYG)? z?2M*Iz-F40%kLu~32*QFBy`_P^Bk~wnp|iFb)m%@pPPvQwaW(G^LZ^m&ZCZTfNtMo z!dqw~31AVM?EBBU8Swm!gOVpFa@0QKU*o!MZmHLY;xh&O_b{X3g$=;#X#ndJWuO|5 z0H0O+9G3|Ty3AmGQLn|B${RYbb≈8O!tcgVmMBfE+_qC&Zy|@Ij|Xp*vLh@1b&* z(>{d~4-rq$XB62H%Q69yVYDJE&c(m(3L%Ju!EYYImq47;cdGA#47ao&U$Y_ z3c(G+oc`P= zynsU~x5M(QO0r;iq|H>5JEH*6<63~0O5m?FGBa>;xEO(;wzpzjvQNkw(*BTvj-n@I zcw|M69fJ$&QqJ$lpJ+SKzB5aRyBylcM_y5L#pX$9aJA3dBxC?0JqF$R+QfPGy7&gA zbsf8CE9mnUK0_cM9#GkxItD3}1_nrE80HRJ{jD0-i{RUbYE2Sa6N?=eT3?TYtzFZ{ zpA}V>Eq%w#m*dw;O1-oYwAwk!KVXAeUfmUM**z(c3eoxApHcxhM63iuDjFp8Iv{&v z=%BKDHA5XPZ43Yr$+L-Si=z4o-g&t;dPqlEfT&os$YI08NT~_TP4j z2$^1=O1|sMLOubCeMk{6$~H!1q%do3OF=8qMwhMnz)&U#LjT zEFA3rgFa#=VrFAvXa1ih^8ZgoVrFOKWc+_rq;@k^yvbI!NU*eh`j{@BfgR#@5zrWs zqZ^Fvoj`9#3b5Dxn_Qp)I?lK&=i80X?9XrCAmABYZkyFgY6w+RYHQg-b3@d4wDsMP zh?tC&|2pjDP_2+n4a^mRw7PZy@<49p>gw{M)YMKCNb@5wwhX{7!1tgYy0pNPrSuMf z3J2usYM*dbAgg%3@$(rg2*!Ib%0ONJi(CXEb|AVthI|k=HzQ_X?!FG87Yy|mqRuun zr~06>q?C}7sXQD!Sy=$}_XLIs#5k~=UH{E2yE;&Q54K>eRswcDm>Yma8`yH-1qd~s zY*?Q!yg~pK9^ksoQnKR zAl|Yzbt*eh*Kg_|fV5>WF(LghF#aP;nZs&WV4Bw_fj0$R8+^Om?!w1G0h>Mde1lF&a_a{Wg!dDa#6w%Ga@l#pY z;csX5&(0~JA}U}{z0~3I8W!O4dy5ac*~RVsEt2)C+e)>*yD_;j@@{}G1msMkGf((3 z3wW8H{9(u{EKjUXF3#jm8r<67V4(`!TW+v$yLWu2`B30ff(~g@Iz9leZ)yV7+zVA* z+?B<%wRj!f#J-R7K~Z<$?(omy)%uy!-C7UYx$gO`+u5Gno&Jh@FbC8k(BizP$JP>Q zMf^tiu@L{*H34-1X#x{T0F4hVC!Q#D^_8B)Wn9O-%Rv}8K-Poy<5^z9J+v_Yhu1^u{bBK=m<+c+RV+X>=>gNoT7cw!dIwOU&)+$&&pLhO+SXQSK^~-QfXv69lOPX`&fIVKR~;1H zJYj9Kv>^pg{x#seI=iHrSJ0eVoxU^`Q1QmlR1JG)_at^rXDj63{g5VgG-uW5i@ViU zHFhyK18!)1bo$P)whd*{yT&K%E|W64X<`^?w=h2zccs7wPJ}MGYA4PG-`$H>MjAo)7A38i2CsF7!M%Y!(VaUY9Ne?K4e^fvHt)`u!c$>9EZMDh3^i-uc>99 z99~t>U${LdfQ+x0^}YEmD9$aA9J*gSGn0QFEIhz+d$lyqX@K;OpRunx=udZKexRuH zK4gl2xL!ONfHqST-y@S3#J5(Y`8}DRAUwT*)?X+O6SE(O@n`YT?aAv~xLN`G`l=-X zlwD!kAK*R8uO6IPj@8s$gjAwWfN1wd^?#dx8s8xSK1*kLF1xE1uNWw zep|e49oT+%@oPD_=~LzN0cWqRK$z}#|J}g4{ii3#;rqucFI#Rv{`b#MY2=nb=s`{7 zn7}?j2Y0Sdt7pV3Ap2+KNB}WoGLPf5?lS55R^?S7?$$taW(3+6bg_}y`*Y}skT_Bh zlY$13Z?&x}?t13u<+-m9`%40}vajwMU>X$>z}KC@xi@Th=&=+Q7IZy__xJNhqVl{i zcc8C!r)}%tKlTy_6cC7~zbtYZW`8N6aP<-(b7=DW4{RB=VfkBnT$N*R^9jl>orSzF z*>KYM7BM6=CU`4^sS;)Yy6GGImomt;Oml^By`2I3W@4YLewHuV2KWcn2XdiwWUTM1 z=_=HuvMx;do4C9?uaDE>b89IppR1Qb_4kyn24E@A7rdu9p&* zFf(RD#Z*N446&jvX=0U^w5DyMlx`_zicUEExh|5GnN1SES#`at9oP)>p`K7>>b7Xl zLxWVx`kQv5Zi*;8wvqLv;cnz$z)z zl9~hp?GqTD{rVTrekuVqS6|h4EI@j8V(3En5 z2>oJ1V0pxKxW$LD;y)8Rl( zM<;cT6y;txfpEQE(-xv!t{nBbn!fu&|2Eb!zW4r05*hSlK}MfdMd1a!qGtSycs#N1 zYMw;ef>dq^e0xx|t!DQvVwB6?gMcO);6P)|oezY8#*(0k#i(!v~LH&av zfPCt7fjuU+pi_C4@xnMH<$;&Gzf2UdD1+7Q!pDkn{sjKQUmAr#?|f;#AQKfFVrw8& z@RYJTv?+1cI*)`@aKZ(cmfjVscQe!v+bF-&9sE>kVP+rcq;X30qC}IuI$H+uC=12;rz#ELuzM)R&qK~wFca#% z|DpOS*eR;UY`{}FP^K*K2akRCofD0i>gGZ$GyYN+)&DFlTDyX?jq1b7{Vd;Q2s%$ckMTSUJ;~a_z~w?}^HHrC#5+lH3EO5V z9-{h4O@E>A8(u*F2E#}D#gzq(SBw!0$g;KKEO9X3&6Hlkkw_Q*(sH-B2jz1Tv|WiZ ziSnCjlSS~oSJb{Jil`*Gu73vNSdzwILG-X56L$o4Yix~lP?7-}-~pqC--`%cyDOm$Tf{DJ_{sd*>|H?8!4&HO=WMvF=b2~ossolF1?Z2(X4$jmQN}Zv?-LyKbuBxL+3O7A2jT3 zeJ_ztmnI&2e<^a-B#Lqzp%ENa6-cjEypoL1>Iid@8W(8KvvZ-Fv*6~LD}!dod&ue5rvht5Uhr5%?`7E(=Y^Z({fOV@4%?pd&t- zgkqUf93#x0X!Lwk5Hbx-V>j_f>~HZKkI0NKP&?ikU1c_Pj7yjNP?%l4I~2`wB|I`>?Z@$faUO+J8?|ll!A`MSMFsPArj`jhOc? zTx)Zt%eRD{b`~BR;7}H;3^emZwqjG83dhI>dg1O>z)uSc=BoVV5n&Qyds_KvM{4?; z$Hw;ss03MKW*HFCH@~dy)^-Ws&_j2bplpDd4d=X6k%y55)Q&19^|&cG;Juj$T*C=h z7o2I`gDQ)qW?mK(ZJyx!#d|7*V$ybr206#LyA8JLVO(>SfGu`h%-kce&Hh|Dz7rZe zZ+dnUk6eNsf{AdZq~s5A5nEg44k6w2L<*@0J&>a=D{yc{WShy`$68JE8Z;TB%y)p| z8J7Smg=Y0$k5L>?1EiP%Q{W~Ey?|Z@d;wxLc6R&S=S)H{^Ydkl@FwknPb&VQ260xD z16H7(49*h+n;`%i)tH^4u(G%8?I5)Od5RI;!8VJvR9lh_>RGp(E~${4j!N*UjeCL9 z^}&;!WxJzEE{Jd&4Vo@KarK zElf2f`g+q)R(Q-X8A;bXZ^WLja^4s?Z!^Z@gwBZ_=bxb%`J(%$*)54W;V^)6c`ai- zFsp<%ogm{61xUC|u6BbQR!<%4vA~3{k!0~QVTw1s2=T*Y8j%IzDVb2YY%e868kqy_ z&53lUGYed?avrCozA%!V@pJV?&h{JgWXNqlzX}qG2Cr62URNUTmXKIxh)mavr2RT~ zw$v3swdNz`O*F4R-SN~+ssRw!c>1L=>EJ+XN@h#5g6D*AgLc@nWSipF>)@ELTy}O~ zHxoawe=-a;av}sH_XFg3dIh{b~>1gI+DanxFJE~<#hZqZO2@s<6U zr%BA6C?8MA&{b}F68pmbqX#+6@MBMg8bCL}zRns<^xX*7*#PpIig>Nk{Ha4BvyMod z<>GwptJc{fmRAZl1L9}5kOuZ@;kq?Ql+X$4EWHkfUh*=t-vt9_ocn#}r{`0cM0`<` zlRT~RcIiG2wY06-7+&CUCHgrmMINM`bet7vD=n7`5gh8H9fpM(C$d)LmvT-P;JX|v zT{suSrD|Yju>tM>MN8WE^k^wRlP2&LC*>T^$%rNg_^dIJa;L9zQ?+9gW=5S}BMQuU z*o^E@t%Og(cw1aE|Ik*EnUrthmZJ!K00+_LjgV;=ur{#{+Rqx<%e7ARGq6HapMhLJ zYT8%P-YFiaPuycHy7M)^+`vwm>&sCr z<@(oKzYRXn*{9YY(ptZyuBm;n!4DqwS zqX5>gcc>P-<+}OC@r%3V+IRbP?K&rPmYOpGamwa@WA28d?oacy-LT&J(nXwJ2*VJo ztrzuC<9U6 zmiAs}*D>4Geyi(~nfCgJnOW0Ky;gx&QUN?zhK(jyk;8Kyw8p5H(*^T~R3d^?dd=sD zTR(obuc%5Z@5q`-cD_xDZ^HY_jl^1kOs8z9j=2e|r3Gzo^#U5H7vM2dx!=Xz`saac zAsnZv*q&t(OLLh43^BuSo*CWs2EH73$TN%flFgnNMhW8n?Axtg_1~db4;sq4;Q&0s zimZW9*d1{{hG#LXz1o=RYm59#Lvmu zC^mlrvPMML-4f!a2;_3{INOG2dAQ=6NNioVSmt7s`@#OqOsGQp&Qy3z{qg1oIkgD= z8Wo#T0yhu#Of2SJ$m?*ya)tz=DFBTTfM|Q>T4d$kx0}xww3dY0HxPojia@RHoZrW1 zHQTlJiTN(AC)8WiZSpDxQ~Tj*eF-X5Uq3tganU^kQCOdp@;w~vt>70a%*Sp560IF* z=~)kVE`_TFT+??$Zkr1(;uB?mr!Cc&d|>3nn9kVtN7Dr#@}H!z;b`k^KcE0tT+#$8 zINg`6U0ZAU{FrnI28LqoyjeWv+d z?sOdqWuQ{dCaXZ49l+C#?h&fbw@uo-G|DwxpHQdY!l_&wsY{*+w5$_i8si8X{vOp{ z)$Vu#SPyz(r%7FOzti>?T2)_))@<7@RSSKTQw96Oea~tL3{`OY0hBw(>p}^C%S2jB z8cTS?miQ}Pi1t9TnMA3!r3f2W9jA-e48A9jIgvfTGk4j^=1$HmXU)kQZ%wieuKDcjx813Awy((yM zYqGpj_ly}W9pYJSmGNtsMw2B{=VIj(mK1;WJ>G$!gH{N)vT%2aQ-!HJnB_EVSfol> z0c0dwQ90G3ekZy6Dy(-}kG5A#L-~`g*ZmL*-=I#07sM`EVLbP)0nhjSFVLi>ZUQ|> z(Z9@y8McZuKUE-HM%e^^0-7VSV^t8zopPFK9KMg@E)mhDFOY(~wd;c!u3(Ulfijbo zL`k4)iRMQMszbYFhb3aAWWva#DzT>}^xs3Nc4QcmtdiccCfuFV9(&>we4twCzIeWBO&VVw%7`+6IuKz2nH|ul}!>qcPw`2 z+I{SHj7s1a6wz>_jyV+*vPPU}$DQwUwHP`LN!d9f9YsQOjOx)vr5&Zkr|rEY`| ztNRY6=8P*5(BZn-%*q0BFyLh;6Sz&#cJUoJl@f*>CKKp?UN(I1Ho12Es}-9eUVIir zMm}xSA2+|sSZhIb8q)ksbA3Xn^Ba|i%!O{gwiDmTjFKYd$1q&W6e`B|>{9g19W8nU zOcgqL0)lNdp^6^Ad!{~3fAh`riFD~h?7+E+_x1J1k_qn-$EEVXd3ER8)R#8RK`+z| z3fqQ=vu;d(IyG%PXm~#wm||?`Ny?&j)YB&%3t_5Wg7gpBR0_pSzj95ft`G?}{{&kv zjZmubj#&W`MHt1yR86|yp>;G6M0}yB@>V_KmEMgTqdeOsuk*|`NJN(_*hO~`9`7~{dJ;$T~SasVi3eY;(4}$T0doGm{?wq zW*ZrYjnNOcVFQ&*Qmyyc!=ATVMsHizuITXoRv;4B)Ylh97q3IPe7q2)`J(!+Iz*;p zBx6jS@#r^<6DE+fn8Qif)$IkY2MWs6hhML3swy3nyw{iYk>+DqmDRfEEazBd^jATe z3<`OFg}kX8(pl%ce5<&ty@bXeGgrd&;8EyD?z6bZPREI2|L4=u2@#3|CWPI3qx0?6 zo#=`D57)5y?0c9;^E8nY{l!KE@Jwpn0wf3VCC|*`Dpphi!3|WVI|HAqY01mu02@RU zOh1voqU`9`A*&Yh&2XF%Mi?L?e^xssZvJ6^^2vh6qwpISiS{Az0Ee2na%XqCFVT6j zv})nxbArqH04N6Vh!P8vnWMAM0{2OES!Kt}#%VzCrYU%*VZeb$dgFiyq3kvW!QIe> zJ-p%k?0kVdmV)i)s8`4k6o%)O&p@b!z>I((=ho~RS0Pid+H;#4>e$a zdgn!%8O3O6&fhJ)8@JT7dO`W(rFPoX>a?G$J|NL2YV!^EDGW0^Me(VU_Z)V3tG8kX zvR*_5FI6M!oq-cH(*XX}>iOQjwLd}{n)_U8fksL+a#q-xKj%ksF0*rL({(2}i*3Hs zd_&Pz?XUENeA;`;ItbmD1{ZzoSYBj*lx=P)rdz?Y!HwdL3Tc)^TDQzZgNX4i;Y5T^ z`^`%9!ZbQqG|497rJ-rOV~OrI@yIZW%M(iXUL_!~*LDN}^?bG~F8HgEV4ootmI)&! z(PSw(wKb+M9mr6|@2JI$v@cS*YiMwoKQtK5f&X*1>}L?Ghdy;qau)z;pOxN!;j0&& zr688$?-HiN5?gj#L$cxeZMBf+fUHkIJBs>Rgxy#+3Dt%;p%Y{G2s9xI5t;tRr>2ue z6FjZm#Q_?*gr0UR^w{yW>Mtip2hSnKO&w$jEVO|SA6W$_PG9&ew!T;=#`K^$(@4;< z;-SPjazO%xS_rm=Vm?1Cd?Vq1-qa4g)yj~;uIWB{@-#ePIgUH>vCzXM4WzBQRAi6; z(GFTIOTJ}NPV;@miCj6zceEKZ5g>x55aj6!SXdvW=ofqPH)e5Budj*nVQS_X3;K5JgnzoImbt?+rxJF5gZ;dMduPKa za7IFh35;BUZn*=R4<70Xn!YkYG1b(ldVKdche`dO;7E*7gl%X*Jh|~%Y250WC&U)4 zq!T=ceR|4+LWxATqaxOJOKH4qHfT`fNu>Y9z}AsAl|C>5*?u?$<J`zb zHUw_wgcf|OTz1Z2w8izY_mrRS?HMuAYJ^|N31bethGP7-o8sQik^d7GM%Tyn0|YJOjPuLDPj3s#SR`V;jF(tg))s;SoB z0>PgBh?73Gl079tF`zm0R2N+e!_uF;hh0O0NT|v;#uTafEq&ARIf}90B14(+mMDVz zmron9A1l#szoYkdA<^9aB=f~3fPLQ*9()S;QkaN|l~0}!hpl1p8Hf`G=GHN>_q_^Yj=k{ zJe!8A5S4Rhn4Kx^NT&lF^vY`Pad>O4foGYtfl8nE*Y;-B&H1$E^IseBjrMnAbm)Wha; zf1q5x+v9xwrQ%l{#{+D1c0w}>A)v~=%~(aAiV2jCDs~o3kYW$rz{07;ozuXM=QA)L zQx@l3YQ+D4aSAANUhE{9#ML#`CNFZPGM-X0%psemBCU5xpE29d&#;K59;TXKIH6BMdb*o}3BrgM;wi8T*yQe`u~`ds;<$Fe z%(3%Jd9_7nDKIL_;0%wT2YS&MbmY*y2dTGP-{vc&6tb}N9UKlzK zGW<~^-K7V%o?645O@QbtHTgET8E~8(J7*YwPZc@n^2vAob6)uD(WActo76`B8NV|k z&Ye4&NO#lkx>wv!a;4u*5csazBa15sq+~6B%Qe}u;yC4fW83nW%0s6L!Dw?hgbIDd z&6kj8!cOJfo<^T59r)B<4BMIGvkScMAQGOLgDEXGwjYpx zq_^6hVg?XECPrnoZl zhz?X(JsRi?#YOe~=%<}<5m~PJT%QHDZ z<-{jyeZv~NdPhyM6L2!*s;MJ1GB|lUWxJ8E_Jm%t(#2VofJc2Dv{o??7gtHD`=hjV zL)b9l>4axbSJxt`@W?HL+J3h>#I=+@^p}$KeaZ*>f#cnvNnL%>PZfcG1MT67XDsD+ zUe;+M+RdJvbv>nrQgJ$ZZc*pY~2LFPI4RvPn(&Io}?P}uh z^u_Rm2}Zb2U7VdllWK+oQf1?IH5F9oA-{&^_EC!yMaS<}^07dLmSOBRZTTo3e!!+i7+a=Ws@y0} zXTiD9a$owB7$@q1)X@613Zj$pt;di7f@g8V0jOH2X0AkDD*lN5O1Ws<*@tKhVc364v9 zsb1`73DIj=|C`NNVHwIcbsagAB46cjUTjhKMw^%3SsBvu zL7YHt9^evmxn#N~pTKrcBjE=xWx9*q@L@6Vv1AM*Ju1$BsJ=2Cug*^@EIal_&qzHb z3F_4SG$vl@_6nV4A$E(C&T6g(UZY^)rVIEJoUC^jMex1#-Jn zXJMg4|B`TdCsrks0D~)yC}3fTi^`{fPsvg_%UP{U%G!q&QPP;FR?uk?MeHq>6{?{LKHOy>r0PV}<8E&oU*Kd|Z z8ozHB>W^`6i-9^P>5VD@_9JMBqLJvPpMp?qp1e0YNIxnjTd9zX&H<37>ioQ>W4!Mr z%xiDACM}>ezzqNoqY~W1nsA=0W|h(>N(CYZO$SYX2t?7&jkzzvDzE3kF35QJq#q5X zpXb{j2_{poyU_D#dEXD2$L#7Zy?R3CIjW$&n`-RRrYSI4l2XsAZsPmo;Wbj@bZ0(L zKH9H-zf0tp7kl%HHWpL$&8xrur&@UD2eTh3UZ*!bzQGj1-_iFML=&`s>c+kUP7;AC zeVF%u$iq8vSJSy6w2>NkoLX@`xwvnAu)|NL6>Y|96l2I?V`{Joafd|b=>~e=EaUM2Koz=u$R+s-v)mCDzu*!HE~Y;NC0Gb}Q5q=E+{B zutt`9(g{CUalQIvDE2_xcT?(k7)t1s50?qhmDl()ckjX_eN2U?BQm9b{K#F!{R%%y zh`^)__(o9R3#~DcHCdna0uP7l!fpBPfr44?(C((gOhTw->b5Uj5ML8G%1_vS0zw&n?WhLB*`CfX<=)I4> zK>Tt%YoKik9u_%sL>;48-M0C40y%C8mD(_8;bljE;w&N%AFixXlVrx*FdmqlB2_xn z4&9f_pwraeGDv>!Qn+>&P0@WvD=VOKE)iSh@a0K85E^X^b^nM5ycD+-V48e?S&?$i z4o^F{D1v>4X(y{p25U-_NFGsl)A30Ev?!FOG4(||zdnHs!PAz=Z&@0=tWc7t*V;OCVJW%a#aSWn)HO>rRpQ-KzO*$9lt6Mp8*HEV$7 zgVUZLkYGHL*MAAq+RuX|^(={hJpbDkf{yGZbH>?=Wd^`Nuz z7exzDHrw^>?~O4-t*xQvUqpf4-zc^MtL2rk*@F3qf>+#YP;BwoLKAv_a?Oo2h^qPg z3cmP4;kh~@gA{YqoO~5r#9^>w8zX<82M>i3tI;mZz+7aSMIMn}B-Ut>cs(jups6B9 zXh9yMb-{dyF^$1j^!8WY@OFmL?$+Gi$)S>G&k{*T>q>s`LxC3Jh{a05Ymyc8G(Y1g3EjP4NA#@v){ghyQv?;MRX6h z4n?V`JcF?VMF(Z*|u zkVQlDTqu}9$g^)0-2Fzy`WSroT;E$NG_}xsTOb5~I9C~Jll|3y_4F*peeGFc;ilN5 z=C~$Gu*XH0z1S|KYy@)x^})Z)A)*x5@YiIZ!YjbakIr6R;`KNy4I@D;kWyP8X--cBdl>; z1yX10_2Z|(lQ%AZPM6-@)95mchLmvUk}iijfLA6o05$y+6;hN@JZu6tmduk;EuX~-tRf=zYyrm;UEQs^iUnG zO-?qG^RDs3BVKrk2VgbD(`%fPL#mEI%S`IPS3g|2GT=r|pY&q5RzKKN7qwMafIO_3^v-a^<(Lso+(F9c9Mi7 zhBlhB!oxn`HWq z27Y06@(G?X?N+}rNk3P*C`)_IDD2R(s#Zp3V*23=dJ$#2y5Gig2rf$~e_j&?KRB zZ7Xhn*Hu-y{;tFJ`&w{*6}=hXIYy+^l1evPR^S^T=`$W#hHhWI_8nS+Dna`%^*qXV z72}_C#c-s)VeNcPxw-Q=q%wZAWPqf=^2?2$%Bh)B&jY?_aqJ^3+9Q!Q%@?S;`pA1E zt*1LY8B~(MZx0B-WknqoZr#JdTNiv72UnqgJx2#2jy(_U<6Yeh%)8k|%&Z_ab+G4% zEw(gcq|)c~x30%5xxNF@SqysczVpQk9wa&EkP7@>H84eam4vWf?T+JXbpW42H8HMY z(5GMLZ5JmloD9{RWn@D>ynv2q0n6Q4vy|Z2P*NWdk2&-lgB8W`f!sI!+1IP;C(Y%5 z!8tT-;39$cl6TyL44EgfVWmHn4l}c4Ymapx`W$6wGTmC^@v-krkg2sG+&j5zB1Gz6 zhxzKwrl#%9^`|{nY_Pcf@;9z0xm#31uo(nm-KUwDWW4=VF)cu;bc(eE6{mby{_^ zHFAgA6N#c6q?90w&0~$>`;9fUaKQeWg3Gh%xRW$b4GniuUV7VEL)e{WwCM=vGV+WI zw_us*dq=CQaAL#Oc<~H~qE%Mhw&R7aC#ij8`Y9MzcyzxF#(3^YubW8ad79LJlTct+ zQdc(g#OgET>^QqjZEpq>DwW%EDAM40Z_+`iUrii@TL zSJiQ=!SDPDDSqx7krux;#pw>JnT*K*>O`pdOrks$Cr$ww3Q`Mj%@(4zvdwq+T+XZE zsr=b1GKSfYX8DU=hvVE0+z=IiGH$5pM>nuSJpAwq3qd(OF|gct^^^B1tfowo48sM2 z7!?F`hBr4zCVSic!!txx7G^TyW{(qg4x|kcRNhg2393*XYTISSk2BZ`nQQO!1L=RP z%9)@KUdSBA(5WC+hli$4QPVe5Q6{smx%Xgjv!0=Kd!`c-)Uo7{QDe7%OjGFdvqeeH zvq_W74nI#AFkTgk4fkywdx(BxJRzgcG7K9dC+_gQ*DU&m?FckHiRD`RaRbN)AKMIw zgS@q;oiQ8ezi2=fc@6J3o=X|PTP9@=`HV@#2y1g2|1Q<6r(Sh)PDErBvp#lG$TIEo9DOoz0s}9U!oqYcJ|!ArFD20?mF|~-~{(w&&qxJ zNM)BJPk=KwyMO%>g2=;AAf;%LV|fOvw8k+-?@Yt;EI&$AXBb3(Yfa!^71!M86VRw( zv@7*an#@ja;3DG$GBE)aTP+K2vMf4aYVyOQ=n=?dAX}gIPG=sc$>f`D%nzRdI`Kt`gk#0xLb!0$|%8o2&d)@fSfh~b@Nn$dWXJ3|*{UV<(Y+d;5j}aAj0DUgKM-jjaJH zeQy@q>A9Wwq0z!jv+SqNN)lVKHd0}+ev1R0T*mO;Jw7>q{(+%l9>>iT6-aU6EtDkH zxZ})m2;QXh9@p#kMgm$JbXK>DMEM=#peKxT0+W{4MjPzh+Z1?YMt*gj8>Slyza4;4 zNNAg4p9p>ahU8OPNPLGCwqVHu4ypgH(T}}4LP7!DAgX$_m}K}|kVL%y$~SE%g_$dF zT>lxRUNAy`*Z>za#t&*(eC55q8&{g<`uBo~Bm7rDmg**o!r;Q+nDrfg(nmg5f^`h! zWyyCg253;$v@ha;Ot|5h>^7;3nrnucpP$tfW~PaVamgO`>cW-#$9fEY{xq{7(KQlM zV(db)xQELm#H)3SE9&NVLSDrzf8F!)@{cvahZFbaA5C~MLD zkCN|ymioCfTYk%BTAqgTbIoIPDV)4+4w}T}26xtVmd9U;JWecTUj8 z&=}1Ufs4zdSer#{L9#(j(1GPTT8H|Id5$)J`CRP4L+;S;Dl^{YE0~*Oqp<@0u}ZdV zi|+ZyP?gszLrIH~GIZQ+b(kEP1D&x-tyl7nlS3mx3rtuzqTesdl3<0NGxCoA$Nw@R z%LfhVm3Kff;au2}3|qUoRB8|;!JHx1JaDcW2cgaG$&K25KI6-(NlAkU*XVUN@*G-! zy2wAOgal-&Nqb82{ZoQq>XCX>KKOC+j}s=R=-Q|-!?P-BzydFLgqu_J*vB-E zV*A<+UGX&OThwQ92B)eCf(+dGFwyIOSV|VCOU`hB-q@PQI|~zskDe*mqm6fp2MH{0 zr`0u;4r%I(I-ej$Fr5^64YhH4pjWx;7w_QQcn9#5(*hN~N zhS=k388(TsFG%sE#?s^9d8*=y8(y_$>NvNirqOc=ll6Lm@I0)g!Mu;lOj6ZZ^Y)Lo3s>ffq;}{R z?X?5GK)xUxSGup<4z`eg8xg}e*Wqwlz3@=|%BuFxrP4CU z!*N7EKvTp(-#Z@h(jjl#7T6an(qG=hK!?HJv13&2yO5a?Solm@A;vC3=jrHlPP8^? z#T_pKBb`CEK+-DAC#hxjOyfa#@eVyh!8_yg*13E0M(c{==P~9J?AljI`<_qP)y%}U z4!=}n9zn@sVKlsdCgh9$5AyD01X2zI6p1Upk);WXDiua<2c)tq&)Hct$P+!y;AWax z>LgkiiUTE~$NDPxu~bQ}veugV>#28CI?3*Qduk5=R!M}x4eUAVQxcs50^AQ@)6@H< zXY<)EGQCGIj(n?Pxs-{HvfZZNtM$noKJ{0Lu-|BbM%*}m#wa###S#$a&ESt@RV<89 zit0(chD|R9nnSuMAs56I<&4vrASLxygjKW+}1kP?b$7b`E4?g!V&W!1n7knoUU$vnNasrc75czgP>c9tG!#0(Dwer zksU2U>+P?jpi}!HWMFGwCQfnSU=4Zdb_gN4EaiNKI>LHN6TeB{4=L;c zl~=ZqM-ewqzgFs0QESBXJjp%FFnG%0L#_hvV!+2oFI9S7Y~)9{`CW+rYP=T}8SnvYffODn{G zO;HtYc@Xt+(I8czWnFC?y%c}l&k(l>!Q0(#Tdw52utsOQ(B*{#1b3~9H5llwg_?a- zex!^jDgU$ZMpcW$_~4{@YrDxH3dqAwSkTljs& zU7c4rV~p6sbSppGn=F%mj}Iny=Y6$B=nh9_I)g8%3%ztXZPMDrR47oM zY7*{X4Ixo_%AF1kR;v;*^~kao@vC`h9LiHl`n0D;D+{y~0W6d)iQUZ~4omE_u9_9ykqDIMW zPOMeO*c7p8sEVt`Ap_$*C<|YIc~H;3&WrE?Vi#BVZSkU%>ymYStfbQQyrtWSEHW>gM@LY_dY&R z6=|LGK8UBgIlhF?@)%C`Q$iY>-SgM(g!-@Fif5ba#oatd%^W8b>1SVmBQH$u#)Rtc zN85sxy6~*Xv#)wy@c7I0A`K()gRb)zk?V){^}9c?!$1)Y?ySU?H2+2}8>rlV66BZa z+g`A*8XXn*GykC+Y|ex_O6LKLm!1-hTTmB6Uc_nnIhLOAT9=!Zjl_xI5Tm(llwd#L z*&{vOA~z$TnlO+hkXiWA=HZYy2~Nl?G0Oy|!XA1YX-wgH^kEf$#4*Kk=>oHv1LeWw zQQ7#4QcS1?-<$gM9%yC@>9S#uY6 zfr!g!%!Z8iP@9=#{T*pX8%_WTEmXH)#s$V{v~eDd7BPA~k}VGb(dYdFe4k;**N7Vn z_2D*$RC_O1uJFCfEEQN^@~Q;H9^J{!&Ysi)Hi{0uB)rCdnKJIX7|Mkg-p`F;jj(G? z8BVWXwK_Uvg&7I;c`uv`M0{VGtLyzEuF;0aD&i8f1$itXlTsoZ5g~2O*dcFT(=aLB zxf$hxv}xQJx-*6EI>^5ojhu(lz6^E)k|-TXEcvpI6zHK@11&at3z>-)W(PxXqEuvp zeIsQz4_~i;{W8*qUD}w}Eo_7Nlar2a%4}wwraAG%_`d^0Tp8}=(&MXR#kN04B$Go! zf2Qo2XW3UVfxZGIvD6j<#Q-rry7)6fZF-)inZ!tz^hM7OA866^FNezzN(VcB*W;<( zR1R>E)il9T`2rvSvm<9gwv3gBWI;N`4>2XgjQs_FlumfZkj_ve`A89p1teT$D@T?_TF$*XT2c4&C0Sw=fqz1km6o~b1^FnMoC;;d|MMJ<_rX#7ZUa(Z9f2`yu` zwC#X@8Y&GHuH}^PClH8;=AB(Y=cpYz=vI4UdB{5|DE}$G2GmX*js>mNzPw%}d_xVW zd{8vLjRhB}{MPk0d(3GNrOFnQGW4HsU_*&7VU3YK2s*lha~k&LNZoAgy->SnWgXK- zI7`RwUN9ib*$p?B+hVr?xGkUjbWi<%%!!qRlSl6VJtI`C40y^*&u{F|AQ+8o zM$hNy+<3n*?Zb)KKNw=!{aB2$%5MAXrVh8)_-d1wthqt_Vea}%Q9K)^);6iB8zHhk zlgW>AVQ4&Sy1}3|q{GnWBNUzK8J8T9Wz_-)>4oAKHuv-}&0x3`5oidaUhYm-oIam_ zOjR-K6TPiA}mlpUk+VZVmak)5xLRa1Wi3bI3G!s0P5?A9zM+ipmv ztMc6W@s}&k?<#R$et`0rsi<)BlINyj6bR1hFcyA=4{H3BQHLETmGsz$fzs1z)%(&) zKgEA;@nmNV$9a>%N1>$7i{P@UqQl^S5Zg#16nI516P*EGWCEA(*B2s6Lq&cMJO;%sG9RnF?M{;S zoyQlLv%!m|Z=DG|IvG3!B*`WV>d?(m++cEy`GzO!26apTwHdC6u0>^5Qfp0r-an>) zrwpMQc90XnCECVe)~Rf}mx5n&hlOFCD~kdOAV7-Wk3iYAAeol^Zs!h~*}BeU;&r{x z^cdUs{1E06XkCVAk0njzH(Cyva55);ejt}ciL>9wVIVCzS9%JL9pP5dam zp{N=Sj&+C*^VfNb`!q4B$u_+@&7j(?z}DZo`dQ27<5WqJD65bZc1NZ)qyF|)u09*mnf0~LX>O@Tbzm>G302pQ2L^b+QNJzb(&Vsfm(|~7 z$N-%DpMdEkN9Nmn)&+Eir;c`WZ@5|NB%YgwTPEJ;J~j)YMgyjQv?G{_Umum#_py5y zBK=jwbfa)0`q4CjX`tGzFgHvCM_ZhOBAiv2;#W_!3{M8jO9)8XYB6W|JYq&4GPtIL z*%`IrIxG(~p(4{p%g$O|eIaz}^K(RannW9TZ2oklKeZWg#rjb1Andc_2oJ{Cvz1gW zUPev&kKrXY9?Zyp5mFPBM=k1Uo-W%9!iC!i3%CxTuHnHK>OOhZ?6t)9{33dxToS7& zwNLN{&x{Z^bTTEQR4PITj!n^j+Je(~OoG-JfrW<;|KR|wp0j`uepWbgQ=PPI7xCK+ zLXyxm{yKcaq<2k*rE0q_;<~KDyiur4*?z8$h!AfCYcY&}pm6G(7K5c~P& zQOZsDNonnm!#dz2vD5Kn?1}Kr+r)Gr%Ub=^X>N%(A|MeD*~sav<1B{@El$42arb)lI7(@bZgrbvKR>w9?Cs zu^nB=C4;~0CPFa}FW^~fcUR?eJ55Y>qVDgO+1wi? zVI>81;+m?|YE#z^4_`hN@Js*6kSByDS|_mFJm0)4JCQ@<==*L{#wqV$Wmh`+p?R14 zzGΧuL6gAyf!ab0$aKhTfHX4YS``9Ia0^5_sD(gbjKm8^d2pUyC^m^-6;B?@p z?auc15fFI#Z9Q3U;lb`&WV7vV#fD)#7mQ|wSb|Bq#?6z~A{0)wQvvdjKBPjbL7V-WNo*;pVb)|9cF>@CAcVT%xnz112A-(n5F#^Xo3AMb;I zj&^b$TM>Vv_sZ-zrGPR2Th$c0W-ex}?0J~D$Pey#h?{$BW|E4RUZ3cgpzADp1Q7LQ z9RJ(Py1SyZK%gwnCq9a*9*#g4&Mw4R$eMNub34570p3g^Hq$&re$YzFow_uDcKF~5 zl?~MKL4Li7^%Hq@_~qQoi!<2PTa~bXlri7kkh;IMEyV zqv){Z2nrdLw`iVT#SdPxsXSUDV=dAJuLp*wn*G}&B7Du6E?qO^h^y`wEmtN!7E*Aj zGu(Zf)G^+DM$OL{YS1m#~G3^CF(Ore$~%9wJZuT>V`t+g;b z(AYWqp5jevv5d?(K2!tnrw+kR@Xm)Xje;vF@S{KB-1Z|uYvWpvnxeb z2>T>IFbx2xxU}QuD3{l|^FUB}spubrat!g?XYJG0&Hf@GId3f$_0|u6BU3%*$g{Al z4t1`EMtDnDjNsmhzUAT=e#GH_5s_|{8ptdyM{d}Q=b4>kq|_A6g0?hPE|$3OjAp;T zNyNpp9!_gp&V5CQ+eug59w==^XKbI$oQX{Ri~<|5$6jz!%ur8g(F!ct;aV@-T`syw z!+6|%ML&TyIPaxp%HKJESQt$5HvWA+V;gl_E*fELjLU9yXpsTwkJ;Mlp8c~bS^k>d zwl*{I!X8vXuq`uf@2`L_@Qs}~cbKj2su{>mqYz&1l2@z%owD0>&Wmwqcm{0lD78!L z#({%?<1%R%20t>!ov3fep=nhMi^&!#HKP=fYo% z2t_?-Yx`q!TLs|yU}H!qjJW0nh|#kc!*cle0i{T(Fr;n21I(*+N_P2GiabtRlcAWk zp$W>J+ush_tPppVu-7u89}B*nESX9CvUz7=W%+DZIG8vd4XkmfAFjba5K@q8i;^AgiF2q%+O^$R+a|5}=&ZVhWQXavdFj}Ge$nKBm5GivPYcE>^OulCZR zcamf!fBFjg1;T_ZQ%XS}6>GC4x|K@kNaec7vdoSaEwNo%*VQ&d_v2Kpaa7EqhWp>v_ z!LG$h-Xxa-8Y>fzA$lE5 zJ$PDwF~QGUB+)0&&>_}@uY|9utIDN!{nVht%T;1U$z=yhjnv~&RG--A6}+z1!$-$M(VbFTdgs@3L52fNV3Ofh~b*ik`sU z@r&;fHm|#kDdSK~Dh81nf3t30xD5%%Nc=;fm;mxApaa^AS5MVz^r_3Q-zyuv!jEQh z1l#$at3?&sDe;4>g?_aTVrFk-lrubk9hDzi0Wv{fqlRqzxj8Nh8~VJ?4XF9{ZIF1n zyZ$ap^kpgpR24OGIdpZi+q?9?;jV>Qa}t?&_Xo&yVzzY-=l1lO!3)2=&l64^9=Nj< zVF80@G(+}uk{17DQT+Yw8r?oGm@d8`i>cv3&bMnQ2xg$>=@K+ZH-?HOadYi|4KVM! zwvWZ$QY8MFD(A~)#+|-+hS%*sBaLrA!syuHO%NB{geZ9&kPbI{ctxDhy4wpbwQ~q7 zw{*QackrLDTnp}}z4u)ufI7y*Q!Im$d7W}Nx=uh#eQh5o_U{mg>=6W}g?eL3r>62%P+T}Llm>Gr}U54{x>KoHnBr++2#=}hLM zx~`2g{s}*UWu;*Ld7_}DS(1_50ui+QCCuQ~5!x)9_Ez&rd=e3bL}LVHC?G0DEHohT zt2)mau2DFe+(C~A{0N?oPdp0oiMh9Uyrce!EvCpH$HC<@9^M@nOD;2idA}U~hBIr= z`uq=ZLpw*cQ(3J*)QA4x5aw{3S*$_5P9=m=eQ(O$G;W^M=tSif9JKMnZ=MHhoyS{SzNtfZgl zLcL@jHPUC%HB%xo{$B#~?zP&EOFx~gzXvB7GvK_QFH4hBVC3X~HI!a*B?2}K-L%@d zD(51@CI((e^z47XE)iJpR_LFaS(uu0Q7kop%&kUbG*+r$QxQY{;Q8tf>S*4r1{0<% zFaW?e2`=Yzw;9*v{-}y`|6h(y~Zl^x*HYUs*`fE!Q__`NU zgDr3AjD|z=FP#^E{e}WFn;7|CHb1Ac?)Ag97zrKAhAqWI&O!+KG}_7t&U%Dm>P-qR zykK?yQ5-oCebB#fEmTv()_vRCCL>ZjT;T~MRMUuT3y=}tHM91c90g;qZ&!`gJ(XY| z`QMOEyU>ZTi6A}qBlJsna8-LFxzG)SM_dUtCF@ry*soB3TPFFQta7Mw#GI>;R&RDN zT+5Mfi!?vXtJJ^d z;!x^{@bEHgZxNSvq=9;bYFto53P*6v@lHdRogh&32lV&Q;PD%tr7dmN)7+{Na-dE` zS|}RwV12*+`n~++GX=Ci{+ABYQk-50a7PkZp!I{t(tYO#WOqOwj~P@?~1GEaov z9s04Xu6g09^Q z+9^|i?RkE6Y4;UoTKI$uxFG?-n3b)~oFYCbWB;wX|e>tq4j{?eeFyc1tpA z>PZ54@Oxk?e-x``@a#uO*tiOngPQ#oujBl8k zsbOYjX2yn-hOuGhY|nnXD;=GSm3%YSNcP28;~Dwq=OW7;*v&<)z!Et1m(wJ%XfPak zK*E(@ayb;L?6NO!dE5zYk(;E@QrgzrM%pZN<*B)`B%y44SyOj%!a@9D?8L5OsYdHR zuUCE9ZM=l=(^!-9J30;QRlx>WA1bV$(+O&4N>H!M-ZSy`K3HI^V<`H9^J?|-RjIax z%dK^fGwc38DsRJZuwnf46&OMvxnci2*Ojfi{%6NWw`MNX2xGJJ!9ddN#@F;LR}=ND zevq8kPhB-vS;f$K!urFe$EL@uHTH^wNCel-u9u=pNZn>-sd}G_6NV?*`xNi_Q>xhY zSYeuDEV-pF6bQFk=SNl?B%E4Er;yvkNlx@D-Fj&@01uvNAd>$F=^yM&KSp;ZLOMSMHq#`TpH196@1oE`=^G;6Tqq>4PY}Jlkgy`C&CyeF8s4bLfd2 zuA2UR(6#<=+D|e@t0RGR)>F+3NyHrB-T)mDHu1gN35_+46P>!1gr-{tDR&8z3ZW_x z1Nu>3LZR*JH)$ben5s&`w3&G5@~8Zp8_FFHL7KU=hvX-=8T$9}jL)PTH$D51K@$j4 zV1rw>WxNjRmgyGD__w)2#PGVHfNIkp{uGA=mZ+69?N`E`<>RKjK1>LVSpEssf*;aV zpZ&N_u52&ma8I>ZPAN0S6KCsT=986Qt_+Pjb$|8o2>JIdmH#f4LPJzyo$ta4;8lp! z&yBd*qv;9Q*7XZkZHi+VTcF>%3YzQP2ZDU^%{VRx3`lhc;{s-+!c={E#0uBohCY&k zWD*pu*^9CPAv`=sf%pOANmslC0A`x0DF6t9cI`w^bM%I0f-#RV>*T&Im=F9J=Hg}n zenhOHqcGJ$EmZ8LR0>AJoN7cpgDfBY{p1ZqUJ=EIebVq-Hl6i^Aj1YOB__rV__TM^ zpV`X!?#$C1he2hqb#4+ zxti$>yfsUb-MgBrhn3W;f}h?uzOwHqwBeK+kq56k)gnS??T^`2(B+{BxSeRlR}8lh z-o9`G`{|XfGSy>^xYqfFoqjuhfY}9U#+-Q#%T^-Ft+)w;ah%- zZKU03+^o;=x*GeN8xJBxCiSnsd|z_QBBvKiLA)?OlHf??U?E@+)=QJ=Ii?3TNpBVmaJkQNaKSjo?=cRYXePBc@Ovpq%AbU_6qz zdgoVq^4!lz_qNt=?v+)e750^~kL-L)M<2dpn^RM0-!DUX|Je(q0sRnz{}dHs%740ok3N*ueW=IK*OpVCBwk-SMtzTE9DPxh^Z&&7!B?g{!tP zH_mTdrTx<0XXvTdv9y;3y1-C@&M2~)yKA15 z+tq7V`r-EPlJvthru!rzL<#tB$${cLZ6IT07s}uVjO26ke^fdDWpazX=hj5yaK^GvexPPx` zIG~nMiD5-wB3!J3XCDHo0nzeVsMemjMdDDlg|dR#bm@CHC;h1+EGl~GD13MLWqLaR z-{O_sV0EB0dpy`4wjV_H{VdC`M$0ILt4v@xtFZgMTK}gFzN{PvknX3%^P_Rxai>tT25WSq%$`Tb>^>y_1tjf0OKUzPxG- z)=?b0t#~{G3MV#7XttL&4FDtjb$@q7MuX0PGeK}7Z7MvE;JM<{VzTj+;{G9q6~%|) zxD7%A!xqy^*5hFkMQ^t!FU5TM)ah}r4S~Om{2+e)^wnfxOElN)P*Z{U$Z%wMN zL=;I@E_2W1aoO#w8kwSqcy#boGgEY82oS(Z!XGW(To z#UuF2L;bTG9DzH8l7Z)EX5d-|m7a~y7punnNj6^BBJwhawQs*%d15qOp{%ksg5^yC zT=wO8FHvzf(l$26I7UUugYTY{f)50f|F<0F6DDWCo&m zpt-V*$2__LegH4?S|p5sOx3sShDIBh2{V^(?pr*|1DhfDvz{|zs;MOe9b$bYin))8 zzn<%Ci&~sIYu^n*bx*GgO2MMX-^{5$#m32;nXv5B9DT$ zZeuWCinrS(sNT6Rm}?Ej+0yUbK{Ho<`}}n67M0>$LAF?Sfa) zPvkTIUPksxC5sp#q*Z13Q9EkD*7;gHpD}u~ruV5+20v?CtZkUTmsa}?qxxB3eqH0_ zZD9`^uT^MEl5g0*$e)n4diMP_r+E_gT&<9saY`>_4_7#jRHT*vt6HBqL)1_Dh3_=@ z>i5s5mSlpCtdPZWm5cAN38CJ%Iy-c;9RsCS#kRsZBu(xe$Hl5^O3&QDQ?7VQ^E|}@ zX*t6D>#NIPczRu;xgTB;|FE=Of4DNvulAsgl_gp{y(k0EzdaxLmAXl>X`}P5#$or( zVpP?vFJ6g+p~~t?C8b$URF76I*;jTlp?l>ai|r_FhGz5gE^k2&QI$%V`$!z(+0v^s zR48{hM}>LP;|g+F%FAGYVC{{Mf#HcR7or#eMJ&;m`^8U-qrt_2N%!G=%38-i%dG#P zQ)EiaLE(f={2HX<@+Px3a5yW^c!qX+lw-m_V`ATbf6;`rK1dG;@WaoUb=2&o)!R## zDEBn{dR-sZj3J<+zD)ql7H28lIj=}=<>HHUoZEb5o`w=rim0CfbkF*$a_H}SLytx< z9r+WAM3KyCS(t#!LfXrIRIt&f*;z!sD!eSXDmJ~S(kAJ;TKtt1YX`x9i8MNAtHTyJ zucoE}Y&1>#pefJV6SpyxQhthiX``|A7xld8qApeE1_&bA8$tk}yNafL)iV%y0aTgKJ~ex9FZ3#jWb_mtJ}C~=!wN~isi z4qXoC4%9F#P(fOR6=|j9thP3Qe9^?hm@pQu6o#D{{04@MWu1P+KR+WF>6`B?6vBAH%*VUYQA#Bl-R|d z(*`zkJ`wx5O3jIe8GXEU;3uz*oI$By)GOkpus(^nZVe<3I7UHByl3JZ1@L^8^~&c) z?%>vuA16aAZP=mA$Sb4{rV)?C!y&5z(;cq9QfS2ld@j;3CF|z0VQ?aNOLi2#gX;4m zrTXx}2ngS*ctf`|rdxg|Bf6HrC>US(o-Gu=y z0t10Jw5kbV`9syHhjQDt!ApJYcRSiPw_mW?3)7ReXiDBIMk617T( z!x2ZiZ$F@qGrTNgAtKUAUjXo7KBT#>2z+#imk}tC$)Ll|iv%VI2Gn@+VDEz3zvfVZ z=TK-lrp9on9@|EpfNq!-N}cNwpJ7gP@4#|EWz43h~VTx+LyF$BHPqbt#PZ5((cFu)6PBW`#&OhaOUmevW zQ_^f^v)$@v*ILsp>4kGrhB%pImrDDE{>ITu%9eH zhR-vu-aOh<$M{x?p}qbzx+*TITq~c9-+NPQWf15UDhtV!hzm~0Dv@bQad)c$jAAL{ z^Jer9=!_B~Ri)78l?=i!71jvcp4??N14%YQ>2vJJ9g_stvxJ;+f^o34m7;%_f39fK z!M|H|Cxfu6ZNzdBu$FU4$~awIhO{griVIv%Y#v#`6ZRcBy)zRO2yg#+6Hd5l;>{?EL8dhcrF5d2Om2SFNCkd95lV_%TdT`OQfV>HLAF8=B$F zln3J6FWixy1G}!O7L?-{+^`-Ob+TnKn`>1aL2EEOp#-rmE?P2$DR@0e>CMMzX@|tw z9ej&J$d~(NYA_{J6dh{CRbBr(VPSz%exnnEsWiR5SKkvmS1=GS9ogeC|}ez z-p6vBA`(x)aF!!kDv2ZwNri5ZFaLU?XQgLV#sY=Y0P9rWouyZ7G_bRvXDT-npdv?b_Ocua3GLbu@>%s!jD*3Ia2Mb zpees_ak2bgz7a7S3kN9my9YQ8psnS!E{^VdRFgL1-ZI%%!=TaZEgFL{p=>fzZ~WK4 zm^d6>YLpM(3wU>ZM1u8m53AQ}V$z{NhT-4j-!vcs1D7;SGgO8#+s-9ToMtJbQ2EoZ z#KLf0n(kU#jl6*7-e@X5pyZ!6a@%Y3n;)L1Q&ARoth2A+hk!em6*eXzfQ(qWrHKtq zNDwKMINb1UyVPHTda6(?GwECkQ9CJs4N*0iZn|B&b@TSvoa^hvV0cM?n(8vWsl78 z#21h%YiZd?5R`<2kYRWm06~iX*E_$&afkw>7#gbYs1_R2#B4UF4)D6l`r_;e1z^rlC`57Ex?(;8_o6Sb)i#IgZs}SY$!qG7dwa;r7wXcpc7KI=D{J! zV)H{6TZrffw2B>(58j5U}Fgnb4!azf{tdHMv+GuvP2L22D1C<5w{cj z?-I~Ku-Hfeh4_fzY*=t}hU#_HESBVv2r4wfyWyk6`*S>Cn4QsJo~ZcY;Kr~gIAoL% zhb3QnOM8vsyYuKIfdVe$HF&XHq2r(sC8Hs@=nAkP%wf_y!^m3gdqTeBYpSzfP>4{x zMsHK=g5S1LQ0@u7@#)2kn0FBYAtydHz_0C0tX)?Uc2{Db5$ug9pnL<=eDsNzx9dG8f$@u)Ze5LDut7oqDMfW* z2ww(u@$N|qRFm3Z1S(+^DjYcVW_MQw%r4_b{BH%y*;eeBA+-XMp0cM;4?u%~(>G>t z;Z`CLbxqwwxpus98~cH(7olKU1+rV=I%#O;RIhPiE%i97%MY$DUlx4*bT`RtCkuK#lyhirAM$(B^2R9H#TFW{J%QKW-T`!%c-a897P2G z#I?z&)hO@oys3*mbtBa*4|Rzbi9`?`zYnwgTnoAtYmt6vQ=X zj{G|IHA`4wH;%*6@m1b7gvjwuP>F> z6P1V?K>au8P`$Gk#u{nniuZ!%-3-h&F{Q*-*KcUc(OlG?xEm=$5ZY;%v(!V_ca)n~ zs(P7tHkY^zZF%k4&I^F1cUr2E!hsc{(1h(uWxt~ybwjw*c?LYWo+RsYugjXC9FoBaWQidxf0c+QgRaPmBz#qOpPdo0XLXSG`XT zp*ycSd*>31i#0PydBJ45NkbjBbu3+mZs`LAhE)L%CVB!FBaEs@|KyIzxGj^AHIooJ zV2Fh#+%Xg9G;@+UbFy~xc;=Bs1W3ejrNfnz{_8WjGV76(YV#{;^`T;~YSp}DqboiJ zb@QgQNH%#K-67)Mr?siqDV@n64IF{pIy{Y1mwz3dMQrLi8Mz=A5`|ly`@5NLG?&_V^|895+QYQ@DXnk=;TT@!t!@ zTxGAG!BE8M=FRao1twX!pNI4XdKR|yapm+cab(G)>%uX2G+MhwAjjk6Mkkm;AX=mo zDR+lhFpHAjB8tQB-g3~q1lRWpcNAwmzesDGgaM7f_{#`8sf3ptY=7WkYx6%DZZXS= zbu4yV5?w?j*SH@h!%Iv(-w#USLxeq|VDNR)!2vLAKXp<3{J3v*VeV4jVNKwnVo^ua z4#EAJaq|7}`wZJ=hm~PSspu^300_34YwnrR?7_PHN(Y6x38ztMZ|CUdJm-8fzOV?? zh6RrZu@t(_Ry*p;=%RSig-VJf2bj+lk}D6^JA;s!ZZzlv74Cw*6o|B_8Agl3LNqZp zPMdGqzu9FdGP&VO|83knx^)46`QcrIfxy?6J$sx(iN1UOdut~tt=JMY5J1|55J7vh z)1XFAGY*;HI?JB#?7_bv$ziQvnXD3XM@H2Bj#h4Ce{G3bZiheBKyD$12a%6sEPYYv zfg-bW3?&0{*jHR(bla6QT2h7PCazlC7s@D!cOBLTXa0h@_NY-%v^y)F7tC(Iu)wE? zwcVEIw6JTj(3bh~qltQ2!dU3ilzJK~|5Cjj#3{M4XQ!&8P+os7CGk;wJh@ zPK=ey0QG$teto2cWcq6en2=a|=L#Px&yVXJF)C^+w2&tU<8b@idG=_=ryotWvy!If zKN`&N^?K;VQesZpA${h|q2e->B!7n)=NFd7>F&t=9WIWMzYU380?>sm<@dDyKGO?4 zcq>nZHq71UXBUm+GQnVIa=)FQjGYVs0)l*HRDorfk3@@mPKMu58ILB(rzP%MAG+aV zHZ`8GzjP!fY3JNO4LpA#T5td6^>)4ea%AP~k*we-qT9hlM(x@BElOe_bYa4ANHn%b zS2b}t7fSt)3CXDOEZ_)A+YwnJdJ_2v`W!@knt8(k`eL70fH8yXt!F=ATSTC8RUwYQ<}n4l7hrS#ByiIb%w#7@u8zjon!+J{LE9bvCv;H&6^H{C->NI>wR?4aJ^ z@DSz{_Eeni@W7B2c(NT;^14mGtDZ%4J%V#${>Ci`m1S_f5XHKS*h*eXa#h`{rItL@keig>B z(#pZE?W$4>fFIFN>@vM+K+sU^+rN=h9nb&+TQ`Dm&8QRCt=dWW1WSoyxyk?7NZ!mn zyNgFV?#z9iLp@4~-FuXQ|1zh+cjcI;9O$?8#OL;mydHP_#iKW}!E$}oL9qtcaRu?Q zVan$tWAmXD-4mVk*U9J#L^Xut0%h4QuY4 z&&d3!Ef$$-XtTY_air9^64l9DgTWk2(u$+%irqy?=yJDXG`jE~&0_<)u0_H=R$&X` zN9cWHcau=1$zxN{I(BppisUkwSx42Ht9Fy}sr3Yu%sw>GMH|p}Rg)w`WEN;7EtpYs zXI%hj`RJky2U&J_eRbe8_w7$t{6l5U>FQ9hbp2_`Q#m8CAk%B&?5OlzZD)PLP5Q7s z_?5GK1r6(T2+l6XiDrOczWibPuu9K5a~!kXA%vJWgI_v3w``xLJt2$K5RUX`8UhCh z6T(bwgnMsWonr@~hhDlNMCXLJv2nSgcsL)h>WA;KS{vo)Dst%?U5*s&(1=j)hUCchO}27X)S5@m8CwLN44bfICU&`QCA#U83T7C-yQK zv2d&)(LP3*(8<6r`YHNCFJDNXeFlNR&ISa!?qqU z&a_%B?hdtGuOa7@WxGMC4Zk#6dr0wFVE;EdC9KoU@D^0x3?wiSYHabdvG(3F9P#WR zubtw#eVOV95#663AO!=+P7EZD6t{xlsdGiz*Ucz>VvM9{bT_UX9zf&9^C+CdCuu#; znm!&)FTIq2YP~gLnL1DMD9xvcXguG_uv zR2QIFKs*|t#zQ;*bUDYDyZh@JVrY8+}HSW_!@o#Tt|mOc^`^&?^`G5al1UUw^{?B>Tv>aWks=I3>VpA4C<@u#hlMG_<19%>{Zg#*sE1S# zf^(eZze!4(2i2h{psUx((xTgDoBRkSyP}ZhI*yZ~SMHRFeAE<)Ut}f+`i!oPIV^KS zay4+t3R%Gl`YMwL5LJHplBUS|1=ljDjBFPYhx@RSOB&_u?2Z7v=HPL;arkvfSV=r) z@|!BnQ}P|$I6uk(W_k$J&tl7E-?tA^(|{BB{r1u@Z+RZ1qYApQ?IRhEW-BYu%-hCtjQYdW|%T9q@2 zP9o#0VygaeM2mB}RO;H&qD*Wngk?tIo`j`CqtK5d%ho>p`E5O*I}zq&K2s`{UM@zw zQ>p=cWmQVsUIz7or-h=5((-w9lUy&zPRWI=wr_K8_Lwo4axZgike}W1>y;&>Z-Q() zA%|T4VS6W=KO_H1erWq{Fk?rN^(k_uz`W^KtaO|EkYAnyFkMDpyNvEr?tUZx3qjQ8 zb9K&@GQ6_^J7zmzN1~G&mG?4aZno=&S%ovrWW5lUo)avy3~#a?Pcjp_6zzI8zwm2r zA^yGZO35Vb(v77Kn1IUQM+&UOEwd6UXVlB=rdmIfw zr0nW8`ZvX&tAv%7=wNfQ`M_hG35_ypJ+EQPsdkB;)5HQdGWtXY1+1S5sZ%H-c80H6 z=b!vb4sTyfpFpn=*R=W(K1o1FisC%>V(_Bp!>*H->IhFqwto8QKwS;vYrVCndVz4L zc&$@&vP5Y9s@qiV4`+l%j*he2Uk-nKKNxroCioKK$M*(eYLF(7BhIe9`$Ld8oJ|;! z@ly*^@F;=$MVzt(D`@W+ntcQA?Z;Kovxbj=oALbAd=u2Hk^`;v#mQoEX-6pk3Mpyr zxUd~YDSkx+fB9yrZY-gm1&zF!51A3od~1EqW$!I)1I^&;Oe##M@gwwKEs6>E(lVDBU}Ug+Fj0k5;DqrnKBl^~+} zZ(sbNuC=*EBeg&gM(ihrp#5AQzk4s=r ztY0+@{7*ScgLmj|;-}|Rly@xkTKP&&#A#N`eMQdHNm%;;`CxB(MnF@vh}@3;@1-=f zG#Q7Hilbjf_~NrJyJeDx)f&22*5S$a=kOYV7%&jhtASP>&E<&Myh}%f7!Mpvt=`@e zzWk=PL5N#i1qW7`lZ~MpoaBOEVQYbep4DqX&5)yQ&@+zEM}p8_54N7}ki=QK$3@Wr zhOSS}$*abXlW06Qw=`Rh-O@kCO@tMK9%J)+jL$i7PYoH~xT6^o)mthLB=)^BuFnTg zo>N1&uSil=F>i}MML}^7TNFJm49c4-!?C&Dx}$%(<71v{7D<4Qjxh|jLCnFvd#&z(17RMqE}j;`ybGZ(B=8d3)(C7ERF| zlMIcF`V&`@CQcH4QOb&$BJ90lY2La+s4XOB9$JPW;y(UQ#EX6(fxkiZMJaz;fuC%R zz3LXh92TJ{D=_Oug3el}f5V+R@?c$sn9hO(R-ikKP|t#WnBm@2Xo+@dVGY4h`boHu z3KZ+uKNw4pv(7JehXlG>w45g9)O$db@|qy>O=IEAhf{&WqJTgwja6` zE_It#*Pm85*qYh&M9N2(iO})w0%$B6BD!9spX}qK?yWC^Wz7^=dTFsqc(BGfnDFt^adGAf^1$eJ6AEvleQbLr-ikEN?p0M56Fz!Qb@ zlotM2hT!o2-)aLV3XPPzH(aUxqDF+vWNh*2Blfcy7E^m}@bXYcj!ddl<+rVZ5bA>K zt>*I>C)}4&#|n2x*X+ltSk0X}`BG?kBMZ5rO8Fxgq!RH3BPs<_bA=;`STQry1o>6O&z!qR zt2p&+KCx14Cb1-J2C>hq+Oc!2YO!UkivK7=GO=2$;;}5O--&8(Kn`_Fx1uEPu(AeO gTmSRuI5`{rbaww~YKFkY@h|vU5GW|b6vPq!7aoYJ!~g&Q From b0adc3b6df97c410628b220a79e39867ce5bff0b Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:12:56 +0200 Subject: [PATCH 3/8] Divide manual into three separate files --- analysis.md | 123 ++++++++++++++++++++++++++++++++++++++++++++++++++++ theory.md | 113 +++++++++++++++++++++++++++++++++++++++++++++++ tutorial.md | 111 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 347 insertions(+) create mode 100644 analysis.md create mode 100644 theory.md create mode 100644 tutorial.md diff --git a/analysis.md b/analysis.md new file mode 100644 index 0000000..b0df084 --- /dev/null +++ b/analysis.md @@ -0,0 +1,123 @@ +## Analysis of examples + +### Example 1 (3e0d) + +3e0d is a small complex formed by 2 protein chains and 2 double DNA strands. It is a subunit of the eubacterial DNA polimerase, but it is perfect for an initial testing of our project. We can test how the program works, the time it takes to run and if the result is similar to the original one. In this case, we have 6 different interactions, as each DNA strand is counted as a single one interacting with another one and bound to the protein chain by one of them. In fact, this complex can be thought as a dimer: two monomers formed by a protein and a double strand DNA, which interacts with some residues in their protein chains. + +The first test to our program consists of joining the different PDB files into one single structure, without further information about the stoichiometry. Therefore, the program will try to build the protein with only the information provided by the PDB files and the fasta file. The command to build it is the next one: + +``` +promod -i examples/example_1/chains/pairs -o examples/example_1/results \ + -f examples/example_1/3e0d.fa +``` + +We only need to provide the folder with the input PDB files (`-i`), the desired output folder (`-o`) and the fasta file with the sequences of the proteins (`-f`). The program will take the different PDB files on the folder (by alphabetical order, to make it reproducible) and the result is incomplete: + +![3e0d structure obtained from: a) rcsb.com b) promod without any extra info c) promod with starting point and selected stoichiometry d) without stoichiometry control][3e0d_composed] + +We can observe that there is a missing double strand DNA. So, we will take a more elaborate approach to complete the model. As there is a missing protein, we should force the program to include it. There are two different ways of indicating this. The easiest one is selecting the starting complex of our model (which contains those missing parts). Different starting PDB files can result in different models, so, it is important to select a suitable one. By default, PDB filenames are sorted alphabetically and the first one is chosen as the starting point. As we said before, this is an important choice that is evidenced when we compare not choosing a starting point (Figure 1.b), choosing an adequate file (Figure 1.c) or choosing another one (Figure 1.d), which in this case results in models with additional or less chains than expected. + +The other available option is to indicate the stoichiometry of the complex. We will focus on selection the starting PDB file, so the running command would be: + +``` +promod -i examples/example_1/chains/pairs -o examples/example_1/results \ + -f examples/example_1/3e0d.fa -start examples/example_1/chains/pairs/3e0d_ZG.pdb +``` + +This is more similar to the original structure, very close to the initial model. However, this is a very simple protein, and bigger complexes may be harder to build. However, from this example, we have learnt that: + +* The program can build a model without any indication. However, it might be incomplete or it might have additional chains. +* The starting PDB file is an important choice that can influence the final model. Thus, it is important to remember which starting point gave which result. The same starting PDB file will give the same output. + + +### Example 2 (6gmh) + +6gmh is a very big complex formed by 24 different chains, with a double strand of DNA. There are several chains that interact between them, but there are not repeated sequences: each monomer is unique. + +![6gmh structure obtained from a) rcsb.com b) builded without assistance c) selecting starting point and stoichiometry][6gmh_composed] + +Our first try is the following one: + +``` +promod -i examples/example_2/chains/pairs -o examples/example_2/results \ + -f examples/example_2/6gmh.fa +``` + +In this case, there are several chains that has not been added to the model, and some others are repeated. We should select a different starting point and limit the chains with the stoichiometry, so there is one copy of each chain, at most. + +``` +promod -i examples/example_2/chains/pairs -o examples/example_2/results \ + -f examples/example_2/6gmh.fa -start examples/example_2/chains/pairs/6ghm_VA.pdb \ + -s examples/example_2/6gmh.stoic +``` + +The `6gmh.stoic` file contains the stoichiometry of the protein. Basically, it is a csv file, without header, in which each line follows the same structure: *,*. Each chain_name must be in the fasta as a sequence identifier. + +This starting PDB file was not in the model in our first attempt, so now we force it to be included in the computation, and start building from there. This way, the result contains more chains in the model and it is more similar to the original one. + +### Example 3 (5nss) + +![5nss structure obtained from rcsb.org vs composed one][5nss_composed] + +Here we want to show how to use two parameters we have not shown yet: distance and threshold. The distance is a value that avoids the situation when two atoms collide, which is not allowed because two very close atoms would have high energy. So, if the separation between them is less than the indicated distance, it is considered invalid. However, to avoid some errors, up to 10 colliding atoms are allowed. The threshold makes reference to the identity percentage between the PDB structures and the fasta sequences. This allows certain flexibility in sequences with some mutations that may alter the conformation of the chain. + +``` +promod -i examples/example_3/chains/pairs -o examples/example_3/results \ + -f examples/example_3/5nss.fa -d 2 -t 0.9 + -start examples/example_3/chains/pairs/5nss_NG.pdb +``` + +In this case, the result is quite good, but an important issue is the speed of the program. As the number of interactions grow, the time it takes the program to complete the modelling also increases. However, there are a few considerations about that: + +1. Adding a new chain that satisfies all the restrictions is the most computationally expensive situation. The number of attempts increases with the number of chains in the model. + +2. It takes less time to discard an interaction that has not an homologous molecule already in the model than adding a new chain. During the first steps of the program, until the model has grown enough, this is the most common case. The number of calculated alignments is relatively low as there are a few chains in the model, but this may become an issue at certain point. + +3. Once the model has a considerable number of chains, and particularly if there are several copies of a monomer, discarding an homologous protein because it does not fit with the current parameters is a very time-consuming process. The program will try to introduce the chain using all the possible alignments whose scores are higher than the threshold. This is an important problem for proteins with a lot of copies of several monomers. + +4. The stoichiometry must be carefully selected to reach the desired structure. It may happen that the program cannot construct a model with the desire stoichiometry, but it is also possible that the stoichiometry was not appropriate for such case. + +### Example 4 (6om3) + +In this last example, we will cover the software behaviour in relation to the number of input PDB files and the influence of stoichiometric restrictions. + +![6om3 builded from: a) rcsb.com b) promod with stoichiometry][6om3_composed] + +``` +promod -i examples/example_4/chains/pairs -o examples/example_4/results \ + -f examples/example_4/6om3.fa -d 0.3 -t 0.95 \ + -s examples/example_4/stoic.csv +``` + +We can see that there are some absent chains in the protein complex built with Promod, which can be added using custom values of distance or an appropriate starting PDB file. Nevertheless, we will focus on another topic that has not been commented yet. + +If you provide a stoichiometry file but it does not exist, the program will abort. We could have changed it to a non-stoichiometry builder, but we considered a better approach to explicitly state this situation and let the user manage this type of error. + +An important thing about the stoichiometry is that it is critical to make sure that all the model is built following the user preferences. For example, without any assistance, the modelling in this example is good. However, with a selected stoichiometry, the result has some missing chains. Here we want to stress that the user should carefully check the stoichiometry file to make sure the chains are correctly selected and the number of them is adequate to our purpose. + +## Limitations + +1. The sequences in the PDB files and the sequences in the fasta files must be similar. The program admits a certain degree of tolerance (which can be adjusted by the user using the `-t` parameter), but selecting sequences in the fasta file which are significantly different from the PDB structures will not allow the assembly of the complex. This is particularly difficult for PDB files in which protein tails are not properly modelled and, therefore, they are not present in the structure, but they are contained in the fasta. The pairwise alignment with the tail will yield a low score, resulting in an incorrectly failed homology identification or even returning a fake corresponding sequence in the fasta just by chance. + + Our recommendation is to prepare a fasta file as similar as possible to the structural sequence to avoid this kind of errors. We provide an additional script, `pdbsplit.py`, which can extract the sequences of the chains inside the pdb file, preventing this kind of issues up to certain degree. So, if you are investigating the effect of certain mutations in a target protein, it might be adequate to prepare the fasta file according to the PDB sequences. + +2. The running time of the program is proportional to the number of PDB files selected and it is also affected by the order of the structures and the starting point. The most critical step is reading all the PDB sequences. However, as the number of interactions increase, more chains might be added to the complex. This is a trade-off between the number of interactions and computational speed. + +3. Selecting different starting points can produce different models. It is uncertain to state which is the most adequate starting point, as this may vary according to the goal of the assembly. + +4. The fasta files require to be specially prepared for this application. There should not be repeated sequences (with at least an alignment score equal or higher than selected threshold. This also applies to sequences with unknown aminoacids, which are usually marked as X in the fasta file and will cause a mismatch with the sequence in the PDB file. Those sequences will be usually missing in the final model. + + + +[3e0d_pure]: examples/example_1/imgs/3e0d_original.png +[3e0d_fail]: examples/example_1/imgs/3e0d_fail.png +[3e0d_builded]: examples/example_1/imgs/3e0d_builded.png +[3e0d_start]: examples/example_1/imgs/3e0d_start.png +[3e0d_composed]: examples/example_1/imgs/3e0d_composed.png + +[6gmh_composed]: examples/example_2/imgs/6gmh-composed.png +[6om3_composed]: examples/example_4/imgs/6om3-composed.png + +[5nss_pure]: examples/example_3/imgs/5nss-original.png +[5nss_builded]: examples/example_3/imgs/5nss-builded.png +[5nss_composed]: examples/example_3/imgs/5nss-composed.png \ No newline at end of file diff --git a/theory.md b/theory.md new file mode 100644 index 0000000..aa288eb --- /dev/null +++ b/theory.md @@ -0,0 +1,113 @@ +## Background and scientific explanation + +### Introduction + +Proteins are the executive molecules in all organisms. They perform a wide variety of functions, from small compounds transport to signals transduction or immune responses. However, proteins do not usually work individually in cells. Most proteins interact with other proteins (protein-protein interactions, PPIs) or molecules (DNA, hormones, drugs, among others), which are essential for a proper development of their activities. That is the reason why collecting PPIs can lead to a better understanding of protein functions, biological pathways and mechanisms of disease. The identification of such PPIs has been a challenging task for years. Experimental methods provide irrefutable evidence to test interactions between proteins, but they are expensive and time-consuming. Nowadays, computational approaches are being developed to reduce as much as possible the necessity of experimental data. Some advances have been made in this field, but the range of successful organisms is short and general frameworks are lacking at the moment. + +Different experimental approaches for the identification of individual PPIs are available. The most accurate methods that allow the determination of the exact atom coordinates in the complexes are X-ray crystallography, Nuclear Magnetic Resonance and Electron Microscopy. These techniques are difficult to perform, they need very specialised equipments and a strong data analysis procedure. There are also high-throughput methods based on biochemical properties, such as Tandem Affinity Purification, which can detect multiple PPIs at once, but a high rate of false positive results can be expected and the information is not very precise. Protein microarrays are also being used as a screening method that can be easily automated and parallelised. Other traditional techniques, namely co-immunoprecipitation, yeast two-hybrid or pull-down, are still being used as well. All these approaches provide complementary views of PPIs and have their advantages and problems. The main one is the cost-effectiveness of the experiments, making in silico approximations more feasible and adequate for understanding PPIs at the atomic level. + +Regarding the structural properties of the proteins, there are three main categories of methods for computational modelling of PPIs: homology or template-based modelling, *ab initio* or template-free modelling, and hybrid or integrative modelling. Homology modelling is based on the fact that the evolutionary information in both the sequences and the structures is important for PPI prediction. The latter are preferred, as most existing predictors use surface patch data, and only the residues in the interface have to be analysed, instead of the whole sequence of aminoacids. A pitfall of this method is that not all the 3D structures are available, although public databases such as PDB are unstoppably increasing in size. Template-free modelling needs more computational resources, as it explores all the possible orientation between the interacting molecules. The combination of prior knowledge about the individual structures of the components may help reduce the searching space, but still this approach is more challenging. This kind of techniques also require a careful evaluation of the results by various means and refinement of the candidate models using biological information. Finally, integrative modelling combines experimental data and bioinformatics developments to narrow the possible complexes and save computational resources and time. + +In this work, our aim is to develop a software that builds protein macrocomplexes using as input pairs of protein-protein interactions. Interaction with other molecules, such as nucleic acids, is also considered. To do this, our program is based on homology modelling. Therefore, the homology of the different chains is analysed, tridimensional structures are superimposed and energy levels of the final models are considered to propose the best possible solution. Another option would have been template-free modelling, but the required computational resources and the steps of evaluation and refinement excluded this possibility, because of the limited means and knowledge that our team has got. + +Here we propose Promod as a first step approach to protein-protein and protein-nucleic acid complex modelling. It is distributed as both a Python package and a standalone application to be run in both UNIX-based and Windows operating systems. Several parameters can be tuned to adjust to the user's needs. The final result is a single model with the best option according to the algorithm that is explained below. + + + +### The Promod method + +The algorithm used by our program can be divided in three main steps. The first one is the analysis of the homology of the subunits. Then, the superimposition of the homologous structures is performed. Finally, energy levels in the models are taken into account to discard unlikely complexes. + +#### Homology of the subunits + +In order to build the model, the program begins with several files, each of them containing information about two interacting chains, whether peptidic, DNA or RNA. We can overlap similar proteins from two different PDB files to check if two chains in those files are alike. If they effectively are, they can be joined in the same model. This is known as protein superimposition, and the whole process can be compared to a DNA sequence assembly. The main objective is to recognise that two sequences are the same and put them together in the correct spot. + +Two chains in different PDB files must be homologous to be considered as part of the same protein and overlap them. To check if two proteins are homologous or not, a pairwise alignment is performed. This alignment consists of calculating the similarity between two sequences, taking into account both mutations and gaps, and then returning a score between 0 and 1. A score close to 1 means that both sequences are identical, therefore high score values are evidence of a large proportion of sequence identity. Only the homologous proteins will be overlapped and used to build the model later. + +#### Superimposition of the 3D structure + +Two homologous sequences will probably have similar structures. However, it is also possible that two proteins with lower identity score may have similar structures in the tridimensional space, as a result of convergent evolution. Therefore, when two proteins are significantly different in sequence but they have similar structures, the may have the same role in our model and it is desirable to consider them. For example, two homologous proteins from distant species whose alignment do not pass our homology threshold, but they still preserve the structure. + +In these cases, to test if two proteins really have similar structures, we need a measurement, such as the root median square deviation (RMSD). First, to calculate this value, we have to superimpose these two proteins. That means placing the atoms of both proteins in the same coordinates and orientation, to check how well they overlap in the space. Similar structures will have atoms in almost the same positions, while different structures will be more distant. This similarity between the superimposed proteins can be used to calculate the RMSD, which is the average distance between the superimposed atoms in the chains. A value close to zero indicates a perfect fit of the structures. RMSD will increase when the differences between the protein structures increase. + +Due to all this, RMSD must also be a filter to account for those structures that are different, even when we had considered them homologous in previous steps. + +#### Analysis of the energy levels + +Lastly, it is important to consider the final energy levels of the complex. A good model should have minimum energy, as functional, naturally occurring complexes are the ones with the lowest energy among the set of possible foldings. Situations such as two atoms very close to each other, aminoacids with hydrophobic residues located in the external part of the macromolecule and several other cases can increase the final energy of the complex, which should be then corrected. + +After that, it is possible that the position of the atoms inside the model is not the most adequate. Sometimes, despite having evidence of interaction between two chains, collisions or clashes appear when they are joined in the structure. Taking this into account avoids impossible models, such as those with chains crossing with each other, which will be thermodynamically unstable and unlikely to happen. + + + + +## Features + +The most striking features of Promod are: + +1. Building of macromolecular complexes from basic input data (pairs of interactions between molecules). + +2. Optimization of the final model using MODELLER. + +3. Graphical user interface (GUI), with the same functionalities as the command line interface (CLI). + + + + +## Implementation + +The Promod package is implemented in Python3. It strongly depends on the Biopython library, which is an installation requirement for the correct execution of this software. The GUI has been developed under the Tkinter framework, providing a user-friendly interface and, thus, avoiding the use of the command line. The implementation of the GUI and the CLI are independent, so each one can be executed on its own. This software works with well-established bioinformatics formats, such as FASTA and PDB files, so no atypical formatting of input data shall be conducted. Additionally, two scripts are provided to divide a given PDB file in separate pairs of chains and join them if there is an interaction between the molecules. Some examples are also included for the testing of the program. Promod is freely available from the following Github repository. URL: https://github.com/Fabian-RY/SBI-Python-project. + +The formatted documentation of the package includes dependencies and installation instructions, examples of use and a full tutorial with sample code. Each function has got its own documentation that instructs on the particularities when importing them to be used independently. The users could learn all knowledge of the library by looking up the detailed documentation. The main functionalities are further explained in the tutorial section. + + + +## Discussion + +Promod is a basic tool that relies on information from both the sequence and the tridimensional structure of pairs of interacting molecules to build a final complex model. As we can see in the analysed examples, it is successful with small complexes and it can even handle nucleic acids interaction with proteins. The MODELLER final step adds an additional refinement of the proposed model, returning to the user a good quality structure. All this methodology is built in an easy-to-use package, well documented and with a GUI to save unpleasant command line work for the standard user. + +However, the main objective of modelling software is to provide new knowledge without the need of huge amount of experimental data, such as the determination of all paired interactions, in this case. A limitation of our program is the necessity of input protein-protein or protein-nucleic acid interaction. This type of software is being broadly developed by means of various approaches, which are proving to be very successful. Both docking and homology modelling paradigms are being applied to achieve this goal. + +One of the strategies being exploited is evidence combining methods. The core of this strategy is integrating evidence from multiple sources, including them in comprehensive databases for later integration. They are called gold standard databases and they contain information for both training and testing of the new methods. The annotation of paired protein interactions goes beyond structural features. For instance, evolutionary relationships, functional features, network topologies, sequence-based signatures, structure-based signatures, and text mining information is recorded in these databases. Finally, machine learning algorithms are fed with subsets of data and performance is measured to propose the better candidates, depending on target species, data sources, demand of accuracy and coverage. These evidence combining methods are performed repeatedly to find converging results with different input data and classifiers. + +As we can see, template-based methods are leading the in silico modelling techniques. It is reasonable to believe that there are a limited number of possible interactions and that, once we have big enough databases and curated interaction catalogs, most of the PPIs should be easy to model with high confidence. However, *ab initio* modelling has also yielded promising results, mainly from competitions such as CASP, CAPRI or CAMEO, where world-class groups bring their latest developments to test their performance with real problems. Of course, these solutions are computationally expensive and require a long time to return a final model. In addition, complexes with weak interactions where the conformational state changes upon binding are a big challenge for docking software. + +A large number of information is nowadays available from high throughput experimental techniques and a lot of structural bioinformatics software has been developed to integrate these data. The best performance of in silico techniques has been achieved at the tertiary structure level of proteins. Nevertheless, quaternary structures are the ones responsible for the majority of biological functions and their knowledge is essential to disentangle protein interaction networks in both physiological and pathological scenarios. It is clear that hybrid approaches, joining atomic-level experimental structures, database information and the newest machine learning techniques, will give promising results in the near future. + + + +## Future perspectives + +As we have already mentioned, there are diverse strategies for modelling PPIs. The development of refined algorithms to perform this task is far beyond our current knowledge in structural bioinformatics. Therefore, if we had time and resources to expand our project, we would focus on improving the user experience and the accessibility to existing data. + +First, we would like to implement different ways of modelling. It would be useful for the user to choose whether to use a template-based or a template-free approach. We could also use existing Biopython packages to perform certain operations, but further research and testing would be needed to decide the better candidates. + +Next, the result of the modelling operations could return automatic reports about the process. Not only the final model would be reported, with the tridimensional structure and their characteristics, but also information about the reasons why our software has discarded certain possible models. Therefore, direct visualization of these other options and energy plots for user reference should be displayed. A good option for the format of this report would be a Jupyter notebook, so the user can interactively check all the available resources. + +Another functionality that could be feasible to achieve is the automatic download of structures from public databases, such as PDB. Both sequences and structures can be obtained using Biopython modules and some keyword and ID search should be easy to implement, as well. These would be the input data for the additional scripts that build the pairs of interacting molecules to run the core Promod builder. + +Finally, it would be ideal to develop integration options of biological information to help build better complexes, in line with trending research in the field. We are not aware of Python packages that would allow to directly implement this kind of work, but maybe external tools are capable of doing it. However, a broad review of the latest literature and a challenging programming development would be needed to know the most promising approximations. + + + +## Conclusion + +Although computational approaches for building macromolecular complexes are far to be perfect at the moment, a lot of effort is being made to develop strategies to overcome the pitfalls in this field. We have provided a software that, despite not using any novel strategy, achieves notable results when using already defined structures. Further steps using machine learning approaches and a broader range of training data would improve the performance and confidence of this type of programs. Finally, the integration of different types of biological data will also be a key step in the progress of in silico modelling of protein-protein interactions. The achievement of better methodologies will definitely have an impact in applied fields, such as drug discovery, biomedical research or food industry. + + + +## Bibliography + +Chang, J., Zhou, Y., Qamar, M. T. U., *et al.* Prediction of protein–protein interactions by evidence combining methods. *International Journal of Molecular Sciences* **17**, 1946 (2016). doi: 10.3390/ijms17111946 + +Ding, Z., Kihara, D. Computational identification of protein-protein interactions in model plant proteomes. *Scientific Reports* **9**, 8740 (2019). doi: 10.1038/s41598-019-45072-8 + +Hayes, S., Malacrida, B. , Kiely, M., Kiely, P. A. Studying protein–protein interactions: progress, pitfalls and solutions. *Biochemical Society Transactions* **44**, 994-1004. doi: 10.1042/BST20160092 + +Liu, S., Liu, C., Deng, L. Machine learning approaches for protein–protein interaction hot spot prediction: progress and comparative assessment. *Molecules* **23**, 2535 (2018). doi: 10.3390/molecules23102535 + +Nealon, J. O., Philomina, L. S., McGuffin, L. J. Predictive and experimental approaches for elucidating protein–protein interactions and quaternary structures. *International Journal of Molecular Sciences* **18**, 2623 (2017). doi: 10.3390/ijms18122623 + +Sarkar, S., Gulati, K., Kairamkonda, M., *et al.* Elucidating protein-protein interactions through computational approaches and designing small molecule inhibitors against them for various diseases. *Current Topics in Medicinal Chemistry* **18**, 1-18 (2018). doi: 10.2174/1568026618666181025114903 + +Keskin, O. , Tuncbag, N. , Gursoy, A. Predicting protein-protein interactions from the molecular to the proteome level. *Chemical Reviews* **116**, 4884-4909 (2016). doi: 10.1021/acs.chemrev.5b00683 diff --git a/tutorial.md b/tutorial.md new file mode 100644 index 0000000..d51a452 --- /dev/null +++ b/tutorial.md @@ -0,0 +1,111 @@ +## Tutorial + +### Installing dependencies + +In order to make Promod work, there are some dependencies that must be installed before executing the software: Python3 and Biopython. + +If you are using Windows, you can download Python3 from [its website](www.python.org). If you have got Linux, it is likely that you already have got it installed. + +Independently of the operating system you run, you can install Biopython using `pip`. + +```{sh, eval=F} +pip3 install biopython +``` + +### Installation + +The recommended way to install Promod is by using `pip`, exactly with the same command as we indicated with Biopython above. +```{sh, eval=F} +pip3 install promod +``` + +You can install Promod easily by downloading it from the Github repository and running the next command in the downloaded folder. That is the recommended installation procedure. + +```{sh, eval=F} +pip3 install . +``` + +Alternatively, you can also run the `setup` installation script, as follows: +```{sh, eval=F} +python3 setup.py install +``` + +### Hands on: how to use Promod + +Promod has a command line interface which explains briefly how to use it before execution, when using the `-h` flag. + +``` +promod -h +``` + +There are 3 mandatory commands, while the rest are optional. The mandatory ones are related to the input files and output folder, required for the program to work. They are `-i`, the input folder; `-o`, the output folder; and `-f`, the fasta file of the sequences. All of them will be covered later in this manual. + +It also has a graphical interface which accepts the same parameters, but graphically with message boxes and graphical aid. It is an independent executable, so the CLI can be used for automation by itself. The command to start the GUI is simple: + +```sh +promod-tk +``` + +#### Input + +The input must be a folder with two or more files: each file containing two proteins, a protein and a DNA or RNA strand or 2 single DNA strands (that do not necessarily form a double strand), which represent an interaction between those chains. Two structures are considered to be interacting if the minimum distance between them is in a range of a few Angstroms (1-10 Å) . However, in shorter distances, the forces between atoms are strong enough to produce changes between the chains and promote a different conformation, and, thus, a realistic model must keep the residues at an adequate distance to consider it correct. + +#### Output + +The output is mainly one PDB file with all the possible chains joined in the selected folder. However, if the optimized option was selected, several PDB files will be created in the same directory. These are different approaches made by MODELLER to optimize the energies of the model and the files it used. The model will be saved as `final_model.pdb`. + +If the model is required with minimum energy, and thus the `-optimize` flag is used, then several files will be saved. MODELLER will save some mid-step PDB files as `final-model.DXXXXX.pdb` and the fully optimized complex will be saved as `optimized.pdb`. + +#### Parameters + +##### Input folder + +This is a mandatory argument and should be indicated with the `-i` or `--input-folder` tags. The input folder should contain, at least, 2 PDB files. This folder may contain other files or subfolders. However, the program will ignore other files and will not check subfolders to find more PDB files. If you wanted to include some other PDB, you should copy it in the folder before running the program. + +##### Output folder + +This is a mandatory argument and should be indicated with the `-o` or `--output-folder` tags. The output folder is the directory where the output files are written. At the moment, the computed model file is saved as `final_model.pdb` and, thus, any other file with the same filename will be overwritten. + +##### Fasta file + +This is a mandatory argument and should be indicated with the `-f` or `--fasta` tags. The fasta file contains the sequences of the chains and the identifiers for the stoichiometry. It is a critical file, as the sequences must be homologous to the chain in the PDB file. We can hold up to 5% of differences (default threshold value which can be modified using the `-t` parameter). However, if one of the chains (if stoichiometry is not provided) or one of the chains in the indicated stoichiometry differs largely, the program will exit. The reason for this behaviour is to avoid guessing possible chains that could randomly match, as that would produce unexpected results. + +**Note:** The sequences in the fasta file should be of a similar length than their counterparts in the PDB file. For instance, the fasta files and the PDB files downloaded directly from [Protein Data Bank](https://www.rcsb.org/) may differ in the initial and ending residues, as they are difficult to model, so they are usually removed from PDB entries. + +##### Distance + +This parameter can be indicated with the `-d` or `--distance` tags. You can indicate the minimum distance to consider that two proteins do not clash. The model will discard interactions with too many atoms at lower distance than the value indicated in Angstroms. Being too restrictive may discard some meaningful interactions, but being too flexible can force chains to overlap and produce interactions that are not energetically favorable. + +##### Threshold + +This parameter can be indicated with the `-t` or `--threshold` tags. The threshold is the minimum score an alignment must reach in order to insert the chain in the model and determine its homologous protein. A very high value (0.9 or more) is usually recommended to ensure that the chains are correctly identified. However, for sequences with less identity, it might be needed to decrease this value. + +##### Stoichiometry file + +This parameter can be indicated with the `-s` or `--stoichiometry` tags. An additional file with the stoichiometry can be provided to make sure that the final protein will contain a specific number of chains. This file must be properly formatted, one value per line, as follows: + +``` +chain_id_in_fasta,number_of_columns +``` + +For example, this is the stoichiometry file for the example number 1: + +```{pseudocode} +3e0d_A,2 +3e0d_B,2 +3e0d_C,2 +``` + +##### Starting PDB + +This parameter can be indicated with the `-start` or `--start` tags. Promod's result is highly dependent on the first PDB used to build the model. In fact, results can vary a lot when different starting PDB files are used. Therefore, we provide an optional argument to select the starting PDB file to account for this variation. This allows to use the same PDB file with the same parameters, to obtain the same model. By default, the PDB files are sorted alphabetically and the first one in the list is selected. + +### Uninstalling + +You can uninstall this software by using the following command, if it was installed using `pip3`: + +``` +pip3 uninstall promod +``` + +However, if it was installed via `setup.py`, all files must be deleted manually. From af4a89cb409ec005957a252f4f7abfec7be765ba Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:40:29 +0200 Subject: [PATCH 4/8] Rewrite README.md from scratch --- README.md | 345 +++--------------------------------------------------- 1 file changed, 15 insertions(+), 330 deletions(-) diff --git a/README.md b/README.md index ca5c5cc..a385846 100644 --- a/README.md +++ b/README.md @@ -1,353 +1,38 @@ -# Manual +# Promod - A protein complex builder -Welcome to Promod's manual. Here you will find all the information you need to run this software. This program is a project for the Structural Bioinformatics subject and the Introduction to Python project of the Msc in Bioinformatics of Universitat Pompeu Fabra. +![version](https://img.shields.io/badge/version-1.0-blue) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) ![](https://img.shields.io/badge/platforms-linux%2C%20windows-brightgreen.svg) -## What is Promod? -Promod is a python tool whose goal is to form macrocomplexes of molecules starting from the interacting pairs of chains that will form the complex. It is compatible with protein chains, single and double DNA strands and RNA strands. It has several parameters available to play with according to the users needs. -In essence, the builder approach is similar to a genomic assembler: seeks for similar chains in the different PDB files given as an input and overlaps them. After that, joins the chains in a single model. Pair by pair the builder puts all the possible chains inside a single model +## Introduction -## Background and scientific explanation +Promod is a Python tool whose goal is to model macrocomplexes of molecules starting from the interacting pairs of chains that will form the complex. It is compatible with protein chains, single and double DNA strands and RNA strands. It has several parameters available to adjust according to the users needs. -### Introduction +In essence, the builder approach is similar to a genomic assembler: it seeks for similar chains in the different PDB files given as input and overlaps them. The builder places pair by pair all the possible chains into a single model. -Proteins are the executive molecules in all organisms. They perform a wide variety of functions, from small compounds transport to signals transduction or immune responses. However, proteins do not usually work individually in cells. Most proteins interact with other proteins (protein-protein interactions, PPIs) or molecules (DNA, hormones, drugs, among others), which are essential for a proper development of their activities. That is the reason why collecting PPIs can lead to a better understanding of protein functions, biological pathways and mechanisms of disease. The identification of such PPIs has been a challenging task for years. Experimental methods provide irrefutable evidence to test interactions between proteins, but they are expensive and time-consuming. Nowadays, computational approaches are being developed to reduce as much as possible the necessity of experimental data. Some advances have been made in this field, but the range of successful organisms is short and general frameworks are lacking at the moment. -Different experimental approaches for the identification of individual PPIs are available. The most accurate methods that allow the determination of the exact atom coordinates in the complexes are X-ray crystallography, Nuclear Magnetic Resonance and Electron Microscopy. These techniques are difficult to perform, they need very specialised equipments and a strong data analysis procedure. There are also high-throughput methods based on biochemical properties, such as Tandem Affinity Purification, which can detect multiple PPIs at once, but a high rate of false positive results can be expected and the information is not very precise. Protein microarrays are also being used as a screening method that can be easily automated and parallelised. Other traditional techniques, namely co-immunoprecipitation, yeast two-hybrid or pull-down, are still being used as well. All these approaches provide complementary views of PPIs and have their advantages and problems. The main one is the cost-effectiveness of the experiments, making in silico approximations more feasible and adequate for understanding PPIs at the atomic level. -Regarding the structural properties of the proteins, there are three main categories of methods for computational modelling of PPIs: homology or template-based modelling, *ab initio* or template-free modelling, and hybrid or integrative modelling. Homology modelling is based on the fact that the evolutionary information in both the sequences and the structures is important for PPI prediction. The latter are preferred, as most existing predictors use surface patch data, and only the residues in the interface have to be analysed, instead of the whole sequence of aminoacids. A pitfall of this method is that not all the 3D structures are available, although public databases such as PDB are unstoppably increasing in size. Template-free modelling needs more computational resources, as it explores all the possible orientation between the interacting molecules. The combination of prior knowledge about the individual structures of the components may help reduce the searching space, but still this approach is more challenging. This kind of techniques also require a careful evaluation of the results by various means and refinement of the candidate models using biological information. Finally, integrative modelling combines experimental data and bioinformatics developments to narrow the possible complexes and save computational resources and time. +## Installation -In this work, our aim is to develop a software that builds protein macrocomplexes using as input pairs of protein-protein interactions. Interaction with other molecules, such as nucleic acids, is also considered. To do this, our program is based on homology modelling. Therefore, the homology of the different chains is analysed, tridimensional structures are superimposed and energy levels of the final models are considered to propose the best possible solution. Another option would have been template-free modelling, but the required computational resources and the steps of evaluation and refinement excluded this possibility, because of the limited means and knowledge that our team has got. -Here we propose Promod as a first step approach to protein-protein and protein-nucleic acid complex modelling. It is distributed as both a Python package and a standalone application to be run in UNIX-based systems. Several parameters can be tuned to adjust to the user's needs. The final result is a single model with the best option according to the algorithm that is explained below. +## Documentation +- [Background and scientific explanation](theory.md) -### The Promod method +- [Tutorial with examples on how to use the program](tutorial.md) -The algorithm used by our program can be divided in three main steps. The first one is the analysis of the homology of the subunits. Then, the superimposition of the homologous structures is performed. Finally, energy levels in the models are taken into account to discard unlikely complexes. +- [Analysis of examples shown in the tutorial](analysis.md) -#### Homology of the subunits +A PDF file with all the contents is also [available](manual.pdf). -In order to build the model, the program begins with several files, each of them containing information about two interacting chains, whether peptidic, DNA or RNA. We can overlap similar proteins from two different PDB files to check if two chains in those files are alike. If they effectively are, they can be joined in the same model. This is known as protein superimposition, and the whole process can be compared to a DNA sequence assembly. The main objective is to recognise that two sequences are the same and put them together in the correct spot. -Two chains in different PDB files must be homologous to be considered as part of the same protein and overlap them. To check if two proteins are homologous or not, a pairwise alignment is performed. This alignment consists of calculating the similarity between two sequences, taking into account both mutations and gaps, and then returning a score between 0 and 1. A score close to 1 means that both sequences are identical, therefore high score values are evidence of a large proportion of sequence identity. Only the homologous proteins will be overlapped and used to build the model later. -#### Superimposition of the 3D structure +## About -Two homologous sequences will probably have similar structures. However, it is also possible that two proteins with lower identity score may have similar structures in the tridimensional space, as a result of convergent evolution. Therefore, when two proteins are significantly different in sequence but they have similar structures, the may have the same role in our model and it is desirable to consider them. For example, two homologous proteins from distant species whose alignment do not pass our homology threshold, but they still preserve the structure. +This program is a project for the Structural Bioinformatics and Introduction to Python subjects of the MSc in Bioinformatics for Health Sciences at Universitat Pompeu Fabra (Barcelona). -In these cases, to test if two proteins really have similar structures, we need a measurement, such as the root median square deviation (RMSD). First, to calculate this value, we have to superimpose these two proteins. That means placing the atoms of both proteins in the same coordinates and orientation, to check how well they overlap in the space. Similar structures will have atoms in almost the same positions, while different structures will be more distant. This similarity between the superimposed proteins can be used to calculate the RMSD, which is the average distance between the superimposed atoms in the chains. A value close to zero indicates a perfect fit of the structures. RMSD will increase when the differences between the protein structures increase. +#### Fabián Robledo Yagüe and Claudio Díaz García -Due to all this, RMSD must also be a filter to account for those structures that are different, even when we had considered them homologous in previous steps. +![](https://www.upf.edu/documents/7283915/7376028/marca.png/906677bd-c708-255e-92a3-99524273d872?t=1484925837278) -#### Analysis of the energy levels - -Lastly, it is important to consider the final energy levels of the complex. A good model should have minimum energy, as functional, naturally occurring complexes are the ones with the lowest energy among the set of possible foldings. Situations such as two atoms very close to each other, aminoacids with hydrophobic residues located in the external part of the macromolecule and several other cases can increase the final energy of the complex, which should be then corrected. - -After that, it is possible that the position of the atoms inside the model is not the most adequate. Sometimes, despite having evidence of interaction between two chains, collisions or clashes appear when they are joined in the structure. Taking this into account avoids impossible models, such as those with chains crossing with each other, which will be thermodynamically unstable and unlikely to happen. - - -## Features - -The most striking features of Promod are: - -1. Building of macromolecular complexes from basic input data (pairs of interactions between molecules). - -2. Optimization of the final model using MODELLER. - -3. Graphical user interface (GUI), with the same functionalities as the command line interface (CLI). - - -## Implementation - -The Promod package is implemented in Python3. It strongly depends on the Biopython library, which is an installation requirement for the correct execution of this software. The GUI has been developed under the Tkinter framework, providing a user-friendly interface and, thus, avoiding the use of the command line. The implementation of the GUI and the CLI are independent, so each one can be executed on its own. This software works with well-established bioinformatics formats, such as FASTA and PDB files, so no atypical formatting of input data shall be conducted. Additionally, two scripts are provided to divide a given PDB file in separate pairs of chains and join them if there is an interaction between the molecules. Some examples are also included for the testing of the program. Promod is freely available from the following Github repository. URL: https://github.com/Fabian-RY/SBI-Python-project. - -The formatted documentation of the package includes dependencies and installation instructions, examples of use and a full tutorial with sample code. Each function has got its own documentation that instructs on the particularities when importing them to be used independently. The users could learn all knowledge of the library by looking up the detailed documentation. The main functionalities are further explained in the tutorial section. - -## Discussion - -Promod is a basic tool that relies on information from both the sequence and the tridimensional structure of pairs of interacting molecules to build a final complex model. As we can see in the analysed examples, it is successful with small complexes and it can even handle nucleic acids interaction with proteins. The MODELLER final step adds an additional refinement of the proposed model, returning to the user a good quality structure. All this methodology is built in an easy-to-use package, well documented and with a GUI to save unpleasant command line work for the standard user. - -However, the main objective of modelling software is to provide new knowledge without the need of huge amount of experimental data, such as the determination of all paired interactions, in this case. A limitation of our program is the necessity of input protein-protein or protein-nucleic acid interaction. This type of software is being broadly developed by means of various approaches, which are proving to be very successful. Both docking and homology modelling paradigms are being applied to achieve this goal. - -One of the strategies being exploited is evidence combining methods. The core of this strategy is integrating evidence from multiple sources, including them in comprehensive databases for later integration. They are called gold standard databases and they contain information for both training and testing of the new methods. The annotation of paired protein interactions goes beyond structural features. For instance, evolutionary relationships, functional features, network topologies, sequence-based signatures, structure-based signatures, and text mining information is recorded in these databases. Finally, machine learning algorithms are fed with subsets of data and performance is measured to propose the better candidates, depending on target species, data sources, demand of accuracy and coverage. These evidence combining methods are performed repeatedly to find converging results with different input data and classifiers. - -As we can see, template-based methods are leading the in silico modelling techniques. It is reasonable to believe that there are a limited number of possible interactions and that, once we have big enough databases and curated interaction catalogs, most of the PPIs should be easy to model with high confidence. However, *ab initio* modelling has also yielded promising results, mainly from competitions such as CASP, CAPRI or CAMEO, where world-class groups bring their latest developments to test their performance with real problems. Of course, these solutions are computationally expensive and require a long time to return a final model. In addition, complexes with weak interactions where the conformational state changes upon binding are a big challenge for docking software. - -A large number of information is nowadays available from high throughput experimental techniques and a lot of structural bioinformatics software has been developed to integrate these data. The best performance of in silico techniques has been achieved at the tertiary structure level of proteins. Nevertheless, quaternary structures are the ones responsible for the majority of biological functions and their knowledge is essential to disentangle protein interaction networks in both physiological and pathological scenarios. It is clear that hybrid approaches, joining atomic-level experimental structures, database information and the newest machine learning techniques, will give promising results in the near future. - -## Future perspectives - -As we have already mentioned, there are diverse strategies for modelling PPIs. The development of refined algorithms to perform this task is far beyond our current knowledge in structural bioinformatics. Therefore, if we had time and resources to expand our project, we would focus on improving the user experience and the accessibility to existing data. - -First, we would like to implement different ways of modelling. It would be useful for the user to choose whether to use a template-based or a template-free approach. We could also use existing Biopython packages to perform certain operations, but further research and testing would be needed to decide the better candidates. - -Next, the result of the modelling operations could return automatic reports about the process. Not only the final model would be reported, with the tridimensional structure and their characteristics, but also information about the reasons why our software has discarded certain possible models. Therefore, direct visualization of these other options and energy plots for user reference should be displayed. A good option for the format of this report would be a Jupyter notebook, so the user can interactively check all the available resources. - -Another functionality that could be feasible to achieve is the automatic download of structures from public databases, such as PDB. Both sequences and structures can be obtained using Biopython modules and some keyword and ID search should be easy to implement, as well. These would be the input data for the additional scripts that build the pairs of interacting molecules to run the core Promod builder. - -Finally, it would be ideal to develop integration options of biological information to help build better complexes, in line with trending research in the field. We are not aware of Python packages that would allow to directly implement this kind of work, but maybe external tools are capable of doing it. However, a broad review of the latest literature and a challenging programming development would be needed to know the most promising approximations. - -## Conclusion - -Although computational approaches for building macromolecular complexes are far to be perfect at the moment, a lot of effort is being made to develop strategies to overcome the pitfalls in this field. We have provided a software that, despite not using any novel strategy, achieves notable results when using already defined structures. Further steps using machine learning approaches and a broader range of training data would improve the performance and confidence of this type of programs. Finally, the integration of different types of biological data will also be a key step in the progress of in silico modelling of protein-protein interactions. The achievement of better methodologies will definitely have an impact in applied fields, such as drug discovery, biomedical research or food industry. - -### Input - -The input is a folder with two or more files: each file contains two proteins, a protein and DNA/RNA strand or 2 single DNA strands (which may or may not form a double strand), which represent an interaction between those chains. Two structures are considered to be interacting if the minimum distance between them is in a range of a few Argmstrongs (1-10) . However, in lesser distances, forces between atoms are strong enough to produce changes between the chains and force them to adopt a different conformation, an thus, a realistic model must keep the residues at an adequate distance to consider it correct. - -### Output - -The output is mainly one pdb file with all the possible chains joined in the selected folder. However, if the optimized option was selected, several pdb files will be created in the same directory. These are different approaches made by modeller to optimize the energies of the model and the files it used. The model will be saved as final_model.pdb - -If the model is required with minimum energy, and thus the -optimize flag is used, then several files will be saved. Modeller will save some mid-step pdbs as 'final-model.DXXXXX.pdb' and the fully optimized will be saved as 'optimized.pdb' - -## Tutorial - -### Installing dependencies - -In order to make Promod work, there are some dependencies that must be installed before executing the software: python3 and Biopython - -If you're using Windows, you can download python3 from [its website](www.python.org). If you got Linux, there's high chance that you already got it installed. - -Independently of the operating sistem you got, you can install biopython using pip. - -```{sh, eval=F} -pip3 install biopython -``` - -### Instalation - -The recommended way to install promod is by using pip, exaclty with the same command as we did with biopython previously -```{sh, eval=F} -pip3 install promod -``` - -You can install Promod easily by downloading it from the github repository and running the next command in the downloaded folder. This is, for now, the recommended install way - -```{sh, eval=F} -pip3 install . -``` - -Alternatively, you can also run the setup script installation commands. -```{sh, eval=F} -python3 setup.py install -``` - -### Hands on: how to use promod - -Promod has a command line interface which explains briefly how to use it before executing it. - -```{sh} -promod -h -``` -There are 3 mandatory commands while the rest are optional. The mandatory ones are related to the input files and output folder, necessary for the program to work, and they are *-i, the input folder; -o, the output folder; and -f, the fasta file of the sequences*. All of them will be covered in this manual. - -It also has a graphical interface wich accepts the same paramenters but graphically with message boxes and graphical stuff. It's an independent executable, so the CLI one can be used for automation without confusion. The comand is simple: - -```{sh eval=F} -promod-tk -``` - -#### Paramenters - -##### Input folder - -The input folder should contain, at least, 2 pdb files. This folder may contain other files or subfolders; however, the program will ignore other files and will not check subfolders to find more pdb files. If you want to include some pdb, include it in the folder before running the program. - -Beyond that last thing, no more things are necessary to know about the input folder. However, it's a mandatory argument, and should be indicated with -i or --input-folder tags - -##### Output folder - -The output folder is the folder where the output files are written. This is mandatory, and no special folder is required. Just a warning, as, for the moment, the model file is written as 'final_model.pdb' and thus any other file with that filename will be overwritten. - -The output folder is indicated with the -o or --output-folder tags - -#### Distance - -You can indicate the minimum distance to consider that two proteins do not clash. The model will discard interactions with too many atoms at lesser distance than the value indicated in Argmstrongs. Being too strictive can discard some interactions, but being too flexible can force chains to overlap and not being energetically favorable. - -You can indicate the distance with the -d or --distance tags - -#### Threshold - -The threshold is the minimum score alignments must reach in order to insert the chain in the model and determinate its homologous protein. Usually a very high value (0.9 or higher) is recommended to ensure that the chains are correctly identified. However, for sequences with less homology might be needed to reduce this value. - -##### Fasta file - -The fasta file contains the sequences of the chains and the id for the stoichiometry. And it's a critical file, as the sequences must be homologous for the chain in the pdb. We can hold up to 5% of differences (by default, using -t parameter can modify this threshold). However, if one of the chains (if no stoichiometry indicated) or one of the chains in the indicated stoichiometry differs too much, the program will exit. This is use to avoid guessing posible chains that could match as that would produce unexpected results. - -**Note:** That means the sequences in the fasta should be of a similar length than equivlent chain in the pdb. For example, the fasta file and the pdbs directly downloaded from [Protein Data Bank](https://www.rcsb.org/) may differ in the initial and ending residues, as they are quite difficult to model, and usually are removed from PDBs. - -#### Stoichiometry file - -An additional file with the stoichiometry can be given to the program to make sure that the final protein will contain no more than the indicated chains. This file must be properly formatted as follows: - -chain_id_in_fasta,number_of_columns - -one per line - -Let's see the stoichiometry file of the example 1 - -```{pseudocode} -3e0d_A,2 -3e0d_B,2 -3e0d_C,2 -``` - -#### Starting pdb - -Promod's result is highly dependant on the first pdb used to build the model. In fact, results can be very different by varing the starting pdb. Thus we provide an optional argument to select this staring pdb to select different starting pdbs. This way, the same pdb with the same parameters will provide the same model. By default, the pdb files are sortered alphabetically and the first one in this sorted list is chosen. - -The starting pdb can be selected with -start - -### Uninstalling - -You can uninstall everything by using if installed using pip3 - -```{sh, eval=F} -pip3 unistall promod -``` - -However, if installed calling setup.py, files must be deleted manually. - -## Examples - -### Example 1 (3e0d) - -3e0d is a small complex formed by 2 protein chains and 2 double DNA strands. It's a subunit of the eubacterial DNA polimerase but it's perfect for an initial testing of our project, so we can test how the program works, the time it takes and if the result is similar to the original one. In this cases, we have 6 different interactions, as each DNA strand is counted as a single one interacting with another one and binded to the protein chain by one of them. In fact, this complex can be thought as a dimer: two monomers formed by a protein and a double strand DNA, which interact by some residues in their protein chains. - -So, the first test to our program consists on joining the different pdb files into one single structure, without further information about the stoichiometry. Therefore, the program will try to build the protein with only the information provided by the pdb files and the fasta file. The command to build it is the next one: - -![3e0d structure obtained from: a) rcsb.com b) promod without any extra info c) promod with starting point and selected stoichiometry d) without stoichiometry control][3e0d_composed] - - -```{sh, eval=F} -promod -i examples/example_1/chains/pairs -o examples/example_1/results \ - -f examples/example_1/3e0d.fa -``` - -We just need to give the folder with the input pdbs (-i), the desired output folder (-o) and the fasta file with the sequences of the proteins. The program will take the different pdbs on the folder (By alphabetical order, to make it reproducible) And the result it's incomplete: - -There we can see that there is a missing double strand DNA. So, let's see how we can complete the model. We can see that there is a missing protein, let's force the program to include it. We have two different ways of indicating this: The easiest one is selecting the starting complex of our model (which contains those missing parts). This results that different starting points can result in different models, so, it's important to select a good one. By default, the pdbs filenames are sortered alphabetically and the first one is chosen as the starting point. In fact, this is an important choice: the differences between not chosing one (Figure 1.b) chosing an adequate one (Figure 1.c) or chosing another one (Figure 1.d) can result in models with extra chains or with less chains than expected - -The other not-so-difficult way is to indicate the stoichiometry of the complex. In this case we will focus in selecting the starting pdb, so the command would be. - -```{sh, eval=F} -promod -i examples/example_1/chains/pairs -o examples/example_1/results \ - -f examples/example_1/3e0d.fa -start examples/example_1/chains/pairs/3e0d_ZG.pdb -``` - -Yay! This is fare more similar to the original structure, very close to the original model. However, this is a very simple protein, and bigger complexes may are harder to build. However, from this example, we learnt that: - -* The program can build a model without any indication. However, it might be incomplete o have extra chains. -* The starting pdb is a important choice that can influence the final model. Thus it is important to remember which starting point gave which result. The same starting pdb will gave the same output. - - -### Example 2 (6gmh) - -6gmh is a very big complex formed by 24 different chains, with a double strand of DNA. There are several chains that interact between them, but there are not repeated sequences: each monomer is unique. - -![6gmh structure obtained from a) rcsb.com b) builded without assistance c) selecting starting point and stoichiometry][6gmh_composed] - -Thus, lets try to build it: - -```{sh, eval=F} -promod -i examples/example_2/chains/pairs -o examples/example_2/results \ - -f examples/example_2/6gmh.fa -``` - -In this case, there are several chains that hasn't been added to the model, and some are repeated. So, let's select a different starting point and limit the chains with the stoichiometry, so at maximun there is 1 copy of each chain. - -```{sh, eval=F} -promod -i examples/example_2/chains/pairs -o examples/example_2/results \ - -f examples/example_2/6gmh.fa -start examples/example_2/chains/pairs/6ghm_VA.pdb \ - -s examples/example_2/6gmh.stoic -``` - -The 6gmh.stoic file contains the stoichiometry of the protein. In essence, is a csv file, without header, in which each line follows the same structure: *,*. Each chain_name must be in the fasta as a sequence - -This starting pdb was not in the model, and with this we force it to be included in the model, and start growing from there. And this way, the result contains far more chains in the model and it's far more similar to the original one. - -### Example 4 (5nss) - -![5nss structure obtained from rcsb.org vs composed one][5nss_composed] - -Here we want to show how to use two parameters we haven't shown yet: distance and threshold. Distance is a value that considers if two atoms collide, which is not allowed, as two atoms very close would have very high energy (if the separation between them is less than distance, it's considered invalid. However, to avoid some errors, up to 10 atoms colliding are allowed to consider a valid interaction). Threshold, however, makes reference to the homology percentage between the pdb sequences between themselves and between the sequences in the fasta. This allows for certain flexibility in sequences with some mutations that may alter the conformation of the chain. - -```{sh, eval=F} -promod -i examples/example_3/chains/pairs -o examples/example_3/results \ - -f examples/example_3/5nss.fa -d 2 -t 0.9 - -start examples/example_3/chains/pairs/5nss_NG.pdb -``` - -In this case, the result is quite good, but let's focus on the speed of the program. As the number of interactions grew, the time it takes for the program also increases. However, there are a few things to comment about it. - -1.- Adding a new chain that passes all the checks is the most computationally expensive moment. The number of checks grows as the number of chains of the model increases. - -2.- It takes less time to discard an interaction that has no homologous already in the model than adding a new chain. During the first steps of the program, until the model has grown enough, this is the most common case. The number of alignments done is relatively low as there are a few chains in the model, but it can be an issue - -3.- Once the model has a considerable number of chains, and particularly if there are several copies of a monomer, it's also noticeable that discarding an homologous protein because doesn't fit with the current parameters is a very time-consuming processes. The program tries to introduce the chain using all the possible alignments whose score is higher than the threshold. This is an important issue for proteins with a lot of copies of several monomers. - -4.- The stoichiometry must be carefully selected to reach the desired structure. It's possible that the program cannot construct a model with the desire stoichiometry, but it's also possible that the stoichiometry wasn't what the user meant - -### Example 5 (6om3) - -In this last example, let's talk about how the software behaves about the number of input pdbs and stoichiometry related stuff. - -![6om3 builded from: a) rcsb.com b) promod with stoichiometry][6om3_composed] - -```{sh, eval=F} -promod -i examples/example_4/chains/pairs -o examples/example_4/results \ - -f examples/example_4/6om3.fa -d 0.3 -t 0.95 \ - -s examples/example_4/stoic.csv -``` - -In this last example, let's talk about how the software behaves about the number of input pdbs and stoichiometry related stuff. - -There are som chains left in the protein builded with promod, which can be added used custom values of distance or an apropiate starting point, but let's ignore that and focus on topics still not covered in the manual. - -If you indicate a stoichiometry file but it doesn't exist, it exits. We could have changed it to a non-stoichiometry builder but we think it's better to do things explicitly as you may or may not notice this error. - -An important thing about the stoichiometry is that is critical to make sure that all the model is builded as you want. For example, without any assistance, the work heres is quite good, but with a selected stoichiometry, the result is missing some chains. Here we want to say that is important to check the stoichiometry file to make sure the chains are correctly selected and the number of them is adequate to our purpose. - -## Limitations - -1. The sequences in the pdb and the sequences in the fasta must be similar. The program admits a certain degree of tolerance (which can be selected by the user using the -t parameter) but using sequences in the fasta which are fairly different from the pdb will not allow the assembly of the protein. This is particularly difficult for pdb in which protein tails are not well modelled and they're not present in the pdb but they're in the fasta. The pairwise alignment with the tail will drop the score of the alignment under the threshold, forcing to decrease it or even mistake to which sequence in the fasta corresponds to jus by chance. Thus, our recommendation is to prepare the fasta file with the sequence as similar as possible to avoid mistakes. In case you needed, in this package there is a script included, pdbsplit.py, which can give you the sequences of the chains inside the pdb file, avoiding this kind of errors up to ceratin degree. So, if you're looking for the effect of certain mutations in the desired protein, might be worthy prepare the fasta file according to the pdb sequences. - -2. The running time of the program is proportional to the number of pdbs selected and is affected by the order of the structures and the starting point. The most critical steps: Reading all the pdbs sequences (However, the higher number of interactions, more chains might be added). This is a tradeoff between number of interactions and speed. - -3. Selecting different starting points can produce different models. It's difficult to say which is the most adequate starting point, as may vary according to the goal of the assembly. - -4. The fasta file require to be specially prepared for this application: There should not be repeated sequences (with at least an alignment score equal or higher than threshold selected). This also applies for sequences with unknown aminoacids, as they are marked as X in the fasta file and therefore will mismatch with the sequence in the pdb. Those sequences are usually missing in the final model - -## Bibliography - -Chang, J., Zhou, Y., Qamar, M. T. U., *et al.* Prediction of protein–protein interactions by evidence combining methods. *International Journal of Molecular Sciences* **17**, 1946 (2016). doi: 10.3390/ijms17111946 - -Ding, Z., Kihara, D. Computational identification of protein-protein interactions in model plant proteomes. *Scientific Reports* **9**, 8740 (2019). doi: 10.1038/s41598-019-45072-8 - -Hayes, S., Malacrida, B. , Kiely, M., Kiely, P. A. Studying protein–protein interactions: progress, pitfalls and solutions. *Biochemical Society Transactions* **44**, 994-1004. doi: 10.1042/BST20160092 - -Liu, S., Liu, C., Deng, L. Machine learning approaches for protein–protein interaction hot spot prediction: progress and comparative assessment. *Molecules* **23**, 2535 (2018). doi: 10.3390/molecules23102535 - -Nealon, J. O., Philomina, L. S., McGuffin, L. J. Predictive and experimental approaches for elucidating protein–protein interactions and quaternary structures. *International Journal of Molecular Sciences* **18**, 2623 (2017). doi: 10.3390/ijms18122623 - -Sarkar, S., Gulati, K., Kairamkonda, M., *et al.* Elucidating protein-protein interactions through computational approaches and designing small molecule inhibitors against them for various diseases. *Current Topics in Medicinal Chemistry* **18**, 1-18 (2018). doi: 10.2174/1568026618666181025114903 - -Keskin, O. , Tuncbag, N. , Gursoy, A. Predicting protein-protein interactions from the molecular to the proteome level. *Chemical Reviews* **116**, 4884-4909 (2016). doi: 10.1021/acs.chemrev.5b00683 - -[3e0d_pure]: examples/example_1/imgs/3e0d_original.png -[3e0d_fail]: examples/example_1/imgs/3e0d_fail.png -[3e0d_builded]: examples/example_1/imgs/3e0d_builded.png -[3e0d_start]: examples/example_1/imgs/3e0d_start.png -[3e0d_composed]: examples/example_1/imgs/3e0d_composed.png - -[6gmh_composed]: examples/example_2/imgs/6gmh-composed.png -[6om3_composed]: examples/example_4/imgs/6om3-composed.png - -[5nss_pure]: examples/example_3/imgs/5nss-original.png -[5nss_builded]: examples/example_3/imgs/5nss-builded.png -[5nss_composed]: examples/example_3/imgs/5nss-composed.png \ No newline at end of file From 502fb78ff2254f18f7f1e72cf9fa07cc60a51e10 Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:44:45 +0200 Subject: [PATCH 5/8] Add website links to MSc and UPF --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index a385846..5c1b3f3 100644 --- a/README.md +++ b/README.md @@ -30,7 +30,7 @@ A PDF file with all the contents is also [available](manual.pdf). ## About -This program is a project for the Structural Bioinformatics and Introduction to Python subjects of the MSc in Bioinformatics for Health Sciences at Universitat Pompeu Fabra (Barcelona). +This program is a project for the Structural Bioinformatics and Introduction to Python subjects of the [MSc in Bioinformatics for Health Sciences](https://www.upf.edu/web/bioinformatics) at [Universitat Pompeu Fabra (Barcelona)](www.upf.edu). #### Fabián Robledo Yagüe and Claudio Díaz García From bd134efc09f60c1e9d550164ff00758bfd8c531b Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 20:51:02 +0200 Subject: [PATCH 6/8] Remove Rmd language annotations --- manual.md | 12 +++++++----- manual.pdf | Bin 870747 -> 870011 bytes tutorial.md | 10 ++++++---- 3 files changed, 13 insertions(+), 9 deletions(-) diff --git a/manual.md b/manual.md index 0dda021..ffc4c02 100644 --- a/manual.md +++ b/manual.md @@ -70,7 +70,7 @@ The most striking features of Promod are: ## Implementation -The Promod package is implemented in Python3. It strongly depends on the Biopython library, which is an installation requirement for the correct execution of this software. The GUI has been developed under the Tkinter framework, providing a user-friendly interface and, thus, avoiding the use of the command line. The implementation of the GUI and the CLI are independent, so each one can be executed on its own. This software works with well-established bioinformatics formats, such as FASTA and PDB files, so no atypical formatting of input data shall be conducted. Additionally, two scripts are provided to divide a given PDB file in separate pairs of chains and join them if there is an interaction between the molecules. Some examples are also included for the testing of the program. Promod is freely available from the following Github repository. URL: https://github.com/Fabian-RY/SBI-Python-project. +The Promod package is implemented in Python3. It strongly depends on the Biopython library, which is an installation requirement for the correct execution of this software. The GUI has been developed under the Tkinter framework, providing a user-friendly interface and, thus, avoiding the use of the command line. The implementation of the GUI and the CLI are independent, so each one can be executed on its own. This software works with well-established bioinformatics formats, such as FASTA and PDB files, so no atypical formatting of input data shall be conducted. Additionally, two scripts are provided to divide a given PDB file in separate pairs of chains and join them if there is an interaction between the molecules. Some examples are also included for the testing of the program. Promod is freely available from the following Github repository. URL: [https://github.com/Fabian-RY/SBI-Python-project](https://github.com/Fabian-RY/SBI-Python-project). The formatted documentation of the package includes dependencies and installation instructions, examples of use and a full tutorial with sample code. Each function has got its own documentation that instructs on the particularities when importing them to be used independently. The users could learn all knowledge of the library by looking up the detailed documentation. The main functionalities are further explained in the tutorial section. @@ -140,25 +140,27 @@ If you are using Windows, you can download Python3 from [its website](www.python Independently of the operating system you run, you can install Biopython using `pip`. -```{sh, eval=F} +``` pip3 install biopython ``` ### Installation The recommended way to install Promod is by using `pip`, exactly with the same command as we indicated with Biopython above. -```{sh, eval=F} + +``` pip3 install promod ``` You can install Promod easily by downloading it from the Github repository and running the next command in the downloaded folder. That is the recommended installation procedure. -```{sh, eval=F} +``` pip3 install . ``` Alternatively, you can also run the `setup` installation script, as follows: -```{sh, eval=F} + +``` python3 setup.py install ``` diff --git a/manual.pdf b/manual.pdf index adc6697aeef68f27b10e4534e7b68a5b8e5af42f..67c1ac7b37550f78307a9d305f5d8f5a746acc37 100644 GIT binary patch delta 31776 zcmV)NK)1i!k~RB`HL$ol0yQ<4!9NWte_30X+qMyY&#%xUb&iDv3GkM;coVycyQv#* zPfpYHflFc~bSYBh#cTch&R_-5;_p%hZ3oBL|>ui&n=~4%BEL>1+O!O0^wQRUX)$ zsQx1GHuo?f9gE6P%iR{fmHCV(n7Ta?p&ze`P9sa_ATX zpT2z@aGPfwCJ6y`I24Y7Dhetmez+wMHI9=ofEv&fn{YeF^|PPj4kIq{IgUsB^EOk# z8(u={Weq){pY|^vYCfzp{a6n?@_aB2wZ_@Gb!Up|?B5^G->-9#Nmh&s1+E+SdN>WP zGnT^$`y*pP1KuHH#|~7ifAcfUsi&qt4w^@Q>4qg3Oa7Y%AeI0&E8Pc=_h=FUACEyt%{*=Pn06h=ccW}v{foHex;LQJJwFdNnnp&?pwAm zAfgqy#)*`@1EW3`P07>(AHwU)wn45{W*T+ZbDKE*#1?LSK1~)pf13n{@4MpYSgf<) zodx$&g9d2QfZ0`md-!Gp)6okpx2Y`Js_D?eXx^F^h+>iD%m$~?XLclTPaVZ>H~c=> zwr=;fL{Eo()Nkaj>pa_<$wl3_ynwc0GJT$$YM-ftnnEqKUi`XuVdK=Hcin@UUERXZ z>}_pJKIquGlPXi!e@^i8`M}D-*k(sx#|Urnb0&mJ?dHCUj&JA-I*P+3 zA8eTnHdI^kRm%B5w70tI-w>rA>Gs5h83!bU%uEvpik@%Oe~o)?*}iNqH|Z|Qs8KPF zrb6C2-9Qq^4Q!iFBEB#i9*GCc34ir!1i#{3oS>+@jNx(v^F)>;6vE%yhuWfroDN&? zWapJY4(2{{Xt9P!hg;#}ZV%uz@aj?rzX;b*!56cR^LdyReG?DWe!Q zf_gNBs+`YMnWzs)JX70XWOH5$w}eQGGJ7BodZi+W<87A-z+19Jpeorlcj(~IIjs4HD57RJo%e)6sGv&d$;y;?JQ0K{+T?Jx4iRer*e z1!nl1e^X81W0-9Uro1Z8@De*?35tpCLbAhjBI3)E(0+6BI&I?!41ZXYTM7_E#a<@) ziqv9}M%k6nvQ*23(DJVdEtD#|>2(WLSo9>dLNl8UZ2*>Fs@6j~w+@j5-z?9Dww0 zxOKLAFh}bQ)LZkX76lJVF`S8TzfIGib5ok$-*?59bp}ci`u(+jtu5!d-i1Et1Kf9{MEWhpuEhXaUtCb;4`;>G;HNCLD2WkV}r&J zfg$dmdWdfk*Dw|)2p~B*1Z`WFwY-)!5f3h(fK@$VNJ=k2-e~_mQhfhH*g6uSQd4rn z5Q~5X$74)C3j~xG*a5j@-y2l2-sg#{e_kTVAZUB(%)I%k??!b>r`eHut+=CD#S1Tx zebaSwa&7uZh0?8kiGp>c1&7&RO;NWE9+igx5a65U!QTWEJ|1k{9s!VZ|F1mZlb>)` zz8}F8Ce76Z+9{i&I3;;RBhEv^eDO)V^=x)Vg>zOiJ?q!t+=G(K-*tkv0eKxpQI za+U?ZF6X=-yS}Ab8?cXwKkyvym7QLs@e06B;|XAcqM~B3i-dv4u0>@?lS_dT+D@UM z%U%mdTbe)BV`oo|G7-o5%t2Gp6Bt}`p`l*WBhVQ&3GMa;6#LD&f5^|)Jg>RQEEJKN z@LU)sYcdC~ki}Or*IA}_Ahi5-859UPyeZ0FT8f*LJ9Q3i_HIi z3rXq-CTYnHT)jtbb`Zx^lKQN(hww=?A(df)1?$ehj|Nj->m$ck{{e`=C&qB2+=nZw zjJF-tA2rwEfpjdMe|$KH-bXCxQ%Q=ZJ>y{8Jvl^v> zIx-fQY3dO-baiAVOtm(gXX)LhT8x38;&Jl@93AVrR%_ZADFDP$`;_3()hd0Zn?i)NR3cn%z`;Tg9|^V1(nuV;*`t7}Q>fH$QWpJ-e4){%znWd&1lr z5(MWO#xD#LEYmG#;I(}s6F$G-G-NGccwBxZ0WS|$Q>mV6g2l2fJ>5d_sWbT#|G_H2 zh;sz}WEG@df3V(FtQlwenxii<48Gji`f@kNaaTn=_k%f3vdgBcc4PBqdch0fOjldc z6&AVwZ^T(<63;AC#R?or?mf}Tp4QAa7D+y3kna7{pM(B;wJ(HfNIRE2g3Uo)VpnhG z3&&IfKhb4ZeUA!7ma2vCq1w%=o4Hhv=6@5PKmw zyhd|QA1IA#vF8;7AvqN>sP|4w$2Ht6UM%?O{=5GH$-?E$lX^!e0x&U`!9NWte_2_R z+sF-m_pi_+r%K|o4_@9UJN9~=a#I3H)p#|H&kQS^P)vt(&&C_TfZu;0o|?n z#vAOrroqAgy8knrl}{`;NuGrfsKB1K!t*KRrfWEn2tewTr8Xt?AG*3Ue|>nj(^?1j zr(&e$YM@y>U4g{7t7-j2R$MVw9dC8+QFmFeH@Mf-n-C@RH|G9-`Xz~%*S6V-G|ZLS z+?OEtxjL)6T`c4IX7I%>b#dXlC{$_A(N>0N`MI z5YgHIJQwxQ@hPW;+U$-ie_2JW!wfZU_ZrMNov+9x^H8VKje4%9Vc%Ehk%cN)(FeSq zx~3Z%UjK;Zb|O<@9;pijK8>fYWgS`(!}HTxqgG^@tuH}|FiUxjU%|RiF3?@qzJ|#{ z1gGv!$>0%N!7MuJod!b>>^U%^yrL0s%{-Y6HO|0LwI)M#K9>b-e^@7{l7erFt=)Fk zRm=7a1`c!JwhzUgZS?U}?XmGhBh0sRUBH>C^P#WImq^7=)OjKgW-I6Gr{mz_>;?~3 zFmvd7b_k#8e*!+W*u**trBo#2{mGCL(x7k;ivI9=3WsBfyRczPy>Kfvk7qPV z0ZoSSVy*6sYAtGnqBz3sXtmYiFswDh#mwG-cM&Y5e3gr>s`iX%G+MO8LoV(>W}!?n zIv*%l*LH288H(Ydo5u4r(lbFNo=PBs|1nj)DP31j540^je}KM^ebI1hNB@s^i240j z3%4!df+$e+F07&{_Y6X?N9)3H#V`i$ zV2K1eid9h<$yrMpD1c%me?*1GnTu|5&d7>Q^alod2FNUG>b^u&7(s}dT13Tp`@K1j z0}Ur-Kq!QCvOe}cA0Bt>>!V}_wP@*<=o8wkIF zKVCUxmV{cyvnhaL)`vlKs7kz;Bqu$<0McN_Q6Juc%YZ~@g7_9W3jjIn%Hs*@i-+25 ziFOf5$ea~q!Jj^Tqz7V?uq($4dVy%dvJpJs-qYAMo(g0OP6_gigQ*WJ^gYFor3dI# z1FxV{f3IR9K^_>TwcdWbU~V7&_HoVWhN?N&c-q%joG7{$Hg@CJ?~{g)7E5k|fr#=j zi?fZEI!qGn0}Byc2S)JNSQ{VsZK2hiSV7J(LCA z1!w`qOd9fq+vlKV@(3Ui9N6s+Iqs0z9iRa;e>56k8QYb3+`1(8w)1%q(n5x5#6J7( zuWtw%DV}7(@2+7Odk#u-r{pdW&Wa38@7Ci7YW)($(Ma50*yP3jx7|qHt=Xdk_LS-f zRBqo;SK+2i541&2W`>Kyssc|f*USrML3#rb-jMdLojHoaH1=SQYos&$*ehWlk>pd5 ze~HbCH->wF=vu-Ck9+JrA=|TX3=V|Zp1&s;0ScM+BV;85y%T-Fsa=als!9f7$s}Ul zdX_Io`A|4(4%m2bqVT8&Y^A}0Q$r191*99|gi}MF&U#Qlgwvewo~_)Ousf%9W%SmD zqr%|%FC`H43Kw!*MJvyr59X2D-?T?ie?r#`F2Kc+*olZbzo;n23#Go0{(kLdVK`20 z*Sim=uJ-|pkBY($+Z75~hO|d#8|Fax?66KlYM~*Z$(eSRGP3Z=O#bjh^{&$<^q`8-Soz#(-}SxOkKv8Tp(%3wM)dCrU7O>+5Lp$6h`pLmK|`7(QQC1p+le; zss}fNxDlBJzW=Ag^i*IbYb9yv->(7z9yDUOn?snO;)aDty)Xj_r;j;7Tsk65So&g4 z804!>@PMgQQ~l4B^J~YU$i_xWuyZfY*1ht&2*9o)hKgE*O&4%o#cUOtfA?UQqubR5 zKQ9&2M68~ngP?6-R+m|{(Xqta9E{_8VhNK#8$`r+&bXajx*72{XfH6|1HUZjb{yU} zXwP=Q1GYa=>tB9*_x{tT_rFt*!+obfZgtphTGJmZx1fdDxjQ2_OxX=F!dQ@4E-))m zX5=T5Xkxeq0ZPARo`yo_e=C;s+}U49E{0-GE=pf&x{@@3nZ7bn>YDN(ii?VnP zPZ-~snES#hq9sJ$VczjZQVl>NqYGho{^qaqSa(R3DXX>BWGc~ce=V3j1uu8&#o?MW zqKNY~>uAt+&A)8Xnf&WHhmuJO3ALcD{cVGGY~B-#rD2446|c^0jo`G@$GB`|>PQv4i zdKUpz#&M{k#Aio;QF!0mx^)C{TGaCi~5_T&Bv)RM1>h| zTV~C`1*f{4nR7NX=Mob0(XnfGZ3f00^(Fz?qFRgjy{eZ6f3MbAx}MbaH>+A;9odC8 zb~~Gl?tVm_F_?T7T`Vvb$KiQ_hHzZ~Xe7}p`^_Ss{bpj$ZK2sKHH#=8O>I7kn38Ma zY|vAiVi>ti(jw%%nMW2tg*Z>$Jwy9)m!@zr5oRvZUTWyzp8C~d-0ozY9%!-QT1`rT z;U{`QWKv`yf2kufDHEAEcAv{2dxkDwKh-lHLa;ctl*z-h6H71IiT2;omea6YZ?j!^ zW?EqXzETq(0Aac3MJyHImEC%c-ZG>rTV^1ge^iPqxdmYR*m((tppf7=B)fA)EBuBXgjJy2=x!v}qn zgqGt)62XP;t294U;1AcRzWs;-8o( zN`YQRe=K3srr9e6+)5HJ(P{@Vm+=S-Xzr#U4Wf^eY#&T+q@(u8p;ZYx36@_d{QWFHLC(Z40!9o`5V(a9Y?R4r===Lu{|6kc zob`jUQ-`us0f(|v0*A6w1BbFx1h=wN1@c(|HkUzp0Vsd1SWRynHw?Ycui&x0NFqh* zqc8#_4xAcj(ZY6tw1y9fvq;>gyT*FmB>%oo>%h*A*Vfq1VKLg7CyIQJqBz48C{kof z#KM$`tx_Sbkt&JczmOy)1}F%zQj8TKia8IQNHOG0YLS$*GUkvKT$ECdBu05{fwJLd zrd$I_p(1~6lp;F81;DIv-TClZ2=@H>Bj1C=l&%u|2?6|I>PA~pn1vXPH+fFA&AZa;xK3{EvN990|tV4I6%%=+5nSP zdEkKAiclCUL)|uL(xx_m)+sc5O|&LkOVSMf8Vy?7n#domBI|_Ln%V$Lg1%vAhx{S= zSBs+s`ImGYu{#dP@13nq=-D$o9{kal^BI5n-`~8S4SshqU(M#L3$lhA{K4#Eak)I5 zU0^S)Z~QuYclPGR;xiow+(>XPS2-CleTjKUp*8HQ<(u_h|M^!_K3jGL?#(YhemuDY z1&grqbVPQL=zdT#a|L^(gNnr=VeSW66sJ(AgJN}xLUKO}+uCLBxdr82NTU_^pdWuV z&G*0%hp%FDOPa4^lq#6M8n;?g*JDigzb_Z3hqDzOBN<-qkw2P!UeUGXtyv;*oF>AL z%M;}5ltkHpflm!COXh?oFiq@4wF}eeDk@0_hHB=_x*QUmCO#zuuZHB(gNjc7bex^egMYeV(A8;HM+8` zMvY(@ai*?s0vOKB9?Fs{Dt8u*Z;Hc4PF)w1N9L54Tjxo7Tuz~7n37W)Fp!9x;w|)y zVIexM)`OLHw%C--cvE4&1BsT#B-(>cW&eCWU%<|BYxbHSqvh6*2mfNRd^dkvt}TTV z|FhrsyZD0!TEIhhyIU){r3Ce)2C1F{LAXYV(u2p_dmjy8^+Bh$-Dt;r2`%m>%me$HPzgMSrHW7 z)PqGyfR!#R6qmGzLNDyvYe0V&CQh~#zY&ZRT#MZZ809ZUd1x!Kk=)-@4m2gno|5P} zj1xVT$5o3|C#BH_4EdoDK6g=~E^bMW5ljs*iCtJsE^+O_ViH%4yD-&l?e7SM7d01Z zmkz;p=l+^_`@*qTx-isO^uawC!?$c*b{UK2Y&}@bF4Z)pV&5e1cW8g5%B^FsOgr{C z$f#r&ds@kE1p^&U3Wlz!y-BvpfLJ-9l$Ys6`@?E=e({s{t#~g_FW$D&?Kf+8nRY~% zJgMy)!K|t?NnIG)AyezZ@IqE%*EhUVyRu~mjA5FwpSmF^+Q{}=-+@J(t|e;^Ce>V@ zU5Pjux74&Tj1BKLU3Pz286wz2+sG#$3>@-!L$%K$586?$mW#{t^*5O7qv9&nuO1-P zkCt!d7w2u-=@;_5hyLa4zq8ZX!K)YK_i44foZS%9KI%+K_D$T~oysj-;B~9j!=+Z? zwg|U@V@98*?Te;ibEGNf{WK|G|MvFpR%Y7t{eRoHs(8Q5KtV6r-JaZhTW)^3+*ic# zMBzA%3QNE_o%%Qvm!jQ|Lo{ea6ZUX)?PK&c_N4vj~T5 zvk8Z6vkHf7vkSLuvkY_+D>F1O3NK7$ZfA68G9WTFGB7d3jQ}?)&?Fe6Sf)M~#|QYi)93B~^N1Q+s2eguR^$Ju?Fn4?s>{-rmlhnTcKnXzprb z zswk=}sZuk1e-u`Ab#SnE`X4T$s%q+zbO1471vPO1P=gL2sjjN_=dT*j?t{NM9Y8_t z!~dtxN8q1yd2uyiHEktvX2w5j05AjGfKJYqf7<>RH;NBtfWN7Iq?$R|+y2!6KxN_L z;=sem=$VD9Sd!eH-Y&fsA47e6%%OJ{()y^}TIf8)~$XaoF97*{*f4>?^dfd4A+ zXHx)jmL@ECVsm%51k$I5hV zynT(_fB$>OjO<*Uz5b!ie=gg^-p<+5*~R%^6@dUVOB>)H_RfFy%+l^Jo4l}sw1l{- z8ok`d;Mvj3+kdEI$Kc}O@>leqdctCIAEU_01z=%e0Wf_Gskoh~sJ*T22Ww}zKk|uL ze(2<4@8rq&f9BfS&feY5`~TX_EbUCq{wQzif9k-fZfEJ}3X~T6pXd)0+}}2HpbLNr z0CWTZJxnYZ|FrvSTK<@s|Cm2y@bz}EcL10f**F7zEzN)*A8_8zMs7fWi<2wR*ZZH2 z|3q-ioB&fx6PJ%c|2Qmge^r;ZGqVS9|I7TrYJ^e?@2l|8S*Juy?XGviUEc zrL%;k2hdc>(#6E$FBSh~mUc1v7=2+oa~t5tp8RD}|8tsbKF0Rr>aqNDa{=g?IXM5z z_c4(s)^3jf9TL;62#e*uh&l1j>oO0@rHWc~^hw==OfwX`z_u&{Fg zjGUZ|JmHu==75Er9pKIUG18_$kG}>5z{p@{@A8oXaBy|;1(?}8!Ts4x4t4;e@E_B^ z5C?!!z!~XH}zyIm|M+5S|ujGGZi-_2J zc+<0eXhYBXAr~hr^PeH)@coZelYc$p{<=~>PX7PmKhGBc5aCr`VGZ920hv4s$zK5`_= z-PZXS%QF6sdX++=TiWvbJEVKTEOR4D7RQ)0M0n%T}q_-wVq;Gxn709@IjRjNk`OQALOthSP_PnxMu@mR9 z-;dtS-!y{DNT!Fq?A^%%&o|o!nTx*<@b_vQXkk0z*gof1m1`-yg%(<8%exUn^MKY( z?kKUwB)As_e*|h^3g+wFnj)HtimbOjFrz!iN97i4WcLUoM}LB-FK}AXQXl$hvl@}~ zO`v3Qd2?UjfVUtt{B%v?m2Dt@<8~aqbZAx{UpuBmue#+%GYFAF#gLnY0fwivS7bvZ zVTWg8KrnYYvecPejPr3k7+)d0}$=`e*b`GF+XE>fqF5*HsO*I`EdAxf$lvJ#CW<7~d~$JwP*Tr3G8$%pbOeX}1jsL>-Pkf@-3l6<56`zVqs{7A{Uj z5>w7i9c1Iu{#VZQ-UVY(iCwrB(j3{Tyx|ZoW&!e6B>}?ToW^E-p$*s7-h`yx z)*_xg-cgnj;FjnXNCK6E*TSu0IcB~~p$#iQ1JfnaHU1ouU=_&QH@qKOFi@Q^a*J-? zf8~yk16ij;uJ)`-v5-+gEXg;hsAPB(ps|{Vl zJ(Y78LE@1M)7S@V<9XgLeOul}hssPH7r##UltFcieOtD2J$YuuQPs$h(*XpiOgpk4 zTVB%RF^5U-MC^bYM2^Ay)QkunIGH?qf1C%lxPIcuI+E0|sOjDa>n)z|!TdBT{=ag< zl1u!&nh@}w(v})gCs$q%MIeVA3?XY9@aY0LE)6B2S4#&lnAhiU`fMfGgxBIxx)-lb zl1SVofC>u&)83GNt0DojQ1zmOrzEv0wn*Ivz3z?5Ezv1y8B-yC&B!4J9^xf=e;l$) z3acyP<60iPVvy{PDm&la3fR)>itoR96?;RG-Fo#RpFAXGLf{6GKDQ+~PR%u!5X16^16cn=;H* zXd9}sVh~gSconr>dY)(nj(7UCfAKPvMIB>TWei|5>l+TwWy zy0q;yy&b1B7~0JWgVu{)?#C0bpc-t*Tp{+&o=y~tKE)rn5Xe!T#=%C>f5oSo`;Iyd zwQ31kKU?;RZ|Xgg(356(Nty)h;_WTVy$RV%DekdGfkM=S9=(%nL1gp-_>`Rwi_VUv z!fb^F^7$CfE>4GuVWPSI^3mzi`cU79RRuu&OSlL>8%*pC7U+24a^dr<^w&DI+}jBV z`&Sz~CNNX`F;M5IvgUG?f40e~ZaUxDx>r3ISpYe`2I}_|CDdwmGS~3>=2~lG)UZNc39keRe-6()Mv&$3axoE< z+IqleETF84Aex!-!}>7d=A@NdV1$0)22@G4AGNJ4s)aOcW!KY?26Lru1 z36)0fD!%@o9m}BDDKHfBP&n?w-U!a=Rp7*G1Wmk4hJ-?u33KLWELx@bj5C3UnvqqG z6A)_9($skZT^M`Ue^Yy#wm$c$rK;jRH6A!%!EFVaV-05hN@CNX^y|}gm>DD9{zFLP z>8}COHk4Sh5$(p7GJ2Ek%?O4CM{H{)O6(Lg371B&dTOk#P>*?fN6_yp#Olijx9cQp z{DrXkq2@Y)xd)AU_$-^AR5J|P>GBmrKc=IX0&>ovUnR2Ae^NdPad((>@HwxdnAFU; z2}UeRe{$a`$X_5s%ZXZWl!N& z8gDu_q75tyyzNA#ZrJcfajMBzjn&EQ8$+P1JxO##c$euKjk9vX*Y;&+rSY}ca6mC) zv+~r^)8MJVf6Z*eD(!w!qFr-RSRbiHO@q;aLFHtE)vP2MlO$<@W^dOe4l3s5U=yX* z4(`o%pWDvoMDczjVW%Ha3B6LUy*gRelS-#P>gZMbDTv8_IKlqYl(j0yeFvR-E%<(C z@OL%*{+4Q6d9?eY&Du7%kQwpy?~J9=XL! zkBXvwf8=3E5PR9Yi-08^p~~)?vKpVh2)2!G zV2-d%SMgIMBA#wQO^DpIr#pD>?%{7Twj63B$&en_tj>?GZ^^`!Bv|(MzF8`FBe9lG znm-YRJmupUva5!15b}mimlkaTNu}i_wl~SBe|t~2=2P|%-*z24q+LOh89U#p_S*6S z^*EwTwWy9!Y4$uWcHqhfpm#zv(Pd|M0w~3z$O}Z|LA(e1A zLw9cTRIr?Gp3j(K&hwQ9Q%yU?;5uRZv-XE3i~8wp2;YT@epBvrkq8Ok3DD@{tu4r9 zf1dM$<_7gxvbCQXAAXe)6R=4lvgtoXiF4BpPr}$`S&XRg-dT`cm^y1hL5!4W&_O{! z7yK>q8;0}!t8oXvP?m3d#Qwsk4qcT4e$0{zcv^Iqw{24bA$dNQ~fXsTb}e{5f-U6~j4H0EU35OlMhav}5=e9MdAg2SRp zq`zie-_ZN*hVh(?y^Y<}Wor@#_GD6^6xTa{BPiu3W6e~=ZL=35p!uH3E;n*iU?)Gp z-0}3KWIfic_vV}KLnmr<#@ffb*0-5&NBmR`BvElZDIO?+9@hdRaM5~fb{H`^e*+zT zmKf#tOpnW8_Ra0}Wp_cVhlyp;U}~wM4B&0h1f`_OVd??9D3<~Ke3RTW>D5cUfatYk z{?+6HwLO_mHR4y+c&vKkL-o32Qc5O4%1nB(v)>q|G_^q`Bk(=Xmj&D1TVjk&t|hF=8orZs1Fe?@#v-*3_`P)VS9QW;F3>!kxIB@9anNffS${q)OX z=YY5b*>c9H2b;o0pn*qSDjkuV`=$g$7(i3T(6=r@Ng(uiO)6P2pNmpz1~9b$ERI^CQ$ecnFO&e&s(sF>ONE*>08w;ABR zue?@mZ5S?zxyR8JK|_}cIlL~i%D)$W@88XYGoKU2R3*A;9=jw;Z_eZ5imR@6F3v=8B{fMf0f=1n2Zx+GSmNk;`p0eOLm~pEqj3ddR)_HAyevJxBap) z7V;Co_fE<>FWv6s*APbr$3rp^5BnSG&=|_WX+f`*Jb3^aD0)1p5h*I0kD&CN;j)GZ_n`;pH>$D&Mbmx2v_m$vcmV9DLw_~f3ftYw-}swMx-et zk$8zfNaoS`!Rk`9`tW$Zo{&V9#B@9CVk*7*PlOjrkwUwx)`praOer6}C;Ld6i{B#$ zH7Ch_1*GkU*M3`}UT?jAb#8Vtt~uqZ?uidxv_;v=JK*>I1=OsH#jZ9A73^ zJ79?6n1|m^NuRs!f6Epl5))*A$Z#O2_}|~Y9}k9KNdA-*&SVI*eB?tne#I~JJmjRG z97Q5oLlaR*$dN`L_!4Nu^qNW#Pi1I_5g6Mr_CzK3cyJ3jx3bW>CGrMMbOiFXKz|OS zutzHI>|VqCOIqocSPA*7@2r9NiB*!5bA+@0H!l>p_NT}ef6!$AGf(FeZz0&I^5cf$ z3i(H}5>^E2n)7WxQhF0<4pQg?xGSJJ98X~$Z3$33;1rbr8s3;mxht$d#QV6T%*w*&p`UttiZ zyBa!B3?PnPH3&q&x$n(7j9rxpw0*wvGaExTXhNV_n9bPk{kty_ote=f{i`eEZ!9} zzSu?de>_Jk`G+2&Qofvsr?+yB7Tw9*ea8`B-P&YMRqg6t`N_NFI6$1gb}M(7-nJ*A zYX!e^RUmFzkYy4tg0LC|X7?0|X(9OI2b8pw4G8UXnnmSUxUi61pbN&NW)h4M%@jKE z7l>ieO#012SrVt8`R7qPm8K&7TmfXnG|Y}Je~|*AKK9+ewnu1d>>NPKSn=|I_F93*E&q7%RNaXdZE(Zk?TA9Eq#{{G2{&-KB8Zbj4-~kkX4FsM=odVg=5^CvBQL@VV%+k zorNYXp(>dH6{Z2!_JwU$Sf+SI-{QK8vn4WAhXKVG{Y(NzhhkqYA`!*zYuQJziyn0NM(Cz|PHapn z3j%%^1dLfP4o=KQ)b8~`hwnw(LURxu$VZ)I^=;wVj&i?W-Q;6czJmr`3SYeZf66D@ zU4X}SQxANw;0gYm_IfJv6TE%Mx25mC)IC1m(~yh#+YQSZ(X^gcvU*Ug8bF4~`TUS=e~1oK;fO01>{=CgK5eZA8P9>;u6(V!WY^5UW;c!~4d24gP+@07&vlscg zx?7&O>7JuNyFnZyFSpRq;?gmP`wVl8DUp=$oDk0xsw}`!#m^s_{!#8&zg}o;*o0T|yzdnIt2SqIOrS-VlPB zx{uO{R)H0a+$t>)H3+sQ8S-~4+<1}*jp^D!A&4`0-nm%(*s43GoPo6{8%_ATi#G6g zVg_0Rzk~8BnP?Iie=47B$Wh1LiZ-?Awec}WjdqxW(xhlrlEn_+u|i$hPjNY$Mn!mL zSU`P6l`ptcc|8s(|7>kivr6$TXtm?3__*?CAEa+-wU z__r9~K)22Nf5;qY;vocoULIDf^Y)pUnc6qY3zWO}9Rw?vKQx}{AjS56@8a6YXTj1S z)-OvFhU<@iwxjHRPH*-nMQae(t(R$Ug+nn32KGhEWB_KwWvx`08JPux{*|M>U zxAMDzeOpCv;izeI4J zRjd|4f5hcy%+JM>M1V_*M~SZml%BQW4GiX9&GIg88w3gUfjEoX#s;+KzHgdl4;DiB zr{@5!38Sx-;C8jJ5X#I14B(&kH@UUPG{ulQS~&?~Ad2j~H5*+YBAdr^_$Lo59m@TM zt~S4hD7ZC{2ittT3asDyyzpS?g!7%x8=TxXe`Y4++V3LV8GtRg)q97ro5`(5kVd*N zcS*6tf>9604%6-*Q`5M-+gWn1x{opP5Y)ngCNp)&f)`y49GV5aqm z52GZzn0+Illu5WwJfv2q?;37pYj*ntLubFx!Sq*Bbf@Bfp4^cOyb;>{KI|&g^a`&C z+d7BHtDAvQ29rAUvI1I`WZe_?eM;J$$x4W9}|XJ^Bg>Ji@Q1Ly5r67dBd z?NH+}kddFru_b<5fC&e<8E#F&>ryzD!pfn&#dx~nsl)p^YVPr+lWPpt&Wd@`2kCi{eGEaxe`cL6 zCH7FEz(|45fD|Hdhm2k2l0d^Pa@Kq-#~Mon{us>Fv$5Sb+{tIPI;_Y)L?Iy*<@_n+#}LiY@i>%w547CA%76?3Q> z$JB7`+7gDg-CxJ+b@89SQmgb`e@#<1mI#IRYfklV-v)HkRC>Pv8aQhzQpssmQ1I+o z?*zfDDxNj+rGYL&swmcT^@oYBt=)dYM#dNf2t$$<^<0s-FRayvU%q`i3+i2Iny~aZ zL#i2b18o;F27|1FACo|5vxhUoz4Q=%ZPzn$jNN6ZyTl@$nisCBs#1QlfBso!2IBKM zo2JW>GwI8MjBQ_)p9(b(TRF&4P`G6of)v-bK_y9g>ap486iOiV%zfsd;0Gm|^R;d^T=h+_;J6Ux&I?}A+Ef3zBHfCDU~P=Y%I z%RDcBf`Ep>LJfCt*J9DoYIoc`2_{+|9FxVO5aXa{7+<#16ihMLA-ZYP@1Ap4P?3Zy=O3BFc*5F!}B0pSn$Hf%%bU>(`bB6aEeJhGeVhLT$Ma{FbYEu!d&HQiHVV_-(^6s(uAtp(KeT;!yap1a6?@6TN%-9gI-iKKPQfe_@#wO1g#w#--)&nT0`M z+;)LXR8!_0$Nv&Cz6oYn|G$qoC{sEOUa;YM@2tMgwCc|=lMP$1kOX=|pzkgNX_M!3 z0Mz|AUzHU8nDQB?snDbOhTeg~`%h#)Ui?5?_(p5Nt}a*af64VWsjHwHBps<&7OmV@ zyp|gRf0$tue+KXlXkLx%pIX}ng6?z!7!9}a#hgrg+1U0tCVA6fVJwQUxi&s6Dg-$o z3YTQ-!@_t)8V_6;_NgAAP9SXKC0$1|mcI5#dCZvsFBA5=+Ml1k!ihrl7Og<^8qD|Y z-<%1yY8=61QQo2dlJxY^_S3F?0&$7sO2Ue3f2;7Hbh^0y%5A!kkJyvP=-P%c z0Tq*r81|r9^UNs3dpsnt@U8SEfgR~ZWLeCK^atJ{L+_VuJM>00ZIx7de^2k($5*6_ zT_QsYc^|qRKes>@@p;ArvogJt_`B{dV*FV|{4_U(Yn9N%GpJv=yB%2B0=$7IC#nA5 zJZ@WIf1Ui=pr(SsjGt#gdPI`9NXy;{x3z;XO$Ho=ol|h2P1LPpdt%$R?M!S(}WG)^$Y#^tN(&kO<0_N+mY{e8Yw`oYx$%2eP_Rvd|^(?DJiNKXcRZ*Lf` z-M}o#;rqD-vw=j@dB*}A=j9Ktuv1t5y+kx1$vXWZSHTM8eEicKUPOscSM&$o@PTJ5 zRo@N@DMBO7&D#eX13kvDq-T_?sK6Rw_*7KoI@d#=i6l$-uK)pw>M%BZ)h4vOjE%S* z_<7Ro!SS>1hoQr=%IUPPmM+I5R-Vtd)9oopbLT=LGNhQdvQ~uKv{>YmK=-p%Qm9h^ zt*z_n9kMrg^Y<3xF{QW@$|2)tw8E4g=_$IeMgk{Dwfkxxq0ZEWsqpUAlc0~NbhQL? zwc#XgewiEhlfVUxCvO#Ka%xy)5;GF(4YlcJQ&pn<}nng(&0BwIxp`Mg0WshaDE*SWL|d z$9FK0JK4EqLo&V4jb!Wkf^hD8VL^2Y05O{uO z^;(C0{oMPoQ{d_*y3D4xkxEs)4okwVC>%p;-^IZgl)tKZ0PP&yd2Kqq$}$wO<&$J^ zPBaSR)`mEDb|Wbf!ca{T(?cZ(<|^(Ojsg_J@hF{mvbFKAv zBRP=QVS5++Eg^Q`f3=FPT$&0un+hHS&4EKUIzK5bn-J6VQc#gG$KPq}_N|*h0DM4~ zGr}TY6Mdjc#wF;xbQZ8P28Xs?}9|(A<2vd8boV<()`2Sw+b1HqYR6XXk%Q=NC`Lwzm^;<+rcpm9r)3T97%h{ z9V<9-ty8$QEN0v%>cp6Y5g%UA{i(6?!P_;ck01=UhsLi5A$A^`iilH+Dr_i1A)X%o zVo8~BKLNNZ34=~xqDKBB@nxG9ez5wQR+th_DAu68>s1d~0&<&-lV?0JQ^aFg zw|zYDi*3RCmT=P`8M=1uHB}1@B};IuHg4vUPnD*{1oXb;FbTO74aMGd-_?xvDYS97 z3lr7cS0~o?6nJ1 zOf0fmgjuF}AE4C#L*W_wiTL7#F${O3d@)tsnS}Vgrw_upWJ6H~VN!QY5s|XpE;y3{ z5~op!;+sF+E7(xul>;M1x5_eQ*s1aa+w)hwG2`MmH|>v#V%B=2fhSOkkF3pmNbv~i z)=(4x>W9@Z@!MZ~@#Un67yR1g^taMCt-rb8Z7U&7s%d-|K|!R;1a=Wk-o{tf61)8q zD?7HYkV7BN7TgE28L!wR*;%!xuPohrFDC_Es(8ihE8QW}Oqt8DA>9)X4?z~QDVWg) z2*Om>lTl8~gGOZy#aFK$nbv?iNc=R05}pb0Bxq|b{+hrxbeL5yLXMEt;EK7;31<-3 z`X_S1AP1p+)K@idk$c9@AxRzKz_ykW0lir%B(vLesa*KgHDgK=xBb{HVq9ZeWaIG( z1|0cE?nuXb7--@@yz&dpo^g;`MnU;KLw6^P*~pi%VnmT1(hGl~9WBCI$my#7jSlO2Vh%dey{3hoELmmkh^ zyJZa=IkGDV&)x$kZlQl17NFgj2XO{;B6aqXHQue%sE=dw#T5cwe{S~EKMpT0MXJfA zP-4VQ?#5p5Tpm8CyF2gaC5|=Havh)mKB>O29FgIS69RpkE1tfp3n}MM;`xU6?Ja`n zP;T$4(9ki(7gaHWJp@&T34W<`o^ols_*t^m(;OZ=Eb{{?DSUr#Qfx;vDON-9?-cfC zi`S}uPn&dpd)XwHw7K{(u4fHFo+x4|vy5B)qPIe>Jvl44K8xpBS*l%$Vja%{YJQec zJq&Te-rO|4s$0-Fj}(=YWHTOTv2?WbWq(D~__;;rbLvQi%c1If8xTY2Xm3w-m;3g( z_P35pCnU$6HPH<*B)o2p7b){{ml$r6J7CFs_m3%5oYEqyd*c(w&f!~#dfA4v=}S|7 zm@Fk2rU%t0^!?nO zYx##rNdgyqrhKfWh6O58jPd4)hF2<~;t3^_J}_O&qf_L*u)GeV`}`|&qUvj#rv5V^ zjw63}bHH)&U#Uk9nE{W9H}`;Hym7%{3GRdSpk_(A&jz{uop49n<(Ft8-;elE`#c({ zwO^e0&%VFZ3vs+nGg&$S2B<-dRYr#c$a+G{^zaC<5YsjT>ufyqV=-MEk{7k#e4AI) zO2v?@R^gAT_P!}__i_Z~PLtG}s({Ngw{Jj+ zEQ;>inZ*?X8TSB{0O7E#s-?Tw?8hVx{mhkevL_Ku3T)t9iY<#@ zaeyIzfzQXlFZMFva%xD}qZr#w$Jd65#8Xv2!Qxk|a`49A|MYa5zhCmUUD*e(sae z@|Muy^t>i69h8B+bJHq@wzW#Lvu;kyvG$wP5z_I62iyT5S8)}ND9)Mopx&(i(jS9lc4hTg}b>;=P-qG&y~DY=ZuDGAAd0y-KMQ#Kf4KDKL=#1fU0nc5$x? za3oea(u{WiOE$X4;H=K#0i0$zcK&4M%hgzT7L0>6!}A;r=mk=F+3+U83@nw48&MR@ zvrlD5AOv0+em~FghgomRRuiIUMPuN{tW+yX@XsKM+k2GOz+E-hMB>>Ag8Aci^$C;;j@x!Sg>H7v{#Nq?O5l4$RW9GK< zfa+8rBW$C`CiN~hv8UUx3hAgQMxi-pHP8@^md*visgP3aLbR)~1M6(LfPIi^8!m&( z#Ir54JM~UZ#F3Sv-PMdl-!WuqaZrUoWW}?g7)%#MQfbt9QWXC}Ve}Mg#OU42a0g~d zLZJ*twUjMS=qgaLzMf|Gv__L3;@`>3S>R8gV8-Zo7a9#+Xqf?73%-t(r-G#iF=93F z$lVvhHRrTPTNnAf5|P=ZpcR|Z0Gx>ru98T6ZP^H=$*-S@hjc+SV+2cJeWmkH$ZH;P zYt_)NKOIPMVOl2<>R4i&9$L(Nm>=cQIe8N_LRBOk&ib&8?xKA_ko9FA>p_>8YT5#` zM9HYDpo7o5O+m#w|{y~Rt*Q!qFz;dmGur9Ca5a3m7Onqw%Nc00^n0yI$!{&9qhnd z0hab3e?upwymq5eq$(H*u?+^lWUTpSvfHusA>>0kc$8{x`j7n6H(_!<&_~qWvU_j@ z)J-H+>keiM<}oFusRf%&%g^`NN!*TKbpMkUyS#i%EvuI1J|m)EGMYriH)$02LVl_i z$671efsklFpF%4d_T!kb!HNQhUeQqp>(L!h@2g0kjZ~*pOhGXrpq-~Tv$IeFT#8U=}Xi- z`HId4A6d$ABW3F!7<>E3MX9DLbg3&ba|{{+;0lB4b>S6N88BREY1hDm#gMYZs=ZEm zVnrS5%45|d+&fNS(ji)+ysve1;+B(u{M=YR(Asuon#wdrS=_g(Pn!`Djejj{Y;Gn= zQ=dDGpM6Uo5(lV6=zIvG5km@ub668dO9y&Jjqm|{% z_rQMVr9yy9@clW^3~>{8CTCP%3kk8o~Eie2Dj`l_zo#SXiiVKQ?r7TPYx z^X*{#@_kVt2Ea@Q`)v1KXmL@X!E|$J;+-R!rKib-0cETHq^}%HZ6>*0{b9R(EK(R)b!>4OYSjA_4-p)-9m(pFNUjc349?v|Bco zU=8c<`pFl{KUY$~z}7Qh~^$UY>4@6Ao4@HO@5(0))R zne}JeYyS3i!Y;zi&|=aKUguP^al9C#setV?&8eslT2eWL;;tYC<}sX_dE4RmqFm6% zJMOOT5&I+I=4n_qijn{5+vSMxVFlrQZX5Y1>!jcK#=C@O6e&hJ zkQ>bS4}jYV)d$Yr`T*~*${QEE;Vn9JRI^ibFwN=H)!-uVNBcWhNN_Qz;;B?7hRjkh zrqj93h?jKf7a2@ zFg6(%ExK4Pol6QAN!xHVuiX!rIx-m|w}-AF@&ThqbN(?b24!jM5Z|zdixyvK4{r~? zMGpNKaFggMkt2Rt4)UFW&A&V*#vGLUbahDq}Hhp7t);4+qo+F{YGI!pwDXiunCC_e@wca^D*?kWGOUyLU<8o zLIZ9lj=woQL#E;Rm-Ff!apBmg)+1KSY>g12DDo%5!lre4)d}B~*w3wMX29`pe;QE^gDX`rs!{8sh7HssVI*LLe3~;KVZ^2Aew9Ec8~jl#{>epYYns zWy0>Ibp3$>Ai2UB>&dGuLOt<^v!L>8u}nRifSq6w53e9(q-pG6!5H4p7HWN^1#s+N z8wAZEmca0tphK=23-!p>F|L4I(_5$e*46P`EHj`j(r82&NEd*$KWHf{9GxxKW4b?S z5=L%cYgx(3-XZ%Sf$tTQ5kelUjld?(YGx_@4~Wolo0^-=GA%G+%NwQKb`j;7|sjb)P`%tpHj`vK2h#MdP+5NxC{rKby~$Z+RJ z*iFrE<_WrsBI%xhrgj7PKA@A!t~IA2M!mZ&uOobVPL|8_6!-X1h|HD78xTM(Kf;DH za?7`9dmRf~!%FDyc^D;+w`%DyW*uB70x9`~k%SldTDrHjE~?rC&e!~RG@)fP6Ju|W=_n}F z_*Ne4k&|m2JC|6wcFfAg9`J73&FQcpEHGkm?BT&baceiTaazUz7Sdi@ zk)%xQOEl;`*L`sh2$0zMG&rWjc-WU4U%aBu)JCl3I}x zAzmS#0V(0qymEOt9hHK})ZqS4BG;J+{$T$RI^1`tU*7U#&zoMFmB;kNAWf0yXG$zh zWk-Ad<65t_7jlQX?y+e6<F=W%$$bwcy3TAW@2!TpbSoY8NH+r}c{I5FfKb<1gy#N%_p}QdG(>tCYif=O z@2-PJ8|zgA>rpbiY}&1xRy}^Y&AfxjfHOS}9&6lMiU?PZKLEWz1SNmQ^1xA0l4=OA z_c8Al{;+9=#4f~rg*7>7Z$3w=w+th02+Sh}>(p<_S#-YRW|X#*UbYPAKtn@!ci8T{ zyIgqbWXg|=J@HSit|YoSi=V*aV!6UcX46iZ%pr~d*qOP4$yjISN~cabgzcUAleyf| z0X%|=T1b;`D$p$y)Uq>J&#mYE*oy_eS~O>qKGOx;c4nd2-)^_2fz;pkVb}73J<2*n zIZNk{C0YZ!vb??;&u5ipx-tF9De7Ul140trVYnXJAxD+~6!G#IZQ;Vxopk*x7V0 z&WL?iglq12exch&dwTsk?RZy(WW0?<%%6i9q`ety1|c0*@q3VvL?smVYucn55N2 zlsQ)5FMx=Aoix#6TDG-`y>o8Hn>uZ&Zu7V$rP@BPq6CC>js2ce_g$6ONLOs~!BV1a z_4725`(B;`k1)V2K*NXSja5|vne}%pq>-k=UCH5`->5a4Nj}9H>jpdaR0ii^(^ujvnm&gRuaZFhbiKODrrjR<)ZzluQqPVCZVgD z4LIcZqARyLM{5die&uisI04K`Ghk@c!sbGmt3Znm;d^Z$GC+VzYx-@#!q_4 ze078~6H7YQ#!q6L6k>r@60a6~X(B6_7P$W0|5yNh-bpoXr|8Bnx0BI0^^~LXh$q)Q z*^R>^^uc-jkuFq~nZSxHt-3M_02CzN{$qL=bM{8~Z~Wh`Osqc3whj?#;%H*Z)ZEa) z+p%?}Opg#J!@EeI&_O^>Y~WEDO?aSDDKmC*ZD#f#@BzcrZrx~qu^%K$BH#!Yz;S`# zwx2ajdw!)+qe5~k8Y^Z`UO-+j9`e9;3o^|!k{?1!+~f07cwj#v__UUa_xpSr?|| z(s?0*#u;F6gyboMwxu?%;^5`PCW-&&jNg~!7WF~R0;{-`Cq=w@q;j!HJ!!0j`r5sB z!lj=~p~_)3U@MBZ{J*Ct-}dOWKp#PiC?i(C1jw!PF;d#Pagg9_B+Y#WpdWd|gWRSz zfJAEqBQCz!$N7s{*)PGAD#;Ja)v6RPpDQra?=YwTyHFO5?F=2eHX z-C(P77?e=F^AifMgt~ zbLccpFJ5edB3P<8E>~U$;7HVYqlrqomsT+NkK4d+y;C4Ce*JXf&PMDke+AhKi_gEQ zuk!lT%M15m9kSxHDNiK4m6udKFSa}PhSP!>_KUu$P$hQ;%zvgGXoI3I!BAqrohZ%i zqmc&AJ-H9^o)Bha(GM!*igr22#goy)a(2+!^F?XpxFnn(K5RY!gRM2r_pFldc^GUJ z;GVOuRH@1Ga+|7<9eEGBWmBn|8rTtM^3=fr7- z+Q}>3TjAA2Tyjj6t`csjAB2-H}3cq@g_f z(!)QL6MII3^`A9>0e+pET3{+f=u7maoenCWBVo6YqNRO*^dCW@ucuA;vEQq2=bW{a^JLyOE2y|$V-C>*_En)J@O4}l_fmsQ z65h$fqthV)&$Wm;9QM5_KX*S1tF57Ql@+5X_6#;bZxr^|s!*FEcLMN~S?)o!Bv=1$T?Bi?$pufMM3aqy3NlBMD3l zFJg}1d)iqHWU82(NqpwmO01K5l3SX9qtimVfi#szDHL{>NRsadNKc|YlAL5x+r|M8 zaziVa{O*d(JGt@rEyb_;B_~FEq6i9zP!Aoxz`t!(kxqWow2ATFY1}M)&oJ@AP$d7> zsf4Q7H*&=o!Mt|y4T2&6yd0IyBP11^=*J|G*IY7+iCuJP_IE>5JYgm#faM(n+PQx{ zdGYl@;2xuV(0|voF*i;;VL6eGfoQmP!tV471?79&LL{;{2P~F$tF74Mu?)fH=SzJ~L;T2h zy+Y{5g>^g~;g*sOUtTSEi9@nurhawyZ2k3GJfpr0Pky6`x?RUXM4tgKpxy5tUmosu zkn&@dGRLotM{J@Z`M4S*$9^kZ44@;v&l_O2GQ2@OE<<`5K>YhqQuygpmDMAFB`Jy$ zC>z@*2~$Em88(g0zrMTCqI!yI0n|x}skv?Ut>G}0C~&7auwcyh#b+9Z#c$y^XgQ(3 zG}t4zm^6Wj*7!mX+-TDPto*Zi{vFdEn}k$(za_5C;5x7gukVb)JD5!_V2Uq^?>*ic zbO0~%A0yG+*~ihn7@9wzYD12=>_e+^9WKQgg5$WU!q}uAm~~LnY-L0(u4U{c-tmt{ z>8&+Tr%2B4sUE;e8y+TmPgG(jmM+z@Meg>b%yp315;l!}Y(BMsBmEJc`@r2Z=m+1# zaeqR+p(ht`m@3%+O6ux)Av2Zn zZsxudDK2hr$~!5*RpYjKKwo;+NR+8@t)YV#z%Jkh-K%pF#%P17_NpFl^CR73A9EhQ zqaN#SL3T0AQJKZqhe&dnRf&dKXOPIH9NQqfLAt=qmuq$ZWJfuLCX}f5YmAUE8q(hk zC2Kd=1cH$5uO$;kJ~VSfPUWJN>IxFF!Luy4F=yMK-P$vNKiorB#cj8F&zB|B@1o4{ z-~%%(01P%I-R*@X*NgK|tKT(7!NC<*Rpm&`#KFROdv-NVYR?G}z zP>A@1^0xyNg94+~F}x@5q&#MP%qY7vrIe1@-X*;ajqG9;Ng2}QnTf_;= z>zQ)mxI+Jt`p*-t>iGO#9UwAMpEtdRh}+XFXvep|x|6pRAcRBuML7js9re-s;)nh{ zuDorou4+*Y;TP^WYl;Xr32sC#Iz#;X7eZ4r5*#a_nA3biNwBhIET}R9#wM_yENLJ6 z23EwS7y6%PRMA$%?_+`4E^Z|>0uJkPS-6?$d05*_CYU0U=`sRZ3Q>p6f6 zBJf)caiey_K-A2K0o<#+iLQO8oDzf84sngF=!c?O>#Z0hW%Z#8)W@{Q44gc}n3<<3w)0&`zT2XRx_E^s?w=rGTTAE1?eiRl>C%J!XV< z{~bc)XFly7o4?4LE!wcWVxAiLNBnW^Hz7)e`f41!7u(}VlXS!Tz>ajjl&U)v(&&bh zEEa#}?{D!dlBp66D6RAg#MFvW#-AYxB!)JClL4rJDZ_)?Q(_S~UMt(&HokJK@<2z; znO6$6WZ81Me!le5`;HIx7_#gF7(_qHf%|=@%Rjv(r3&Fo`x|s4?gF{);>gpr={2Ng zs5Iue>rs-X=zG|w&tZ#_Hk1}KWn*YC+_7iE%50Nu*3pmMtFi7?WC-81^TClz$_r8; zKAKp?hvS|v&IB8p?^zKXCkZ`PBuEz}mu4-HD6V*2(`ceLD4!t+u1vD!Bk%n(Q!B(N zEx$LUd&x~*_vcEdhJOy@-$gNwRzq3Xu6lZ23<*W&>}Y5BWhQFzNQLK_4FJirm+wyp z$8kDIdcUTS#)_tj7z`%L~9Bh%Loc^P(H(&|;d;8MxE)nu#pV0rGFK5jAO^z55^H!a%$BTDg% zI`Xr@5nmgLoR|fJibet(`!1{ulx>y9r)rSa7U6prvlJAyVcf~&0xYaRpT`keW7(}^ z$=Q!Z>_#lsqZ#eISP%v*DCSh?dLOhn~6A{7BNp2A@rtiJC`6rf&(3G_fjb(q?~Lw-lOqu zYflS_fyyq752=m^okXerz9%0?SqJPO&1M94Ni33a_Hmn=fK@?tl)HL>Uz6qYZ+l$M zi@hj>-3{$23I|^6q%wt%2;~tea(Gb_I`$yTw>Qee21t#+!OF59o8H{w1Z7;7xlM>X z)|Bz@Ky+1Uesru5wyBhAhBAs0+A_aW9=1NeK?925^s~Ulx4@Ufe7?(8kYYchWGhP2 zA5cW&Qw1B9V6PN;uH;6*!?Q%;C!&&Z;g2?1HRB6%Ux|6Lts?&8bBBNsehr5CEa!!; zbq)!#{4X4ZI?M)=1ACR9i6Q7c2G#I4{=I70de1g}*EF%IC28thrS~`_W%h6~BX4jU zH?~G?PnO+NgzAmw=LW^_IGt@db+XOP5x$2uAbM^~;?=^%7a*0uo2U`E^$5j1-MLV~ zs~6wWtwRhbEut&U)H;{iq3R24Z<1Gt?tm)Nn9R8D%&%g`Pn)$^3<}xeD%1Gr9i*o| z5m}IjvGef3R!Muea?Y5zlxuN}SX??XwZ@(Y2~`tRUqO9gA*!@U50hwQ+wS@)?o^c< z1BtQeJRz&-7b9YTRU^3Lb`&aEO^$n4eZpD7J6)XW&DY)MK}@5{fQfqr8P4Mdxdi;T z6ov~zToj7(ma&3}X?F&3Km~GmBqQX7V+v#du`P#EBy#aJpY3m!)bXENJxLbL+QR?V z1YvQwOYf19E0o4ZlKw@V8el0+t-@DcgUy=W`RJYQ*}awl0EZ7qy`~D}2q9EVuYX!& zpES><)a0nICy49}+Tcm_V4n9TKpa{RoL$9tA(;%}`^ z@b$vpFIv>(OAJbnufPi;8fC&k4^154MA=N9R2|vKrwSp4j{ZPEp-Yv?bK3qyStM^A zPbnRbhMAu$i|D)%K z*;qN!ggrrN0Ud4UO-?kwJcED{(^mGjItI-a9~n2}zmYYov9y0Zi}W}Q zk=MNJtSmOABM*krD()gUA0{s$9}+i(8${$VQneYzKB|0JTT7Yn!;VNj( zh%5fA1Su3E>G=%^asZqcEP?nCU8yOAv3+`=kZA}K?tk`noz>h&P+rhO2>&^0L*za? zrKS7~pb>2qyq-^w;55037@4igc|1**EA3H3h`a-h4Z%TIWA zq}>HD7>b|P>1gtOeUGEx!jxi};Vl%gZVXkereo8hzmz*%7#`cNd86depw5&Zy^9%exBEEux` zE$apI>8=B2LqPb0itz}uJ^!q5wC7Fz4=oTBZ@51Q8V5S5kiil(Q-z|)&`_r!w1E;q zv8nXDqKOj7bU_m)L-`}+sHFak($y}PB2g?b7eN>K7VPgxBM7{Td+8Kf%o;hudi2lu zz^fuDK>($aLp)5wAfco`h=h1_t^3WdI3fiqc()tEgRt_nV0s@=U!}>l4kDRemjG>S zC!!#sn17F|alM}3doxz)rUKnikRZ)o*QT5VJ=K_(1P4@cNH{_@f=uyN4%z~y76QZ~ z>I%J^YQi8v0^cZ$m6TCD7Yxf{fi4lhS@fTP|Kb*8 zS~6Ge5vXLE#^NJ4=wyOOd7J{L5VB}ESLAQ-v;JEqojS&^ zd~Mh8DT7r#H!LkqsN6Iwgk|IzJndr7B_)Q;hNx05dY*_!HcqZ!BliRmKojaW=bCYq zcJtG3%`|PSHnXBx0*!{Ms%*~b>gN|7dDpF523(d1t0*)_^@#sat5F+`bN-aay5zDA zpxqa>-j%h}3@z?Ysd`EJaUFX?Ha6*8Q@I06Q2wb|^%Yx=3(=FA92*U~{r8SSmMi3Y zgUiv9*KO9IX_6M86QmL+pjKRY)?O{@a?Kb%ADGs6b0r( z*K#SWH=bKcalf`txjK^MT5Zj&X&1k)%|+&WT-YozTB4qGXAM=GfelPDFL?~_E?v{^ z4tl2T|2U9a@e#(Y^-fv1=UIzN7mE(^vHDyE&2ore@>Zv=R;zZOlP}7HK1sOdWVp5} zr@o6?s%&{2j-FeOoc4J<@7tap|ABeB-CvI{|BYx8gy*XIdicZmQ^1Qn3&77kBKDk} z%lT-pjHH=^jZn;M0VGZ{eA!hrqC&@e%uyQra z{94lyR#_%vG^PWOwFKohNtTL+7}6{FSHzsJ3jtvFfz3vSi=m?lV4^b7oCcEz?B0RG zsuJbC2JnlZ1M#HTKe2SQ#)SuE;H|p*My+k91?H)g7s)uT$GqE-m0|r_%kiAb=Axlv zczxs)R>rV0>I(jPiaFZTHy>uWF%|?1aMz2%_%M;(ri4?XCo7&s!|6I1st9J~JCZk# zW>y&WD@B+7l3n{LB3yz=!|N>mAI-J(rT`AduY@Ca9{5k9p!N=M@RTSz)W(8|RHGfO zk}lJirHKUJcZ2)yc02{+Q0-H52WZ3a$E0}W?C-LU7*{y;cld}N#j)}=dvCTyo@*I#EE~en9>PX|0=c??kCIWfpZ9q1d*h^DB80Y zmKRQS08(wG{Q9$$7&hV?DS226uzvpF9LrU=vER^!K8fU1ic?Krk;e2$wZ*xo00Pb{ z87mRC(F(2un)wc}n{QTXN{mZlM61dTV$1yUD}z@;Yvo1erq4>#%zH|v&?=@}JJ)Ec zQ`?Xan)@q{O^Q2F3nA^KwdMI5^S^kp3u7(w00=#B4?dXnR$C(A*r|sE2E@AfYWvu4 z7Bz|l%0Nl25piyB01b$d-&*Y0#jCN6lB5lyY;hi*`x1{N;4S+ASuOLC1d(Ec`^|2g zC7GxhURNOhFkDKI`p$qcMG>r$x8Ag0!`*}Lr#2PVG&g0;E5{TC5aMcjjSlM6BTH-KX>eE*8+Rj~41K6p71^U=OGzPt+y z%A#62y{9$U59D?yCWGbz0{79N`{3{;UtEaY7cYD%D2qa|?pNkAH8HU!C1G_J8 z;NRgw$6=qaz&LO;^9BoyDZbtK^a%mCBt!}$3$Q{b7*uY4{Y;X#{~GqO7LK9~?&$>g zP5h>)d{{Hpe8$UWw{P)$x`k>PF)O&pj&brHJf9jeZ{JSY3~v7}Rad_5FC;U)Nqt8E z#|E!J<=W#&@aP^)wUSP%5Xz;jqpgyuH_o!$+r!A`y`4HH=(ld_as14 zuVVN;Y;d^iGqrOo^r{1ijkhf$ zrYq^8L{)*q7a8phB%UD=#QshZJ>gU^Be5&=@I`1vxyFdd+Gnl>`|@DT&J}wPyeo=| zf@6gM!2HTg6$IP53xP{K<$3++UCy#;!ZvHV2Xp;G5t1oM3-Bh2F)+1wdhs8LED$gv zVxcPcB|5_Xi#!=kKXT_YN#WxtVxZly3N&4DjFp=i1F43ZDGhs(5_oCH^#5J~}S1A8wYemd1Rx7@Grs|GW}L+O7%NUmR<=B$e3B zVdek5DcCq}1azAnpn>=Q+Rly>z2!K_?5aXvltZj4m8QPx1mGexX~Q&QoXjQKIk>D| zX{PXdOSJdxHUJ;2qVtbA5)fXwL4NMTF1~D z$gFFA6(&#&cxh+o*6d;S9JT18{Ef?n*OtW`o2`@)td^e^@@9`=KQq? z_YdovLW48yN&s=6l7pF;298PI-c;Pk#f+F*ocBNUHZwCjGaD=Oe#=q-0U38++yt996XZD%%WnPqLN}ftb)Y+|LchU|NN>r*_ny| zUylR9;*@5z5%$R4hZn2-(`0yZ7lz6&*+Y(>;-<3`#1Y9NctGIanM-cW*aJm}OMH5K zJUqOU)i(qB+7OAT$(MDXX0<+Z8LVHA z_bg2GIQVN7`ojn$wlBXVq1AC51wUBK0b6pH9znO#C;B6HfEqvOfEkKvmppqPG)*ff zl+A%)m_glOnC4`cU(X-zXP-nHM3MteZ6CT`h&3}Jm;XwJIs-#jY?A!|LwbOAdJy1Q z-f!Cm?jBHJ`jgotP^abbQxpFQgm5si#EY!bAvRj1Z>f3_EYW|Mx$x90cK}iZ$I-k5)hKz_war4N@aJeCheU(m(e_!&;%kMmb zA{65_Z4KBPXvqP!;+lnYb15D4HWfb#Lr4Vk68sQF$31sQo46h!m1;Wm zwy31+|kq|HTTUmb=+x>iGu`)JLlK2M-=` zYSKp!qhxB?M~{dEG78=u(}be%^#jk50Q_bscA8A|AT*;$=D)Y%IfV*;)nouQ@=D&> zic>XlGGmMeD-3E42&x|VLLv2lNQo9w-6^f@6~ce()kilyf32pS45F60Qg8ud^``lo zmxe-%gd&Y<{<{$C|2l;n`3~ItB(sIH^H42Dr{i--_{7|JZ`0Yup0QaUp3<3$p0F8I zpVDc;pRiBkQ|Q2_a?6$%06F^CY<)O0J5y&DCo?14|9i4Gwt{nZF>-S8a56K8E^!L{5`mBo3b8z+cw*O-2ZsWXHBpPxscVaJE7RZr6HHI!$F`%v z!w6JcSGL&`;aHSidu(e{j&W>$AWhmS|U7bcQOn3>j_@cz;utM z9C+bCUxtIh%Py`bH%Y3(e>95eZpZd$8svI-1#7el3Vev5ZU-8)kMuOZ3#Hl)sVWa_ zPgKtc+|6AKNXMcwG;-X+voas?1XH&sBJ@L747g}-O~0Jm;!K*}fjIn5PNpoy;dZCfIQVs&Nx$+?@aD)d%SB`}!6Z?bCL|Zoe731ffwQRW!s?CDl>g`?72;xec}kWe?Ffk51vhe!}ncrbRyPS z@Ww)Tslfs?S-^}HkRIOIz;uj4%WW!)zG^!3Fq*d(1!Np%Ig7zr^qB(*@Tq0&cEjg` zZR>V#ON?~*j`oexbuZ7hW^z&YEia&Lm`sl+XWD0aK}(?#dM`fRy>M{qFuLx+%&xfb zGkaUxk`Fqzf9|Bp)U{Lmd_IVBFt*vz(+R?R{9Fi;3Uif{5WGygqC}aJ7Ibh=GFJf` zsCPdXH&M3kBgtO*%MXJyRUjMR%D1ABlUC!b}1JQf8(}e*#6%9ChQKN47893n$&h7&RKk z(Nri~XB$WYy@7A@NhB8*!z0OnCE;1GMu;me#D$cVmpNQ)U|yId38nDY_Mx`uA*aI@ zBH4K*k%PO>0$Qvg)8STlx!VH-4P;&D;K$GnRPY-lcBcK!)BaK%i;jtRU<48}>b-l$ z!CzsLe>)Y7X7yC5a!9UMM;p|Y1Rd4kmwA>+9I7N*u72NvluL4Rg<6c=#`1Y3esZEd2wdWeYheDtwRx~u+(Yn zhX#0PEHf*;$ln_isP#A$E`IA|M#Au$$$e@`Du3;ZDwJceu|nixwa3MHgIu$Xqq_kh0zE0a%g5wWsdP@OfsMzZy zU(s3|rZT%yT9#_DP+I;urA0zzH@%MHJlnyvosLB+ciUYQWF8M>Guze~Bmj^3{#3Zs zj9y|~+yu<{A>_dwXbHn^XRCrquS0sier?_Fi0JNpTaTReR-8HxH3R_R z!*J_-^-y8OFm=MzQ&SQhd z5P>1?p574OBCcU9Ob|eDaR}D7u4{3vYa$+8JprqF!jM#6fWFcHeWv*Kg|f8_Bc-O| zh9MRK3(m(_eijHQFR%k@$-Xyee`LL{6IY`|l0nk;+L=Z3)!2>Zluomy`dW3%SjCGd zkbN<9b8&5YsUqRlzC^)V>X6gy&!(u`29L@k00{6*^We`R6W$+e-5vpu^ZQRc;e(%W zSH2w~5+=>n1llQ^vN#oaL?g+=z#zKV>Q0?N_DEl(_iX(s&P#-Q9In#>+nUI-YTuMO5f&+vIC8t9+ z%5=pj<0YQRA~|ii)R)R+e+Ynx^J~#PIV37m4TeS$2*+Fu5)a)QyUaOD>aeVCv(+Z^lzf7UcZ+x0{$_JwWw zfnK0eJ<^Dr5C$K&G4|?gJMy7WG?}9R44%61uM^?Y(c%)fV+RRS1OTjHvrJ+Rg{t;s z%bwnCv$Vq~RC&IEY#VCLIOfhpUx+q#0@}BjtX-}3(rI#lKo=mLeKnBY!Q8Kk#xhFv zO%x}YkVz6jIR`%&f03g|=R&QiRf`b3k#6(O^iS4B)hva0jDe3}6ujPVvnUvw@&Tcl zOUhXm{Jfm=cI^6=dTk&+CVt0rd{lOJk;W?kJB=rR4Tg$_!66bpJa#P_Lz-L~lrVM% z1+N^n@M}x+r+)10sxk}XIG+V*YI=fzYi=~uYjy-Sqb8-@f4+cXzqkIg1rDGXfQM{IVG$5oR0sl$9g8QQ&f)DnHuR|_Mbn<~W7|DBM2?}&xPGz5aN!%;kWabI&5YuH z6Qhod&1IT;#0^~^nF&*`4d+?9x2YE2AWrdj^9ceS+qzb3)+iMKVxfIW@aTG#zSGT_ z?EQH-w2is~2~ZTAT=j$>!xf*FQ4)%jtBoHX$$-vIe?TMe)iLCpy45=P)i&5~bpHF? za~JeXO`PBvC_Z(jf093V z1vqhzf1sbdg3JrnyN)&EOy6_#Er!9jJ6qrG<~;7Yi06JV$4PhD#A-J-ucjNkP|n2K zimkB9{eL6QGn07cnJQKgNOB*EPWH6szHyl3Qw15WYy+7Ys?4QG{@ovd>sVqttf>F+ z=0EfTM<^OiY{%3HKM~8T5lBU0ma0Vrq7Ke#f3Ugqkmyf|2tcP`@0dEyodE!JWOY@O z(Pq>fK2syv#iDBwFGCT2?R|gRwK&{({EWoCnJSTlO6LpSJ|q6wg_MNZda;O2!+Zjy z=k4v;fvm%o(e;JTet7fCUn^?rO+U|C;t(8c3B^fu4T+pY0h){0|IcInQ7gt*P^6so zF1gnX2rx3Fm?`J}f3|Z%eK#fm@PfAe1xtn3PtYey&o zG&qwW4=8_)Sy_+cxDkGzU*ShJU~@%M=PdFtlb$P>WCx25kRTHTYMZu|(c#rGJ)I!` zJyonCCAr=1^@}NzB8zo=^%WJa4y$nW;Z?~0Z|`3H^j;>bT;z#LR(JbVDDq6fSh$j_ zREk95UnoMIuI}p9KZ3vCWO2}rHya&=!M13qQ8s`5P*fE>=sft+wRL;5QF*YV4pa1H z^++@CZ&VgMQva^Shli$Wi(T0qc&UN@_pl008;*v$K9s`|-x%*{pfe}CP1~14+wr1B zv!mJF*ff4sS_6g~^Mg0owsnn@|9ST(I4hr6t&%(wA*jHfw#4&k?Ou=IL^1%WQeQj`vS4d)udy=%CH(1q z63?x)+Q?MoTCeVQAm)Gj4Gu<_vAdu~i1s|0d_T@35rXOM*<2WB0sHE)V0SL)&jpYjtjdN%1Y!o> z0e9IKTXy@0W4Xo7Bh4^BPE`SCrosEJG+#pwa#*E9-NsdzpE zR0;sq4`*w&Csb=uA-csOZbz#v7Kdq#>CfKm48$KHUCO4MZB;fW1h~P*1s;ELW&<4w z6~%NuP;gY)wPnQU4vlskPUAprLSH z-zVnk6c(lk92qrEOh z1hOlgT^Hl2mVZ9am$3po%;|r1PN(N$=0c}FUN-wnAr*=!aWmFuId>1ZBZGlahundt zwEMHE024C!Z8Xi+^lZyFL`xQvYlfwg#v%(RXs+EnPwp6lV2{Qn_L5Ny zqQO!Vj1#LN0MeO~!bs>iBQ1SGg~pl7ka5n)ie2;v4tfH}EGim4M^t~9L5P}KM8)^E zTXPzEnoin)-oB{0bt=H+4Xj{eKS_dnYgL$Rx5B&#yIGC~eGka9?0dE8`{KZhkdiF> z`2q3x^psJqL=>vY7eF)1!yr18JG4xalOA9KsSpm)@H5aOkmy7Z-yjVFActLfK0$qP zUzs&AIU*XFbAv4SqZa8CuAO8Aj$rJmsK2>(YlaRzyR!4uBC8M=GrREf#JD`7! zc}5550>#7{@>$^Lpk)dPU=f_z?G8Eakl7ue12i-kpc(6hc-%WL7VO~DEELI1BpEY- z+rPXaY@~#e1;4tEe&{$WO@oSuKsYNhblooJ_tg6}jKhKI1!)T+``@$!4L4?s5!g}r zBvH9tO+$ryWxRi(tnX%3)-w(ebui$l4QAq{@jVT=Z)F+qPH+esMC zRO{vL{ju$Q%HpGE?0LY5+JG1&Sk6+Stv(~zEN2zat9lYs`3?8!v_@IVxr3BY$d z9*)#@JkWpDxw~=8gp&LewZ&^ar*}^x{X1MtBrc#+g2#bp71W2Bqy!kz;7=R@wjcZq zgQpI@-B`{6HaR`HCg{uE zXqZ~K+k}FhuGkUFDF*Yzy34W&1E^#$*1mSywq1W`wu5*%ar{E`?_&Aw^1cL=jMI1p znGzgC%`!&>wZ9{f*v9yP`~p#%to!VB|I z%n5&j`~?;?u#~FH|CrtM*m5SasgVlo+>5hzkGwA8ue)SJMa{CN%eU?lZjqU{;Fp8j z)#W}f71Kn#ki!5$-@vLavtSd&3a!~2$M@tMC1pI_l`(8*=V3;?4f+eL_aH6{vW|oM z8vWT$xX1oS>izRCx9>iGe)lVlIM{azzvcl%0k%3d5Ac5EG0AiTMJP z5>-ZVqKGDnB?wUYtwSxON){yPX|O+2UG&9NT`)~7`b$-lk|NG?R}}3_D8c^)jL%f! zw7b6$lW8WR*cVZg>&^(e%3>k&h|7@u+104Kv^DHPHlN`|>Qqrrln(kip=Oj&@t%Jd z&G3lnor$@>i$%1A#M@6hUMZ@z7AiawcH=F7nZ_!AOQ)^Y7BABk4PP&Fq@Z!vUL3y0 zMigS)1E-C{BBcQTImv{--X zQc6MuCoYYBq_0|zl76=hOJw5MeSw4G8HW7*X*A&>B#T2sg*>#KSkauFME=v<{58C; zx5X~BnHJc;Z`8#HKv?d05laPlWw&1Kw4bV#knrWPgN;?s=9xW)FpsXl5*v1 z30=Wu?`}z|%Vt}RyEBB={?IV<3&?x>mg}t9*D*agjiuoyXWJP$fAYm~x=xwCdZ5zW zhY$M92|b6iB7!fvFVcLWtreC#XIu(>F{PZk6Os9XC;P|smLNdN30FxUB)E5Lvz-GQpkFP^ARRE%F%n|bDq)L#|Qpg zj}E?7U$t zvW&I0VFXAVI5p6sh3x`q4L>BCMdCKyHP-7U`THKN13NpuXly4yu;ZCSiaex_ zIKvc3NSG3FWXi-=sSsD8N-^T^NHHlfKp_$Z=5k|Ki^ox-1CVZn1tg%ti0mR0!4aEc{4=S&C|%cTLP zT+9d%F5m>0Tr(5WjiDgbjD(S-75suGgcKzRC1fcGA|Yu(+k8Yf>150Z_9<|~Z&N&3M5SoPpio>7>$$==)9@+)s z=74faZ5J@czkvg+6(KXj+L|@D(zZ55vuO&=Un^QGYirV+TOYx^v^Bwpg1k+`YfbF} zl?=g#ogMOrO4zC~Rw&xfd2H z??M`_xPOQGQLFi0Z0+^6*xa<{>ouebt}m~f*3`{4%>D1Hae92W=UqjpM9 zZNWeydWvtMXABGCxLOZZ+Sy`LHsiIzeg_gQjY+fzoyz|CVzGpu<2LNII7Z8D9uNM- za)0%1zS?*SC;n%@?|1PB#>bPk^laUls=A0lEHdbVcFBei%!?340qXGL?HV5aFQfk!5+6EBY&G{c z%7LLI*;5)lhjF8)@;J3fby6E`!P;CHT$})t;P6h-gJF7i61WRf%nLR3V3G=^kJ5!9 zg?dYik6>su^lt6K)EvhNFm{rx3k%iZb)^Tx+nvjzT^JV4)~y0V+EdmkqJ7=J-+y76 zDmPBPGVSE!5J#$A>}jgqGy_9U3WmX{y-YT7K(3si=4HBWe^{^2&VTa0>Gx`O{_R^U#U$C@>^m^Q0b8>6 zV6mF(vl|76W5`_?8w94yZrgSi?SG+@RMYlR)37Qu`_^`k(rqtfr3KJ8EEWkX& z*ccf8RQi)G$_S7(Hnnl^aIvuh7~7ixWEmA00g4VDAfpw4+QA-R0<<)?H3vAD1Jr@q z01Y(>RW*Q=s*;AX8Vw_VsIZ!wqoafK|8NmgQ`eB92Z)O(s!ITXn)CoE4K?*Yf7OBZ zApREg07Z3>|4$!K;Gc8_33U;59c2j?rax-{umIeF&MsDe+Wr?eN)R)^-_$^<=FSdw ze>DJ5Te`YB@-i`bcz7^cxVgA8IyhS}I@oV}Yy$v)eL4eefqx0(W^V?P z)729AuL6HI1t4!_3bc0t{z;N__?OcTBqb;bWOw~HF^~|he`?zPD;(eg1pY@GOJkS6 za^;nkdTn47cnpB@VT>vH~goByRQ z>Hu1qo~@6cvB!UZ@0hW@n~V29wE53vn>yILSh={m{Hr1mU~Xj#{KMYm&z@P?|7BAU zQIwICP*Z1+2MwM*gMtG{9eYMsPuIVq|I`x^mj{g^7dL>Fl@-7Y8d3>+GcgA{I}mFZ zgg^3$TY+?Pb#V4#`ag4RWAET$@AH4{=2rG*=6{qob8}>W(y+I3as$eU|4%f?gz&e` z0_X~01^}G^Ku=Rkra$fenwCFimOo~Y41PY24vqkGV_O%XpOrZf^nu{xV(bnCxH`K5 z{e1rE_)mnu!UZt1GIa$FI_R(<{8e4X-rNDe^Di@q$^W|kX9TGKI;b?DQ)=d5Z|emx z1DYc+DLS}+f(C;6|Ia!5uU?XFwzi7Kc0lU?mh`{ljP0y!z5XNezaq4Nf4EXBIyl=I z+y0l&%0<%36KJMvGvXly}OkJX=>3&6m_ z$@O18&_tTr*aKZ$0GvF3xqu)Q{);b2`af&|Ov++^nktHFbpK~${tA<@H+3+xvbO-R za&Q8Sot=%n5ST%8z{Usv>DLzuR#GYG1@!0f>HpEZmxa+a|dUHKby(P0bmmO zWBM231TcyIjko|zVt*rU0F(H?5f?LnN%C*R0$`H*8-XGf|BbjoH8uZ6AfNWX5f7+} z?%xQ1)X(^D1d25I8?k~AXgpkOja@ANZDnC)1DTzTO+i zNC1Uc`~!k0{Jk1>5Nk^>M@yjnKO#UjtA9X$kVH2BfFO*G9a26{*N zM~H&4tFx8oCuYz!$pW&2e*X7A{r_k{@%Ih?k8Dv<2Tvacc91p+OkPd z%b8Q{_&DNu+oV9Df}rNWEXo20d1+pK|CAv6B5jJ$kG5~?Eng}oZ_HGLcZ~fN{P9sF zL`HKp88w2l6qW+ZI!S4cWOMU%irHtfT9aCd02-smVm$3#k8!LM@90;lq|eSgGEh&ApJS*D#e$`8jHm4$Kq*@dP{mzJ-tecbJn z-hBCEm}o9)Lzf-lD;s}dm2i&g8GfNjCSl2^n4Iq)!CmsBt!!=5r;Oy6Dh(l-hJDJ2 z)xI=PuBbGq5*j_I5uxac1sH&T9W{ms!JX<^li1Oh{tUD}wqeto;*S{RK-Z_PQte7H z3}DY3?or9@$_}^o?f${ZF7{p!f|pw*f^7XF1)TTecZ)a8kaE(Q5pM?%@}Tpr z_92#%Z-WATnupprPPle|&jr;L+KO*sMK(DK?j*3h;Pq3x%51TT9z}scnpi>wy0>P? zW@4fnZ4WFME(y_jC7L zw@?nSIDntWs-+(nL2oLjW!S}tB;s;>fmH)1NH3!E*7ej_q|xKwRY?~aSmxBtqLmU3 z58b+kK<+KBg|Ux+(ru!`BS6b@oRxTpR&OMLU%C0>2=S|Byq~9D`buBN^a9TBvwJVl z97lQ44mIn$T~YdNBNB0^ldq7v*e9hm@5yg`25d!3(@`W;bJ{a?YBAWpre&o039Jy) z*QQeBPV;agLG<9~XVN%$Q7ttu||U7`2G zeu28qrlz;-vK|{zuU?;Mt4K&I3`-Qjs-bI< zHt}3@zvZxh##Nx9*)rL>KrU&BD)j9e{&#ISm@YW^CHHUgCnrH{)1p`V)@AX+&y#S& z+XCc3EVDy7$CJ4sbmD+R4-d%~c#{vlY`9=Zz>KR+J)?cqb5|jf(F?P<2OElz8@1352^q+r*|1~FMS=5hP&B-usQ6VQ5=u6`$zdPo8l7X@d0p#9fG1Lt5G z#E4FRN$XPWPgsdaBRi)*}|;0Ue|j~7MdsR-#h z`fG3pSlvay=8keqk1dHKh`Qzig_|%bdfKLj$Htsath!vIIvY&tg*p9ogmo9>oPM=wlrKm zI$zox85pyv0!V&}6cgk?h`+%Doi1E|FMNNN{oJ6DfBOx_@x|7j8N$qA9NZuVcQ}6T z-jpV}kwmOvnel*I{Fr-(OM1KaQkmzUA4mlQwy7$;PC&Bg9xO?OsAH7j-2g#<9!SBU ziT*8B8NG&s+%2M^rOw6#J-mod(mPR-(<`3|Y$c*XTokRY0q_YMD5om)!Cd8GV+46? z%Gy0B(jaJ*ZQr`QHnGPPj=fxkNNxz`Qx%WUQwzDNhS$NQY7ywXI7$T= zTn`Z+B$te8ND_6zX1--3BH^lk#CeM|R_!tZrrDq)t*C0JNhoz#8Jc{-Zp{7b>HRG` z-+S~jHHqF@Ph5zQ_Cl@kM)Lq=@fmQ2jhT9^%u%0!VU&rC*FYItDr~vP4ihU`{i%)? zB%{J(_VrR_4ocd@OJjI_b+$H`$9(-`*!NWujg`aO4bpXiB6x!^3*Df9yu&7a0@f`r z>RCpe428Wg%yl}I+uXrniKNA<(4v`Jn{#&(v#OOw8 zy4g>i&Ds1bW00H!X-s5)M7P;Gt&2+HmyQ(|<%#vU2tWyPi^}x!)6l8l&0OOe-9d7a zeM@q9Kbd81qw%3()l{POoYV(qDY8PXzV1t0bgavvW-9GnyqleV_uWrP5(CB}&flXG z`=np{bhB-yluvy%FsctykyHF}Ljq=~>Qqnq54-eR3H;9x?`#2ocIpDT@*?y#w0ZT4 zJuDL%l^6bB=u#QFhOKSx`v80DYlVgF$X!B|W3Wbc|$$2NS>&p;6T9^?k z+`jhsgJ)_t4P+!>H!ayII4}^XkxR?Fbn6-Sjt1Rz1do^GTnT*r{>Cd7cJk&W(=04< z@NKkQ{HU6%@i<6-Prz}dSGGH}rV`~xdNcvqVDlGLeG1E$UR5QBsH4zej`9UpK`VM9 z)x9?rb$$a;0`c@`U#F*(SrsLyk<13Hw<41eCO#?%Ob7j-ToKvslBXzSe7(ThQ27}z z56Hf~qhI9gxirR7p}lO`U7&A+DI`^-*beu8*{XM=aaK=%T0fA5y%gdZbE-#hk@82( zmX~aU$Yc~GcecoB`cAhOQumSH_MAFp+`v+py56bx+w+6;Iitp_dPFm5h?~@ zcSE%>9hl$pWl8vx}$WB-F)=}4$miln6q*n(m1}LbhB2@|sCUMj9ymnjguHfK8pmaLu-{ssGCz^Em})2dBB) z&NLq4$+S>8zHi}1NZMc4hPjx>c0W{5>phD@e)PD|USX1@^XYTxMx1-!%~!pLF7%kp zbHX_L#A`gPr|Wm=*WTk1G%kEgcQzcfo5% zN#!vR>S;{t~ zjrtQZDrO<7EC%tjUzldJb-|^hh`rC3g*!ce+u}^rpp(B&WnrNfF%mhuP2P7ADCNP6w!f3Thtv6^~qOxAX@@| zS@~(|XAGpKVj~_hCC-QMr+Zz(6jDVxr_vN% zRca%by+mc5;`RHwJzTc^KE5(8IO9#|SUCr-o}9|JnUH?3eAeyln662ACoz@5!Ha{77d#X19a1xQ zli3TLiWg@#H~972=@*Z-++dM=&LGG2gqH1Mmh`<|$7NF-^hbc-owQAUhW+oK!GSArt1BQA!iF%;7E52A{RxiT1opM#8< zU(*N^sEzC~gW?*;pQz;@4{xF8R~Os1Mc-hFkHNka8q8xB^-AZT-D_HZd`>Uh7B8iE z^_w%4_-&o+>=Nl>@YNd)q2np46+9*2%**Ask1%|6#YtmHrNSe5DH|M63GWOs`#0Tw z6xK7f5*-!Yx5aYI{c=CI#{A`ARCcW5=H!fO(;Ll>4L)APUcJJI|s=xwGR5|KCiQrM)<>?F+>B2VPRWD z(lHBYYJMb`AB8TZELVa~5q5~;>bqsWmD^p*F$`;E*!9Fl89v!8mIg>?u5&cMrSF2? zu4L3^(@(VTPI~LM%up7i76(3AS^GZs!(S(x?F9Kdyuu;RbT@W?q8UORziJYSLh{_3 zcbd4V5bF4T5nx7YUmn~G0Y-Bl50PwKdT1=9l{L7WS03xfYRBzqy6S)532&RC5C_w3 zjo*N8u4$uhmo8@#H3(AP%@UQi`cabjbCv#{6TUF5hB9!L9UonTFG$6w9^=WB%h&1#IhK+isVS0e-xZY z?^c{-wRZ$7XTvY}(T|08kcH|+${~&5UU}xo6LIak zjq?KjnX~7au_bcVM167*Ce#sru^aru7dY`sfhl1Bbe{1jKr9Hx%eDV0c5G`$hLNk2Xz8214L{uW`TPPRODu)C*3APBonAJ`8O- zz8zCamsX~zDg!5{EoIwFPP^*H(yeYg-$%pU0!y(k2Iy(*aN`>-ww@Myir z+m`fWzgCQ0Jw^>0P?0897qiQd?kMCO1#^l^b^>1N3bDv$+VpGUTw*(PiYtPN+V_H}pau!U1n5SuD~{M9A=-NqCx< zowO&2fKBxve8KOxGkl)`MLN4u*cY63*SD?NXs|<8hdK%(+*s%)mb6Z}??a2;m=py= zvy+$1_sf-m`6<~Dn{D|Ov}KpKb5|aJ*=Au^AxTJ*n(f&1q!Xdr_$b-EXx5Ej!xa4f zsJ6sMsS1&}Gtvutu@7i;@YN$y(bhKQvC|CNliHA|&ieEpnj9M?tzFlX0>4|P^n&vw zGyS;DGZhU_rb#2LbrU9c&T6A3@8xt}s(ovB#M+zO`rsh?Wo@q_Y0D#5k#3WJBvwIw zv9r~+a~|&*?gUFRIq^9$fmO!Azr;iyQl34-rThccw3XT8kBD~6=~Wf(Y0AmA3qw3p z9_}Y%H=NhFovWtL-R;y_3Hb`7CG?3!9OhC?K+3v3?FJ(#7Mgx4XF5eT2ny@;AoO7P z+7#$tZ3q*|qO@k~hecp6koo6-;tAtx9$4~*He&3w5$~=#z+Xw39~uQ5Ro2MGk|EIf z5@X~&3hLscn4!4Er!L= znb$u!wW{k3H!3eTH`m_+`0R_3ZAghjWbkxV#+AdWLW6@9UNSgp9 z^yBiVMuV^4+}zBeMM1E_qyI2OrQ)IKOcy1t?^`$bZUHN{CW%3Lx(LER!m~Y9&vQmg z0NIB|3B3lXCZ~&w13W0!?t%WyaMW58toXLv$O*v}ir5weL{3|OZcCNL0j}EMHt4OQ z3r8`_>SRUm43aOBo8X}LE1VFg9?OrL?UPTE%P^m-g34D+RIS#SvlDo!nwXgDO3Cp8 za#Y;9D`ZN02oJw<)YVqyuXjr?Y=q*U1j)OG-718<@{IMVWQrJY zN%<)GrI5<2E~1gq!n;Mm)qRsNu^|X|X~)En?%eN9%lyGonBeprz&&aFr3%u%4jxK{ zg^&^Q&bXF1N@p7vVJuX!y^mIt+e1{#M6STpQI%tVMS$?t)|XI4_eP2k+b>r^ z4cng!Fzw!G(QuxKrhF<$$M7RKOgtq(cF!!=}^a<0+7UwT1msv3z5IEpE0%B{M zR`$9|&(#jF=F8~wFptmnGr~1PU{i!4!3cUdJD9P0eKS=>?VLG+98Pcz3~O=5S}?`4 z#GgM8>=^xOGiiE2$EfB+v($!n;ua5O8j9EO6Kag${@ulWwpibmsHqMP;_b) zBcsuyz>Z-f@kU|6V7#cEgXP99lY7nhgG-pKfyx1=Y72x}1M&%!B&cg9H6 zevG45VnE4-kC9U0k$zo-t4Uk^h6^l^4Mv z6^W*QTm*c-VfsP;WxTYOmX%bb)Kz-CHOrJQ%JUM(IFhRE277AUkz%2-BEKOSRM0Lt zhw3GvrhC*>scneiG}s6+9SWRQqoI(0~=bXN2B=5P|r(#1%%NluO+-u8wElo;bYY@jB6< zNAUcGMz#NHhN`JlIBY;`dSK@^u!pwF=LOKnRa==xL8pp_@4$8^1YuqItXUufbQM-Z zvzc!=N^)!K@fR^R!6ZZ)ma=T%j>3Ckt2z4o_3K%1-)i%umFF2s?YKL5hp-6*bUosK zxFiO<1A;l;rKiYihrY2>+#X~7B{tdgf=G3BwaS~#k8*P`-%mNT-Bw)5pBH8A`m6m_ zX?WQyz>b3>tjdw3xpxe!@N*o-Rbb{(U*UFLF85PX2%L>r^UadEDth>?ws|E1xJ5%# zZji8Me&b)^1~Xy}8Q3E;tlgCAG|7a2?IBx&n~>3e`gA398a3X_;E#E2s}W&;{1)>d zfeP)(j0Xv>1OCcOzc`-~Gi-s%MXGi`Ts=RQm)9B{l+Y=Nw{-NaIMv1n4A{Hgp2MWw z8&31WE`nM*TS)`EL{~_flN(1ymminlOLg}7GqXG%YXIEG6f#5S!7(Y{#V-thWcJQy z;*#n}{M7ZIm^~Pe-rM59oMPFT(VkxT7Uj!k)ae2p;h}{SJ)l?@_y`gOH4PVQc|y9E ziig*F;ulDH`rtiRqjc`Phpdr z5XO!F`-p=vqu1nv7-{gy?(a&k`2;uBxcv%Es6Pz;?mC!0bv_S3KX~&~O%;f(n01~G zJ6>q)8!Wp2NdEoBAH0=+f2
T>P=r+ik&6!Y}<}o5R)g8tU zg@IJOII*>3+nU(6 zZEJ@U+t$RkZQHgd$;3Y2|5u%Jd+xfb-|pMDyZU+7Vtsyzbgq}Q6)WG9>Ak!m9HOo7 zbG6j3`R3iPDa`_f7$Rp7Z|mJ@v(ud;o>=VYQ@O_y8_eD8T7enxJ8bUdd~O<*h!V@@ zp7wc*tF(+y>_)h|2Gz^LvLL=*a}tboGY)wCkTIJAVCX+MqjJ zckVR@$4-d)8gk(wu%D2iXsLahBX0$MG5L8K**DjzJ(!3~1$c3%B-xE8MRFs(qcR_7ummg=Jyqot`KFZFr6SY&3EE^`;!}fo1GJj2k;03Y=@WZ;5s5v%ST9&DsTLl1ACRzC!91Beo#1XLtaW zm6LFLnd(iV!B!~$M#&rUEj<1-R;0bRH$yr!A<6N;@U{TMW^|juf|Ck0$jK|E80&#@ zBHjcU2t>M~iwV^z@Gwa4r)v;mn>ouan+}lB;`?EgpnR4@Jv&1>lsx~s(SE`ZlBWWy^#2rMtygXZ}la- z%AvoO&``dSOw1}X8B1zI%Sa}j!ncM0;X?*8fNJ=OpSjvq|C!kA1s?ttgFK7&gmw)#Z3*jaDeII3EW%RsqJC zAv2smkr;cN(*{#;~AO64#D$udH=8?0qHoA+YA^^z#`z*gh9l)axB9hK@|KKQg zwlERW^hkhEG{E;;G2YhPo?-semo@_W???#DxFBZjgm(bG0k97=C$G|J~ z2!H|#ji+{S<^Zso6wYy{K-z8g#}a>2vv;o=W6Wp_NyOnKn|u^cq9T&s)gp43S|o3l z9L_?Au~b_|lnyCDV5YKQOm4N_; zSVJ+q--8|MbaAwR>UJ06#V-Nmcvszx-DBgVG>^(ErN)XZ;P2|~B^8y5@$QVTA6D-l z_bSJNqaOB`;=5Z8IaYj7E?yVWDJb9C?7SW9H*SN5uXQv#ebQM~t#u6MsbuQx9KUG* z4x>oaIDG^LF@s6uU_ijILyNTUD&Wd)z4eC(J5&E#>vKxR1KvjABo-n_6V;Ixd&;L9 zN2u;(E3azqC6zMVd3n}Xorp%fTqAk3zgw4GR$BRB za`MOyQ++^!_TQz?7@KN{AHj#FV4_viGzZV9Eq?AncQ$B3L^?#zg4D^Ugg{f%da{$E zeo?n8PH^N_FRJvNhdvwhI+O3%zckAf20W|Bs%^>^kbN1VLo4M!3zjBL2uDbr5&CQO z2~HVkE#|5@tk;yHL#2EwW9Lvyhw2vfb6Wz?x`(Jx%jROA`!a9G=k3PX*U{6=zO0ZJ z;l;D7|Hi-o)piqIS$M12#?@K{Q zWsNL0Md;**6?LO^pD$&iysx&SUT3{|5Z>rQf?gW|_si3is28Oorl466h=B}lU`(d% zk%~M`r+WRN05$buo)Vkg0VKI=6*q4Kc7xFZ48N>}wPDZQ(5LW!uGJ0FjF%QPnULQX zGFqMSZ4BEqsV{=Msmc&UQtpjbDCf1Pi`9foGv-;ts2NTQ z8uD~|kDdRkKCm$k(L558b}{7UbnL9-6Jj-9|C7qyPCC;Cf}t4nd{|RalXIsF4(2G% zkoF-SZ#WMUhhImv38EZk?T{CB{{=dG%{Tm_yU@|n#Nh!2S2vix*}@8_*(haqqJaqp z)0@oBZYn=lq6awFgzsRKemhCBF60F`BEfBxaUjPj8pun3n#Nm&Sc%s&x4`?d%FO$938umiew8)yvMh2d@ya zE3ngVzZM>J^!H>H0C$(K2vsZF5dnv@Cb9Iwv#M^*7BCov@#>{&*E`Ij+p$`1!90!k zRo=k}PP5aW&R&H>hO)8D7pllsF=l5%^{)QUXO+F(w-8~M6Xc{J2d2&2c#c?&`b`B+ zh;1q30bxjm5QNs1lFn|gm6B&(th__5y@VlldPD%)GB&XF%dw1c)p8y`quUATIkX1@ z$)RBP)3oDfNh2`qQ?>v1ZaKFVtbX-jsIMLkzCq**CvY$%~0cqoYbx5@4&aU z!_5dp}yM*}66$zAQYd7gM)4yJ=8puaC zrlih-W{2Mr+tI+!$bh43bkalk#E;$C4^{Y$C!M?x?{gzB&Z=jRW_fF+s1UkoJ!cFS z*WOs15_6;e?EQ0S*&Z);7V+r7KlcT!OU>y$MS`#wqF!Iq-RYu%a{H?l)bpxy59`?b zDboJdhs$($nd)p-SY1&CoqGrQ%FaoF<-Bm9#zw z!q7u}yx68%wT>~{d&`w=wA0aFv$k&Dzd6CQgfHvh2($D#BjEPZ z99#&hEFKc1NMY32G6vRlJ#79;u?%$}nWqHU!nj#(YSM02D4UPB2u}nF*S>Wxd1r{rpDAY;_DTK z_PYG31fJDwXW+G@M_=4)ayE3XNfh~tj50CS0o+hC5V0HV&I5$60v==eB!A)>xzT;y%_)lklb}w*`ZmS;> zUA|pk4|FvyL*g6D$SE(n2MC`a=1!te;!FfVyJ1by>CeS~9g)Nh1Sij~OjA@%wRzl; zr)s|GKOd+{d}8o}pD78TX16CiO~obrO*IL$!2x|{vT}fNcj_*$naID#8Vz<*Xcl52_1>i?)b4 z>#Z%no<%{3EFLRDij9-M1rNhf?k>01TAv$jw*J8DvYlIAdjT(74bFSczZ27Ix0 zd$i#$lm-b#2GUrWltzC)Pcr1nQim)dPuJ^P zF?ej6mPgNJTwqOtP>H5~u)buPae|LRkqCi}hGv>S);|PergAf4h3=Q;1zruWr)-u5 z*Kf7jx|+lU5gV=s8WDq>;{%c-pk!*BiGcuu-API-MT&L}-_E>TQZ}euO0yHAmzvD8 zzLJw`T1|@qcgy|xJHM>OS&)*@1;E~`8DvhS8#-Euj(5uWCK>0Ua|CVNno=8cUsaT> zXxWSH+?YkJQl_UCr6GRJnw2*JBw^*H)||JL!QbWAUZ7V&PAA^CeK^}{`mInhEtNvV zpFsv?W!z&g)e5A^d!wxx19bZe`{%xi<&_ep2BfV7I+vcH#eX7r8@@MgV&Rr0ha)Ol z=y+w=tbRF7xSXdiTaREaX^5X1QG>q7T4}887c5imOdNhwCKLE5e$9)DF_D_j&o}&Q zw$pD!0~P(CQUR~+*0K-Hl--+TR@tn=fa5&z&}2She5jR{BW~K#w(|f`Y2*;xpNPqt z*}xgy(3;uDZril?`c_htThG6XvpQqEzDv_Do{8K07|pp1V) zw5z{g16==$QqJo<4Gj+@vzPq2y+~Za+I&fzd(`9rOK>W~kgbgdDmbT_4(IH}VWCS3 zr!%#HqNSdJsRRPX#hS#|z+-lSFp-81cBHt!rG6&2yRalIE7<*nS%Y;hy>d50j`;|R zgl(i0bJo((R4aG>bL;eVru4%YgUUq?NGmv=&b*|hZ2c$v!!BugrnbYftC@=&XFTo| zEn( z@7D~Ts;9)VNF+ux+T}%)KxIJ$b5B&;0o2$0rlX6~oWMzz_0oRX_>^)&^T_+!+_)I% zU+r>JaeNwVxpDPwz(Vmvc^Pe7)K{_Ug5^$_vWoU`=rd=tPky)Kl{OpbcJ!4Ig8mFd zg}z7tAq@*BIX~H8l>)@0#7VW$4=hEq)*xqNpAyUZHTewH2skUm=d50$VOt%$gkF9N zg@`5@{}(}H?%zB@^Vdt?{J$c5ulBY_BW>z2KxO}mF=bj@GS0oAzM-)Yb6ws91$56K zDeJ2mqyNo;lQiTGF42c=hFaxF%V7(G*W;(?5X}235mU&4f+li4VZs+9+{e1t@d6*m zuGJFmrH`G=oBT_zlH2Z49{BsfblS97-b;^}#zl9OM<*lJ>fvs6G&MPck`x&fo3e#{ zD^Hz?o^JN15eBTCtMpI9yA6&i7ao*;Py>f{rdc4-onONYQSqvqwM8-#IKlx)3@|;K zMYe89zran`!(->q>is>YsDwreY$;g{aBn0Svh$+K!O7O6onudf*)h6ri(zu&44L|N z<%HB9PtP-~VL#iOFEjj;VO&+DTWb^8=B-g+1zn>J8!O8^^xCvkm7%9)#TJedBlN;^ zLU{g)>K6jIyLcsmS5a{A7Mccyyd9&u{YNDH$>ZV-D)uv++Srp*>Kz;uRlDs*t0(RE>B?2*_Q*-$b+Dubb)VY`wFz_gHU!NehfgfW zle(DhVf718Lg)kh>2b7F;2h%|E&+_*A;uu#a-E&g%;0YlHstj|QrK3zf6$S%-Dw87-t%P=7%;VQOPcpadG6i-R+Iz6Tkb-4i+r=t0i_Zy?N7u5yL-C9#D0*S55$ zcZ?8%;*fgR+oQ!a6R*Sx7FwAlhtQcZNC!@AcPixl4PZ50sdgsR9cp)Y;O>nTjJY8< zO4MoR)R>Xa`oyJ=l4IWQ;__WFALl z_b6X39Jy*hbs9%h@Gr0q7CC6U0D9Iyh|n9iz+imx1iRP=(G~_$pFPFl*^h+VE6uX| z8AJ71hFp*tDx^xgZEWh%FeH@|4xe_4`9*ps?Oa7~?g%h>K)!N56Ch zq`F4#ib@Pzn;ZE;l(Uzs26(G0{F#EF+Gy1=j(D9n`Yz>_{IH_Ye7gb0%xOUGKbPy; z&h}mj6cKDtXo4SZpaC|!7`uaobQt}r^D*$ctg&p=(-9r*-3=>!=Rq5LU4)dKx!-3wvPkvMkkqEOnsO} z9k-24&v@YP8+eXQUs>3Yu{-GcjM!u{X-;d|0rkA>1CNXwr}Xtf8uZ&m0Z`vtGhUc) z8~~zxgdYFvL?8b0xbtVweuY8)`oRBbwoyn-~uvz)Y)Eb2&Q7vY0%|}hSV(BQIl}}V8 zIORN+4tv0!aT*@D52N$qfX{=UL9MM;<^qbe74Q`_Oz6J~Rr|hryLTz!Gprjpq>&L& zxY4u$CjXu`ErL)Iyz4A!YG0kGZy~Y}kiZGC><;?D$n}bp`*X8vD#dZEWtzW|k{S-qT6ZR;9Vrqb!2&n>I{ATql5kpER`bQa#>bP&V>Lj5RN z(yr|4dInPgW&nXyj`|&ipp+p`NA3Xe4io-d#&&D=rBd9Kj}MPUi)_~dtK@1EMk+r{ z)7aHu%lVi-Y(nq)=5r8A7e7FgWgDb*yh|*QZL9$8ll62b8{C)z*r3#{umpoRTWOl`o9$XEQ&QlO3D zaZ){`z|;ld>dj|v8xEb@wKc^%V_Hc>D4bA{H`Pc7Dq8SljOw_=%8^^sMheKU2afy6 zEt|0!Ln&rPjqrK1Mn8pc_71BL(f)GFO|Es}&zm;S$_}~84jo=XWI_2k>9!xje}Gu@ ze^tg(BB4!g3w?QYnLxik*m46Q(2-hLO+`e>JlIVMo>)E1I*@@z2M?|Yy?J*z2-2y= zkY|Sy{stXMjO>OJo=Xc=zrI0z5mt?J2$Hm~NN=91f&Fo^2P zqIEv0or1w`Yokqk+91R>Y#1)z{OzW6KXg~lu?z{v&6Y~s-=Ytj7q{%mCebpciXn7C zTG-{4jm^c53v`E63}PZ+3|h%k-NvC@Nd;ImxD{S*UN&Ys{eq|a8p5FZ*H3**M4c8O z3z%LbOgy1xH6PHAuv1F9Gek@}5a-If5(@QnPG!uWB`XS@I{$LA8>ndGI-ci5cPq>| z{RMr=Zv-y0%wws#DrQjh1)|T>N*%Qkmhx10eA6_#~%6wz{o5hLvuLg zw;5&`e4vyGv!Cx3y$s5JXUuUJd~kZ5|H6wmVlvA3C%f<`%L}6=uzWv6mQ5X-z9z)4 zhuU@c&#bQ6UyJi6lf0*S1{-}Zx~qz{9f5f%2vZkJOxsqZ61TzDFfA+rjKh6cL#cgQ zwlsTh=oA>O`%$bhd0xS>5?;*T2HaH1N<%M2HM6n7Y35^p=ZrxOvUEts?q#kV_8oWY zQHg*3ed{_OmAS@hr0K|h2cv5Rki~>jog1Y^WA^_Mc|AeO3A@?A#I7#W1(%au@i3;6 z;~?~>mp*vDkANxg1)_A2-?%N)Hl2jryVP2|BN#Y!Mn0*nsr<^uMBs{ExHuB;!GlVRUQB@#lFfF1*cz7KjvtMh#Oad!ZU}ZUg z2-{;WSEjPXCd>LUbM}maCq%9Mtv1kJj|Z!TQckOf_kTcBNp)d|0LBGN6d|(KKEam7 z9wC<#SZXCqvWyhgb!Vaqy&ZD(dyM3*JY?sO2EZ)v`cTY6p=D&u;4212uEP)cwe5U7 zZ=5OOv8V%z;xeAjF6UqsTlpVMTNS%fuzJb70=mId5$VZRSn8`(F;hy2sB=Hg)vL|> zp#R|x&JIY}-vY60Rjuib3rUS@03SJQMK3GMUkYQO0~Gbs)0DP~!!+D#k!5R^(-4A! zNS9w^C-g~OdS>FA?5s#?(dcD!2JQH3|9L-hzUNDu8J0^m4Qa&xg~2q}#iMI!my^5E zK@sW~pllJ~%3AF3w{OX;!Og~M>#`Xa4V)n=+yG<(u+aW}H_1{VMi@CpWWY7sN5D9k zDo_?#W3+MZI4V*rnARvf;x$?qyC=d{o}f-e{EM7n8!_~$yX#(iM%{%GiA{*UefP02 zJe{=WE~JM%%8teY$RpUEMU1lUp3{T+7U^T<2s=NQD_y{AsByknK6#>)is-pz)_#!m zM+3z3ci={nNZsh-JkR#zf#4uUvq(MsfCLP(?VrTNm17^#*-*;MG$X~69<*NNURW|S z6d9S#eilP|xqysQ{A5=Yck3Jq3(lZyyVKzn9r4d&=e|6vbaRT?Zj1`!Yq#C9CiFGy zJ`uu+PN;a5wW-~Uk!)ubv#oh^NNHAbsY>aC{$lMRLa7nwJ= z10n4OcL>fV<2C)(ZcxcCD-=Z~=jIlnlwD7Z(LEqG-ieYO&>yR!>m2@|r#a>@Bu8e{ zW1va`WnB#+Age|6LRm@i+3EkLLn}soB3!jw?t(EXq;A`~s71p3NR7pgyXttizye}P zynEy<@=nuV&$2E055@4|GCOBt`^)_eNaQogwT(ERnS9@MU(|Z-66s9&n<%BxD)_Yy zvbVY{GtH|6;ZXpyX*#{Os)i+D?hvxdm|ryCC!*n}uNzzmbZ*}~&WLCElf>_9{HcR^ zwH{c%rD452IXubtaG^LHX=R_wn@SMt$6aG<2+3gy+cqC@#g4TI+0VU7=?^zrZvrP` zJ@)bJGj_;I;&sD}1ZeKE#T*$g4W&&*ftrgK;VmZ-?MNbH_Li1thJWHXzJkmjo1Ppi zC=~Tnimv&jQK3X&FeE5K6gjys(2#Ku7|_$e=3w_VIcMmmsI5POuv zjS1H{kLK$1wgD^8RD5ma8rb9lk5AX>;G&JaUIQxBrXaV$b$MtNYz*$2zB+vaaBh+F zi4=F^{_Pit6G9}bta;s{=f*8kdkx7^Ha4#8um;DGbJlOVB1vw895r?mg{XvktAIBK zh4+;%JFZ=HyQg1UFLZ}aAOPeeoi*KOCFeqU?Ngeta3+{3i6}3Gj_+lN%0{=yuCFz^ z*vtdbJ#2XafGn=KOWPkY*zhep!m1qwdQJadygI5Dz5hhVXkBy}a&Hh0RHn<9ShO4c zk=jw1yP>q0Q?^7dj}=STILokgJi_7g2e6`VvQ&J}rHuC4?Nnb*JluWAFddF(%8{(z z{+_0!bIb2`Hb%Z)70FSYKYxSQQY)^mKMj#ZI`!Wx5s^Cme2TW;7gKG`gF)iQ1*yfE zeShn}(LXWAn#K-I%W|K^>hbS}b3zF$!$5PAqTT5TthQzaA^^mGU8xUk?3zF9DV{z$ z`Z%sewgt6>;iN}hH)8qf4W%sXsTY|SW*t|aGS6O;%FwYaD3`!sGbiZ~z78x7)(7Xy zR2MJh$^Pm?t8CdT(1BOQlt(>e1$i+1C6#s(ZpmMPu3zb{?y0dL44ZThX=)g5VPqpC z8-{2p!THdZ()Db_&(K)@RU2@K_1ujs1o^-h5`}q8Iv{ZFayofcbu)0348@rjp*L0K zqHJzZpH}ese!MYCuwLd8f|gT&lD2ybfDES&%=L_z^xC7#Y|%=)T!cNi1%(%tM$K}9 z(h2b9LIEy!w3P6-Q<)6MG8H!ai>e&Va2@GN-=bykbL&4<9O@Ns$mcwA#Xc<~24&F8{sm6_pOKCsXD&`J9F9C_~mtV*bqQ z)g3Q`zcvpCNCFl|!LeC3K_ejpWY{*0Kxz5xP46@sRTB4B2HvBvFFh0?`|(&Yz2?D{ z7-SXEzxxx7QdG!vT>seN`~x|}m2L;E{M*cxQwvfDS$KKt^7HJvjtRhDp~^9i(T@q3 z^Ns!_d<|Yo$?S@UPF`hOd^XOEJ$!Go1rMr`r#5e1a_apPBL zCocd%&>E*P{~vx55uNZ{eSL&qBUWPOvx4gOhT=}hy%zBYRtYTOG2o!_?bD0Rq7AIcrqUnMW&i_ z6~6xv?#YwtCP5b;SqNAR@SKaa`Isqln1XUwA7ikFJ>H7nR!O=qu@N@aH5kBYSv<+o zyWCOG6m0YjC^d(dkW!;a|Hw5G8bc!M*^E7ybo-TC@!=VznuQHyAW4MA{V3n5vg**qv!aj%OnP@-0`colwfyf}YjDm+LuVTw)9c(tPy~_j^_fz}G$%HgM9Ya(l(8~= zNkVlL3mv&e(Ivj%1mwUG^IySfJCbT^+GPX`ZLKwk3vrZ9?AG3)2a||yBTtDw;jGo z04wMQluPV-i=>A0pg7?>NlHhR1fomoGC z$7P2;E~ekwEPDk=LS+Zna6aI30a!S1u{=GX>I=lA<#6krSx+Pbi+(nZOLHcwj{2KidPc`A_w9Lr7XP$*RR z_p-0xr1qm05R*DX>bbD&jRg@JJB{ks-$*)aZ+E?}4v03_O-iaigfYnc z?nfsr>fhs+*wA|cj{_qwkz3t)Vo&#=3`)Za!r24C?U}XMSGp&`FDE+cp}*>LV?MNB z!^(!nDHq%AMvI6vV)K%jZ2PV18co6BtTR)^D+FG#ltm8L;qDC#2#(0@au_M(K7=D0 zE8uNxI}E;T)kTddCH%9lqT|>3*6i2`MeiMtsSc~BxHsLXP5NGt*QD9%`qkB-fD=g+ z_7{EYde0P_)t{=rscy2qmZZ-dea1FplJviD%`$bO&Adt}KVBEVbmgI@Kz`cQ6k;E_?C7UW_EKQTlKa za9FnivGc#br$W1Qm9fO&@cexEwEI8!JHAwO(A1R-}tm}z^b z!vET2+p{Z&70}Cw?Nm{!pbGQOBS7k4hUBq)?cU}^qVq}2brXg&+AjE`3Bef-n4O4j z*UVjnekmx|Fu=(*he z>4Q-#2`d1B`dNsd88S?|cN2-6+#luN*1IB-G9c!dT;Hx4ZMoyX{9ZvOnkh0T^-3Ga zHbk$L75FS&kydsqmx1@nPxE9OEm~&D0q)>+n}y4Nbq9;}m8}fw62{j|UUO6mxFYte znE?rQuuF6d;caccT21tz@@(D|b9zAhrcRg=WPFG^ z@c86`t4MU!WwpCtgD7&yy)agUKEV^~#tHHx%)Wu!5`P*XqE8=?FX5s!z&ARcQnr6R zc^2iyT7;-PBnieUo z_C);NDZuS}sb#_s<5>&tK!F$RTIstY_3VGGd6%^`T#INGSpow(Xr$q?!iWVHm zv*zieXl#X+%=+~J8u(R5=sYfvc}?8KC{d_}d@zL zA@IhYjRn!2w$oDQ?V<)n6=RDFu#3_A`gJdv6Q!jhuHM`aS}I*|15E83n+qDZjRlYV z%$JbL)KW8k{Vys)@PgUparozPbFGJGGGPJ{YN-6L83yv(w*~73Mxn)et(1f2*FWVD zIIX==8J3=owSRU$8xL!o%dvz>F>>~)zAQ!fx@0-JPQc-0(T~ezd^x83hQ)OfUsCbv zuSDRTa0^GnL4j_R2>%-X;kg@jxCvrjppuyl;f_^owx9g)cym1tGUlSlQvqm-#Bku2 zkTa$cao1{t$x9URrQbF-dtTs+Ub2UCS!Xw>I(%XxU3$6X5hc!sH(Hxsoh9!$PwU4( z#W4S>0`XH@2oxhb|Q?B>+#e&(F|VO*VY zbNm%T@ExS#6-X7rvuJmTpl+%BK`Z>D*}f{fSHt#YBFWk&X~$VtnNL+=ZuUtxWS52s zs>+nn4aA4_o_jgNwE6X;l+KTlr&@U0$Ex>jy>qObU-c5sMpMeYv1|DG2SHY8-3Kx3 z)Bte|t}d3oG&-tsLL zWedv%+VQgOth6)n0KP0vp)ZD%#wHxZEf;VYkHu4aM#vz18e&6eSZV{ z+%G}aMQb*R6GPn$=zHs7Exdklp56egQ5|@J^CFuZt+y$Xz4aKDv4o*g3IolCKpQu7 z_Fr!qY%_u3uvUz;D0{h~9^O9==@;nlNAjfb)l@<`0E#_ojHDXfEd$6uT{p{a%4;Sv4rwbT%w;qNfR1o8d<+02 zFlhfJujd*xiyn&C)f*!_YcX$G(?&6tG@>;eE^pBR4NJjFz*JS2iUK>;vtO*oex;x-B6cTQ}X0}?!_f7G;AxE>t7}#Q>gK~&QA;*%bbaGqY>)(k-3~MbvT(gm;7C{W_%86?LLW;+SH=@;t zJEtuG&laU%&&I6%Gul{|{NR|H(%`9{f(1X*u>d_=vH&*Q@}I;eB&0FeW*T_eUy_=F z@ZYqC=SoVczY0`SksSyVac#K<9&Fap@wkX}>ou1Nyz8I?z?F?SXUp~$hzWrihKY~~ z4#bx;ff``}yq$x;(~lxpwg6D-Z2R+*=4^#$_PMZ*z~O>nz&=|8U?gF30^msxOew^F z7?lFyKxOcl*~u;`s#^~s_lkFhqY%;B0Ua=IF*V+5zsQ8PgFzj`^TP7l5blP2=cpJs z5Lx`S?96oG9lNe!7Lkq$sv5zavHBpGYblzew z-34u9LZ;Fi(DeXq%s{drkUy)eCQLHVm(4FU7HBX=r34kFgrIgS#DU0t8W1J4;dq2F z5(F&p0nciOHYIjfWX12ATQf)Oh_q8DhL=*0V#czZ0Tq0MAM-sqvmxpWBax0l?4!&> z3D0+W)Q-lMB#FD-=nuy-4S_G*QY`he+GKhYe;+`V6ezfvf^5IfO9_464-SbQnKrz= zV_X*6Z%jD|dMdD&-y?JQ_$t<{KFtD^gD2mF-BuvbT#p|v<&ePV)=W_Ps=XW1HE?j4 z7EKs0L9eddcq?nBCW}sf0=W3{y{b3dY?^^q46}x9{fMr>koY6Q*^)^-u_T5?OO(pz zw!Jykov$zC16}UJH$oPd2K=~Lj)`BIs;)%lJt|Wh{))U)V9)P65HjpHz`bRLBBA!4 zkdJMeQZb^lZWZK}C;>;m)@;@OpoCPZO8oxfC131V%lZK{%pP<}w%s@~ z0&W~A7Oz%BL);~IZcWfTvWD>GyZU>Mgi+DCOZtdB_wk!<#kTXnK%243-lU;}wy>qr zrV#j+#HR;IU;>bw=|i5JT8PQGx6I>+__+PrhS7cEC)vBZ9@cJZQp1hm{TA6}M@9=4 zXw0<0xR&-Ugsl* ze)oKvq_dZF&0}`h?HqLMza9{&|VztEE?TBtEGAd=6ZXn>jk{g#_E#X z)BRP?aoS{VIgeQDvdJ38gqGS5nMAT%Hcq_~sW9zY%^Ea1u8IL7<2_$IV zS>4RB3ghKG)KRv!(FFmrgI3_gNkJw-Ri(SSJY{d;?(Q;-wIDN zyZv7WZqHM(+sZ*ct;Q{P_mEVcZ`!Q?QuZ$ETO*}bR#q*iiOqICI*quWsh@b;G=DZ* z8rXqLpMTAT|5GBcT4J_DJ?hLEE`qFEW5}$hX`d`x$IV@p^nC*)wfQRPMxwR{!tLzG z9GpdKTn10@J7<U z-svYR7rg8q)v)*$?0FR;uf7@_41v)F-iYt+M^*IHAo49C_`0l)a{d?XhXPID%&1Ij zUD=TP3jSfSPVI&#N>Jv1@|Vf5RsQ8GWisr|-LVL7e$WP9o6Y2hfMRn?eZ<$K?KuT5 zS6Y{Hl+5`V9^J@=0wf?l6g7&dlK5l`LM#)-DM2eFPa(oH}#a{hNI4s8aRVJ~ijkK{rs*gKY*N0LGi$EKEx93L} z!-}d=2b0Y#u2UtBOcQu!8N>BoQDH*h7!IYvWVc6hp@d}vC<7rCSuzYw7dnbH3}#{K zB}7Sl0yrdUgWd%n5{1ADP#3}n$Hu7-@Pw16!nlKsyQjvvynR49Km(D+0rX8e=K4fD ztz-2S&$(D?b6hGC1eFS{tQ_O!W!wxYhPLw^$_RAO374$zQn>X-HOaf_qCPV{Ymrhg zGFli5L)f_-L-(?G#5M4D%b)Kg1JY;C>I!~HoGxNuvqEJiSSM8~(Q#0>Yh(7S>K=kA5&SeQtRB>5cGZYT0prFEte@^z*NlJKGD(vlH4 z=HgoM;K~mpa6{g7N!Y&2f<~;=o8aa`W72QuT{oAKDkKT8eM^pTnk;o4SkJtG81WUP!*LuV~Jf9BHIwFe?v?ZYj%@= zV*!gPW+E@le5!o3Q*N~vn;FPA1X673Pjx0t6DFGMPjd#S1K*kSZFHA)|NDK5G~^@E zR9Z@&G#U8FEr7sIseWjn%s*v6*)5LdbyZD*6B;5lZD*-I)9x9EAKfRW+R8a|)%E3> zo3N*@bY3-E&4*9sBirS*xC)}Ri=KKC{diB7fi<)dA!&i8Gm(jgKQ~)ntCia_D&LV- zG7zI)Ddz+3#2-c-ggyFl;ce5EG`<|Xyzb$8>l60!`h)(R<;}(#$k)Q4_oE<}9=f37 zpfH@K#3qJ4415Ur=n#w<-j3R=k%cN-@VkQqhOkdLLI{1B+PswG7-|YbsCK-G)?G zn7n4a^&qhKvEQ~_1PWAF*~`&O>rkCI{okdTda`M(?&CFU;W6v%S#{%<%8fKxB3^^` zw`$*QJ(@!s(a9FJGbfLw-UBB^jJEw)8jT(1QL`#7rg>`?PoaKh-{lsIL}a9rWa*)* zg&*~X*bBA0IUipy|2c|UNv=_?6o`;f7ixqyET!JQ}0rxaM zNs1D_aO>fxd2sVIKVDZGau3mo0$Yjmf0QtYTrj}qGAY)*WRsB?fVYGW4gXCJ)7vAIC>sxzk~bTHBQUnQTlRaR25z2g7a%v4vy7)=^Ck zRie+|F#~7o>_}bH@O$0Hx%*Nmp*PoJ`avVaCYEt3Tv;h45qM@Ioi&D7FTT7m zTY7ukUbk0!N#v{c#L^Cvw zN(_l4m(0S-fB`9$R7c~szrt|`O>_;BFJ_Sb~XX3P|YzXhu9th z}XO&Z7I_cX-dj}WWMZZ3~(TA^C znlFC*&4$mNr|Fs(O6lDopb*0DVC!_}h~4T3!NGY?m`#?(l}?S7|0a7E5)h=z8fx<@Pls!I5YA$-)!%K`G88RO;XS8CqsjlO&EH5uG zIvZ~Kv3J=*zq|q1w7r|FZ(d={wR7s-QC~>4X==A=p2iZpgSZ$-#Smv{vR6jTqAA58 zJlhn#aF<#qX!uDkm&=L#jcxbKfhkSt9sxa#j2CGZf{B|Y?U5IW_5M-76%(E@&BqtD3&H2 zWQ}k}@1B2_uwG{{N2_mZx29U_U1&F{Nh5iSDpjbtnwiM`Qvt$7JTTUw`a}$+5&*sf zn>F3u$&EQr-q(Ep^-x*toO925WGYKP>PokpzM?1qfWe8ysu6a`D}CByoO5z*!!2;| ziXrBffzdkVdmLNG8t2E)VcvtOweh8s~MGAnXkB127#4K$mmQUI29(O)jgiA zAHNe#g19NmxMkv0@ZDhk4ptQp&pOecaKJnp5@Jo|`YV_-emh1T^oTbepVoW#HFI5% z@C8v!_WGn&ZGG1#tfy(*G1De!eJwiMj?0=-vZ$C!=#ny^rDay%wo+-;pL_TQNhfO+ z^Z+&i$$>Ru8im05l!{bPGvcsca*_JvhW~KtfLOXU%mF~pD_i8hrEy6#vKS|KcK_{l zUzW2XWDycUHIg=14b-K4<;8$Ih-$+PAbdHr&L;|!VFRaU`AM@ovFIg7NXTh zYBvAC00yD=Le1w{3k{R%Bvg;kU*g0F+6^?MENjYXpE#JXn-ycfYRc)HZ^miPH;yUY zs}p)nvs-M Date: Fri, 3 Apr 2020 21:13:10 +0200 Subject: [PATCH 7/8] Add features, requirements, installation and execution sections --- README.md | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 62 insertions(+) diff --git a/README.md b/README.md index 5c1b3f3..acc4113 100644 --- a/README.md +++ b/README.md @@ -10,10 +10,72 @@ Promod is a Python tool whose goal is to model macrocomplexes of molecules star In essence, the builder approach is similar to a genomic assembler: it seeks for similar chains in the different PDB files given as input and overlaps them. The builder places pair by pair all the possible chains into a single model. +## Features + +1. Building of macromolecular complexes from basic input data (pairs of interactions between molecules). +2. Optimization of the final model using MODELLER. +3. Graphical user interface (GUI), with the same functionalities as the command line interface (CLI). + +## Requirements + +This software is written in the Python3 language programming. You need to [install the Python3](https://wiki.python.org/moin/BeginnersGuide/Download) interpreter on your computer. + +The [Biopython library](https://biopython.org/) is also needed for the correct performance of this software. Independently of the operating system you are using, you can install Biopython using `pip`: + +``` +pip3 install biopython +``` + ## Installation +The recommended way to install Promod is by using `pip`, exactly with the same command as we indicated with Biopython above. + +``` +pip3 install promod +``` + +You can install Promod easily by downloading it from the Github repository and running the next command in the downloaded folder. That is the recommended installation procedure. + +``` +pip3 install . +``` + +Alternatively, you can also run the `setup` installation script, as follows: + +``` +python3 setup.py install +``` + +You can **uninstall** this software by using the following command, if it was installed using `pip3`: + +``` +pip3 uninstall promod +``` + +However, if it was installed via `setup.py`, all files must be deleted manually. + +## Execution + +The basic usage of Promod, providing only the mandatory arguments, would be as follows: + +``` +promod -i [INPUT_FOLDER] -o [OUTPUT_FOLDER] -f [FASTA_FILE] +``` + +Additional arguments can also be indicated. For further information, read the manual or use the package help: + +``` +promod -h +``` + +To run the Graphical User Interface, you only need to call the corresponding file: + +``` +promod-tk +``` + ## Documentation From 848b7c07acbcb350fd45d653d97e1a3dfb8f5edd Mon Sep 17 00:00:00 2001 From: Claudio Diaz Garcia Date: Fri, 3 Apr 2020 21:16:27 +0200 Subject: [PATCH 8/8] Remove unnecessary text files --- MANIFEST.in | 5 ----- description.md | 48 ------------------------------------------------ 2 files changed, 53 deletions(-) delete mode 100644 MANIFEST.in delete mode 100644 description.md diff --git a/MANIFEST.in b/MANIFEST.in deleted file mode 100644 index 6462459..0000000 --- a/MANIFEST.in +++ /dev/null @@ -1,5 +0,0 @@ -include README.md -include manual.pdf -include manual.md -include LICENSE -include theory.md diff --git a/description.md b/description.md deleted file mode 100644 index 514f525..0000000 --- a/description.md +++ /dev/null @@ -1,48 +0,0 @@ -# Manual - -Welcome to Promod's. Here you will find all the information you need to run this software. This program is a project for the Structural Bioinformatics subject and the Introduction to Python project of the Msc in Bioinformatics of Universitat Pompeu Fabra. - -## What is Promod? - -Promod is a python tool whose goal is to form macrocomplexes of molecules starting from the interacting pairs of chains that will form the complex. It is compatible with protein chains, single and double DNA strands and RNA strands. It has several parameters available to play with according to the users needs. - -In essence, the builder approach is similar to a genomic assembler: seeks for similar chains in the different PDB files given as an input and overlaps them. After that, joins the chains in a single model. Pair by pair the builder puts all the possible chains inside a single model - -## Scientific Background - -### Introduction - -Obtaining the tridimensional shape of a protein complex is a difficult and expensive process to obtain experimentally. Adding that there are thousands of different proteins in a single organism, with different splicings and mutations, which can vary even further in different spices, it's barely impossible to determine them all experimentally. Thus, finding computational ways to obtain the structure of a protein is advisable to know the structure of variations of a protein. - -Here we present a computational tool to ensemble proteins and nucleic acids knowing their interactions. Currently, there are estimations that there are at least two hundred thousand different interaction between different chains of proteins that form every possible interaction in which a protein is present. Knowing those interactions, most of the complexes can be built easily knowing the stoichiometry of the protein. - -### Homology of the subunits - -In order to build the model, we start with several files: each file containis two chains (whether aminoacidical, DNA or RNA) representing an interaction between those chains. By looking to similar proteins in two different pdbs, we can overlap them in the space to check if two chains of those pdbs are similar. After that, we precisely use that similiarity to join them in a single model. This is known as protein superimposition, and the full process of joining is similar to a DNA sequence assembly: we have to recognise that two sequences are highly similar, and join them in the correct spot. - -In order to consider two chains in diferent pdbs the same protein and overlap them, they must be homologous. To check whether two proteins are or not are homologous, a pairwise alignment is made. This is a calculation of how similar two sequences are, that takes both mutations and gaps into account to give a numeric score between 0 and 1. A score of 1 means that both sequences are identical, while high values indicate how similar the sequences are. So, to consider them homologous, a high score must be achieved between two proteins. Only the homologous proteins will be overlaped and used to build the model. - -### Superimposition of the 3D structure - -Two homologous sequences will probably have similar structures. However, it's also possible that two proteins with lower homology score can have similar structures in space, as result of convergent evolution. Therefore, when two sequences are quite different but they may have similar structures, the may fullfill the same rol in our model and it's advisable to consider them: for example, two homologous proteins from distant species whose alignment doesn't pass our homologous threshold, but they still conserve the structure. - -To consider them in our model, we need another measure to check if they really have similar structures, the root median square deviation (RMSD now onwars). To check this value, first we need to superimpose those two proteins; that's putting the atoms of both proteins in the same place with the same orientation to check how good or bad they overlape in space; similar structures will have atoms in nearly the same position, while different structures will differ. The average distance (measured in argmstrongs) between the superimposed atoms is the RMSD. - -This way, RMSD must also be a filter we must also filter those structures that are differents even if we consider them homologous, as they can have very different 3D structures. - -### Energy levels - -Lastly, it's important to consider the final energy levels of the model. A good model of a complex should have minimum energy, as a functional, valid complex usually has the lowest energy among the possible foldings of the complex. Two atoms that are too close, hidrophobic residues in the external part of the model and several other considerations can increase the final energy of the complex, which must be corrected in a good model. - -Two atoms too close can also indicate that two proteins are not correctly joined. For example, if several atoms of a chain might be too close of another chain (or even *inside*), those proteins should not be joined, even if they interact. Taking this into account avoids impossible proteins, such as those which are inside other chains but breaking several laws of thermodynamics and physics. - -### Input - -The input is a folder with two or more files: each file contains two proteins, a protein and DNA/RNA strand or 2 single DNA strands (which may or may not form a double strand), which represent an interaction between those chains. Two structures are considered to be interacting if the minimum distance between them is in a range of a few Argmstrongs (1-10) . However, in lesser distances, forces between atoms are strong enough to produce changes between the chains and force them to adopt a different conformation, an thus, a realistic model must keep the residues at an adequate distance to consider it correct. - -### Output - -The output is mainly one pdb file with all the possible chains joined in the selected folder. However, if the optimized option was selected, several pdb files will be created in the same directory. These are different approaches made by modeller to optimize the energies of the model and the files it used. The model will be saved as final_model.pdb - -If the model is required with minimum energy, and thus the -optimize flag is used, then several files will be saved. Modeller will save some mid-step pdbs as 'final-model.DXXXXX.pdb' and the fully optimized will be saved as 'optimized.pdb' -