This is a minor improvement release.
Updates since v1.9.0
-
Documentation on installation is now up to date and is fixed (Issue #38) -- thanks to Anthony Nash for pointing out the obsolete instructions.
See the README file for updated installation instructions.
-
Github actions now cover both the new installation approach with conda and pip as well as the old school approach with compilation of OpenBabel
This is a much-needed major revision to remove bugs and squash problems related to the old-age of FragIt
Updates since v1.8.0
-
FragIt can now read fragment charges from an external file. This is specified in the input configuration files in
[fragmentation] chargemodel = read test.dat
where the MMFF94 keyword is replaced with read . Each line in the supplied file gives two values. The first value is the fragment index and the second value is the fragment charge. The following example shows that the charge of the first fragment is -1 and the fifth fragment has charge +1. Fragments not listed are assumed to be neutral.
1 -1
5 1
-
Includes some changes necessary for openbabel cooperation with especially version 3.
-
Fixed a few errors in the jmol script and template.
-
Switch to GitHub actions for build system testing.
-
Updated the GAMESS to HMO script to python3
Developer Changes
-
Made substantial changes to align FragIt with how python packages should be organized
-
code updates and improvements all around
This major update brings several quality of life improvements in FragIt.
Updates since v1.7.2
-
Now full (automated) testing of multiple python versions (2.7, 3.5 and 3.6 as we speak) coupled with multiple (master and 2.4.1 release) versions of openbabel
-
includes preliminary fixes for the upcoming openbabel 2.5 release (no idea when it is due) specifically regarding OBElementsTable and implicit valence info which was relocated in the openbabel API.
-
It is now possible to show the charges of fragments in PyMol. The command
ColorFragments charge
in the PyMol terminal will color the fragments with red and blue for negative and positive charges of the fragments, respectively.
-
FragIt now dumps it's input file in standard output so one can actually see the options being used. This is mainly to be inline with other programs and to make sure users understand what options were set.
-
FragIt now correctly warns people using explicit fragmentation points in combination with metal atoms that atoms are shifted around. This procedure has not been completely automated.
-
A lot of interal API documentation stuff happened.
This is a bug-fix and minor improvements release.
Updates since v1.7.1
- Mixing metals/ions with explicit bond cutting can (and is still) problematic. For now, print a warning with possible fixes so that the user not only sees a "bond not found" error. (#26) -- Thanks to Morten S Nørby for reporting this issue.
This is a bug-fix and minor improvements release.
Updates since v1.7.0
-
Fixes a python3 issue with the pymol scripts and template system.
-
PyMol scripts are now named .py instead of .pymol to signify that they are different from the .pml scripts. PyMol 1.8.6 enforces this rule.
-
the calculation of fragment labels has changed to a more efficient algorithm. Per default, fragment labels are hidden.
In addition, some behind the scene updates were made to the template system
This major update brings several changes to default keywords and to the command line interface for FragIt.
Updates since v1.6.1
-
FragIt is now compatible with python 3.5 after minor modifications to the source code mostly regarding imports and print statements.
-
Travis-CI builds are failing for python 3.5 for some compilation/install of openbabel problem so testing is done manually for now. This is also why it is compatible but not entirely supported for now.
-
Templates are now located appropriately in a 'share' directory in the installation directory
-
The PyMol template received some colorscheme upgrades for multilayer and FMO/FD (FDD) runs.
-
The GAMESS-FMO writer now supports the hybrid orbital projection (HOP) scheme for fragmentation as well as the AFO scheme that has always been present. HOP orbitals are located in the 'hmo' folder in the share directory. Limited to support of only SP3 C-C bonds but that should cover 99 % of all current use cases.
HOP is disabled per default. You can enable it by specifying
[fmo] dohop = True
in the configuration file.
-
The GAMESS-FMO writer now has improved support for multilayer calculations as well. Through a new section [qm] in the configuration it is possible to specify the basis set for an FMO calculation.
In the case of multilayer runs FragIt supports the multilayer notation from FMO as well so one specifies basis sets on a per layer basis separated by colons:
[qm] basis = STO-3G:3-21G
In multilayer calculations the appropriate basis set will be added to the $DATA group and leave out the $BASIS group.
For regular FMO runs just specifying basis sets as
[qm] basis = 6-31G(d)
will suffice.
-
The GAMESS-FMO writer now has support to generate FMO/EFP input files. The feature is enabled by
[fmo] efpwaters = 1
in the configuration file. This will, if no multilayer options are specified, gen- erate an FMO input file with all water molecules demoted from FMO fragments to EFP fragments.
This new option also supports a slightly more involved use case scenario where all water molecules some distance R from a central molecule, such as a ligand in a protein, are demoted to EFP water. The options are
[fmo] efpwaters = 1
[output] centralfragment = 1 boundaries = 3.0
which will convert all water molecules not within a distance R = 3.0 angstrom from fragment 1 to EFP water.
Please note that this will disable the multilayer calculation as that option seems a little exuberant.
This is a bug-fix and minor improvements release.
Updates since v1.6.0
- Fixes a default value in the base configuration for FMO input files such that atom names are ignored
This major update brings several changes to default keywords and to the command line interface for FragIt
Updates since v1.5.1
-
Removed --make-config option from FragIt commandline tool. Instead, a fragit-conf tool is supplied which allows for the generation of different types of configuration files for many types of fragmentation jobs.
-
Added fragit-conf tool to generate configuration files for FragIt. See fragit-conf -h for details.
-
Added multiple default configurations for FragIt. FragIt (through fragit-conf) now supports:
- fragment molecular orbital (FMO) specific fragmentation patterns.
- polarizable embedding (PE) specific fragmentation patterns. This is specifically made for use through the polarizable embedding assistant script (PEAS). The default configuration settings is still FMO. See fragit-conf -h.
-
Default values were changed for some parameters for the fragmentation group. The new values are:
- maxfragsize = 100
- useatomnames = False - ONLY FMO.
-
The setup script is now more self-contained.
This is a bug-fix and minor improvements release.
Updates since v1.5.0
- Fixed issue with FMO writer which dumped errornous keyword in FMO input file even if no bonds were broken. (#15) -- Thanks to Anders S. Christensen for reporting this issue.
This major update brings changes to defaults in keywords and a major overhaul to the information printed.
The changes to the defaults is mostly to make sure that users of the API get something sensible back per default without having to make too many changes before they get started.
Updates since v1.4.4
-
Default values were changed for some parameters for the QM/MM group. The new values are:
- includehbonddonors = False
- includehbondacceptors = False
- includecovalent = False
-
Default values were changed for a single parameter for the OUTPUT group. The new value is:
- useatomnames = True
-
A new option to enable printout (especially handy when using fragit as an API from other programs) in the OUTPUT group:
- verbose = False
when set to true, FragIt will print information on its doings.
-
FragIt can now correctly inform users if it fails because OpenBabel is missing. Previously this was not the case. Issue 13.
This is a bug-fix and minor improvements release.
Updates since v1.4.3
- Include within a distance in QM/MM.
- Adds atom names for added hydrogens.
- Correct naming of atoms added to cap.
- Cleanup in the code. Removing legacy space / tabs etc.
This is a bug-fix and minor improvements release.
Updates since v1.4.2
- Interfaced with Travis CI for continous integration.
- Build status on README.
- Moved tests out of the 'src' directory and reconfigured tests such that nosetests is now supported.
This is a bug-fix and minor improvements release.
Updates since v1.4.1
- FragIt now attemps to set the formal charges of metal and counter ions.
- Naming of atoms should be greatly improved.
This is a bug-fix and minor improvements release.
Updates since v1.4.0
- Fixed a bug which could make FragIt enter an infinite loop if the files used were only counter ions or metals
Updates since v1.3.7
- FragIt now correctly reads the configuration file early such that settings are used (instead of defaults).
This is a bug-fix and minor improvements release.
- Fixes an error where naming of atoms were not strings but None.
- Better naming of atoms and a proper warning when naming fails.
This is a bug-fix and minor improvements release.
- Fixes error when a QM region not connected with covalent bonds were extracted.
This is a bug-fix and minor improvements release.
- Fixes error in the way QM/MM regions were extracted.
This is a bug-fix and minor improvements release.
- Adds option to not include covalent neighbours to QM fragments. It is not always wanted to include covalently bound neighbours.
This is a bug-fix and minor improvements release.
-
P-H hydrogen bond distance added. This is used when using FragIt to cap DNA with the MFCC capping procedure.
-
Adds magnesium ions to ignore metals list.
-
Attempt to add residue names in a less fancy way. The old naming scheme gave "none" too often. Now, attempt to use info from PDB files if that exists
-
Fixed color space for fragments.
-
Expose QMMM features through config files. Options were added for inclusion based upon hydrogen bonds.
This is a bug-fix and minor improvements release.
-
Adds support for atomnames read from PDB files for instance.
-
Boolean options are read as strings. There was a problem with options not being parsed correctly.
This is a bug-fix and minor improvements release.
- Phosphorus atoms are not to be excluded as a metal atom. This concerns mostly DNA.
Updates since v1.2.1
-
stabilization of the API for 3rd party uses
-
API has support for a QM-region in the XYZ-MFCC writer. This QM-region can be expanded by including covalently bound or hydrogen bound neighbours.
This support is slated to be included more generally in a future fragit 1.4 release series.
-
A visually more pleasing color scheme. We are no longer living in an 8-bit world.
-
FragIt now exists more cleanly if it cannot find itself.
Updates since v1.2.0
- fixed naming scheme in xyz and xyz-mfcc writers to be similar
Updates since v1.0.4
-
MFCC style fragmentation is beginning to be work however, it will take some time before this matures
-
Slight extension to the API to allow for better interoperability with external software wishing to use FragIt.
-
Squash fragments
-
Found a reliable workflow - including bug fixes to treat metal ions. There are some constraints still.
Updates since v1.0.3
- specify charge models via the command line interface (--charge-model=) or configuration files.
Updates since v1.0.2
- Renamed GAMESS writer GAMESS-FMO due to ambiguity.
Updates since v1.0.1
- Most of the options in the command line version of FragIt now uses the underlying FragItConfig object for defaults.
-
The code is now distributed as a library
-
Rename option for distance based layering to --output-active-distance
-
Add option to disable protection patterns from command line
-
Add fragmentation of DNA
-
Add code and patterns to merge glycine to neighbor fragments. Other fragments can also be merged via the new mergepatterns group in the configuration files
-
Add possibility to generate configuration file without needing an input file. In the same instance, file-specific fragmentation info has been removed from the configuration files.
-
Add fix so custom patterns work consistently in all cases.
-
Add fix to make explicit pairs of atoms always work