Issues with System building for RNA-ligand #1050
Comments
|
I you take a look at log.txt as the error tells you, you can see close to the end: Looking at your rtf file it's missing the masses of the atom types. Look for example at the Benzamidine rtf file which has masses for each atom type: |
|
After modifying the force field and topology with the specified changes, there were no complaints about the MASS atoms. The topology was defined using the following files. log.txt PSFGEN 2.0 from NAMD 2.14 for Linux-x86_64-multicore-CUDA psfgen) -------------------------------------------------------------------------- * psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31095: LONEPAIR COLINEAR LP CL CG DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31182: LONEPAIR COLINEAR LP CL5 C4 DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31259: LONEPAIR COLINEAR LP5 CL5 C5 DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31260: LONEPAIR COLINEAR LP6 CL6 C6 DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31294: LONEPAIR COLINEAR LP4 CL4 C4 DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31295: LONEPAIR COLINEAR LP6 CL6 C6 DIST 1.640 psfgen) ERROR! FAILED TO RECOGNIZE LONEPAIR. Line 31330: LONEPAIR COLINEAR LP3 CL3 C3 DIST 1.640 |
|
No sorry, there was no progress on this. We don't officially support CHARMM builds anymore. The license of CHARMM/psfgen is not very friendly for us and we generally use AMBER only. |
|
I used Amber to generate .frcmod files, but I encountered issues such as missing bond and angle parameters (attached below). However, the same file (.frcmod) works fine, if I used amber for simulations. It seems that HTMD is generating the *.frcmod files (from example files), but I don't know how to generate these files using htmd. Could you please direct me to some documentation on how to generate the lig.frcmod file from lig.mol2 using HTMD? Thanks for your help on this matter. Regards |
|
Hi, did you figure out this issue in the end? |
|
"While it is currently being investigated, I removed the ligand force field parameters and exclusively used the Amber force field for the RNA. However, I encountered difficulties with the Amber force field's default recognition." |
|
I have recently generated files using the parameterize (GUI) and resolved issues related to the ligand. I then set up an adaptive simulation on RNA-Ligand using the example code found in the following link: During the course of the simulation, I encountered two issues:
I kindly request your assistance in addressing these issues and helping me proceed with the simulation. Any guidance or suggestions on how to resolve these problems would be greatly appreciated. I attached two google drive link for the notebook and input files for the drug for your reference. If possible, could you please review the code. https://drive.google.com/file/d/1CFnx4GbBkp0U7-YXmX1cNLRzSAnkd0Wy/view?usp=share_link Thank you for your support. Error on terminal [E 14:15:01.158 NotebookApp] Uncaught exception in ZMQStream callback |
That's not necessarily an issue. You need to simulate very long sometimes to bind a ligand, depending on the
I have never seen this issue before but it does not seem related to HTMD. The traceback seems to be an error of jupyter itself so I cannot help you with that |
|
Thanks for your help. Could you offer advice on whether it is appropriate to use flat bottom potentials and establish the center of the box as outlined below for the RNA-drug complex that I am targeting ?. I was a bit uncertain due to the absence of an RNA-drug tutorials in the htmd tutorials. (I end up with asking so many questions here). # Apply a flat bottom potential to prevent the ligand from entering from periodic image of the protein (RNA)
width = np.array([70, 70, 60])
# Center the box at residue 218 which is on the upper side of the protein (RNA)
fbcentre = mol.get('coords', sel='protein and resid 218').mean(axis=0).squeeze()
flatbot = GroupRestraint('segname L and noh', width, [(5, '0ns')], fbcentre=fbcentre) |
|
we don't usually apply FB potentials to our simulations unless they are membrane simulations and we don't want the ligand to magically appear on the wrong side of the membrane by crossing into the next periodic image. I would just let the ligand diffuse freely |
|
Hello Sterdoerr, I was wondering if you could provide a tutorial for an RNA-ligand system to simulate binding and conduct analysis on an RNA-drug complex. Initially, I used a protein-drug tutorial as a starting point for the RNA-ligand system. However, I faced some challenges, like the drug not approaching the RNA as I had anticipated. I have attached the Jupyter Notebook files and drug files that I have worked on so far. If it's not too much trouble, could you please review these files and the code within them, and point out any errors you might find? Also, I would like to mention that we have additional data indicating that this drug is a nanomolar binder, as supported by our ITC and NMR data. Thank you in advance for your assistance." https://software.acellera.com/htmd/tutorials/system-building-protein-ligand.html |
|
If your drug is not binding to the RNA either: a) the parameters of your drug are wrong If you are worried about the building look at the log.txt for any warnings |
|
I'll close this issue due to inactivity. I assume the issue is solved |
I generated the LIG.rtf file and LIg.prm files for ligand using charmm-gui (https://charmm-gui.org/?doc=input/ligandrm).
I encounter the following errors while System building for RNA-ligand.
I attached the google folder of my files for your reference. Please advise, if possible.
https://drive.google.com/file/d/1ZYpC9kMbsw57EuiiLNYRj82DHrHAGvuH/view?usp=share_link
Thank you
The text was updated successfully, but these errors were encountered: