I am running an MD simulation on a protein-protein complex.
After seeing a similar question on research gate, I checked the amino acids rtp file in my force fields folder, and as expected from this error, the HD1 atom was not present in the HSE entry. The atom HD2 is however present in that entry. So I figured replacing the HD1 atoms in my PDB file with HD2 should solve the error.
And it did. For the time being.
To reaffirm, I made changes in Histidine's hydrogen atoms in the PDB file. When I went ahead with the energy minimization step, I got an error that said there's an Infinite Force on an atom. It turns out that the atom was "HD2" of some Histidine in the PDB file.
I saw online that the reason behind this error was due to atom overlap. Hence, just for seeing if that was the case for me, I changed the coordinates of that atom a little bit (this was just for checking, I can't do this for the actual work). When I ran the EM step again, I got the same error, but for a HD2 of a different Histidine molecule. So yes, overlapping of the atoms is the reason for this particular error. I cannot solve it by changing coordinates of all the HD2 atoms of the Histidines. So it all boils down to the main fatal error that I mentioned.
How do I approach this?
1. Changing the atom name (as in HD1 -> HD2 is not working due to the subsequent error)
2. I do not know if I should add the atom HD1 in the HSE entry in the rtp file (I tried this and got several warnings).
3. I cannot (or should I?) use -ignh because mine is not an NMR structure. I have modelled my proteins on Modeller and refined them online.
Any suggestions/solutions will help me a lot. Thank you in advance!
Hi, a crude measure is to use -ignh during pdb2gmx, it will rebuild all the H-atoms based on the force field you are using. Most of the time, it is a reasonable choice (though not always), as the H-atoms are mostly absent in crystallographic structure (as it is difficult to resolve h-atom positions).
Histidine is unique in the sense that its side chain offers multiple h-bonds at physiological pH. The better procedure is to check which heavy atom of His side chain is forming H-bond in the protein (either it is delta or epsilon), and rename your His residues accordingly (HIS, HIE, HID, please check your FF how these residues are named there).
"2. I do not know if I should add the atom HD1 in the HSE entry in the rtp file (I tried this and got several warnings)." > Try not to mess with rtp entry at this stage (as you are very new), and if you like to play around, just make a backup of FF directory and do as you like.
Maysaa Alakbaree Thank you so much! SwissPDB viewer solved the issue.
Ayaz Anwar Thank you very much for the information. Yes, I tried creating a duplicate folder and changed rtp entries for atoms that were giving me trouble, and it threw out several warnings and errors.
- Badges
- Science topic
- Similar topics
- Human-Computer Interaction
- Collaboration