GROMACS version:
GROMACS modification: Yes/No
Here post your question Hello, I’m trying to run energy minimization by Gromacs for an artificial tripeptide, containing unnatural amino acid (beta homo amino acids). I followed the procedure for how to create new residues in the .rtp file, then I added the residues in residuetypes.dat, modified the .hdb file accordingly, and then moved the aminoacids.rtp into the working directory. I didn’t change the ffbonded.itp, nor ffnonbonded.itp. Still, I’m not able to generate the topology file via pdb2gmx, due to the following error:
Fatal error:
Residue type ‘SBL’ not found in residue topology database (SBL is the unnatural amino acid, b³hLeu)
However, when I browsed the forum, as far as I saw, no one has tried to add beta homo amino acids to the rtp file. The point is this kind of amino acid will change the backbone structure of the peptide. My question is would it be the reason that Gromacs is not recognizing this particular amino acid after all I have done?
The other question is how can I make sure that I have defined this residue in the rtp file correctly?!
To my knowledge there is no particular problem with this kind of modifications, provided the forcefield files are modified correctly. So most likely, the problem is with your modifications.
You should show what you added to aminoacid.rtp, residuetypes.dat, etc so that this forum can assist in a meaninful way.
Which force field do you want to use? GROMOS supports most of these out of the box, although probably the GROMACS port of the force field doesn’t contain these. For CHARMM, we’ve made an extension some years ago (Welcome to charmm36m-beta’s documentation! — charmm36m-beta 0.1.dev22+g938e83c.d20200108 documentation). Much of singly and doubly substituted beta-amino-acids are supported, but the hydrogen addition database is notably lacking. There was also a group who made an attempt with Amber, using antechamber (https://doi.org/10.1021/ci5003476). If you know the topology (including atom types, partial charges etc), there is a how-to in the manual (https://manual.gromacs.org/current/how-to/topology.html). I’d suggest to get an in-depth understanding of GROMACS topology and force field files: once one gets the hang of it, it is surprisingly simple and easy.
Thank you so much for your informative tips. I used your extension and the plugin for the pymol. One problem with pmlbeta was that each residue in the peptide was followed by TER in the pdb file, so the Gromacs only was able to recognize the first residue. However, by editing the pdb file (via editing the text) the issue was rectified. My question is it a common issue with the plugin or I just did something wrong?
I also have another question regarding the cyclic peptides (in the case of head-to-tail cyclization). How can I define the amide bond that is formed via the coupling of N- and C-terminus of the same peptide? I know that probably it is better to make a typic for this question, but I just wanted to shoot my shot here :)
Thank you again