CHARMM36m for DNA in GROMACS

[MKozic]

I have been losing hair over integrating CHARMM36m for DNA simulations in GROMACS. The incongruity of nomenclature by itself is an 8th circle of hell. Added to that, as far as I have seen from this thread from @awacha :

Non-integer charge in DNA using CHARMM36 force-field

GROMACS version: 2019.4 GROMACS modification: No Dear all, I’m trying to set up a DNA strand. I have named the 5’ and 3’ residues DG5 and DC3 in the pdb. I’ve used -ter option in pdb2gmx and specified the 5TER and 3TER endings, respectively. The process ends successfully, as seen in the enclosed text, however, the total charge of the chain is -29.140 e, which doesn’t look like a rounding error. Do you know what could be wrong with my input? The force field files are from charmm36-jul2021.ff.t…

Recent updates of CHARMM36m have lost “DT DA DC DG” .rpt entries. After writing a script for translating PDB nomenclature to CHARMM compatible nomenclature I have encountered the problem of missing O2’ atom, which I presume is because CHARMM36 is for RNA and CHARMM36m jul2022 version has lost said DNA descriptors “DT DA DC DG”.

I have ( I think) solved the problem by using february 2021 version that apparently still has those.

My questions then are:

1. Is it true that recent updates have lost those “pre installed” DEOX patches?
2. If so, why, and is there a specific reason?
3. Is there any way to apply DEOX patches in GROMACS?
4. Can I use that feb2021 version with those DT rtp entries?
5. Is just renaming entries (e.g. OP1 into O1P and OP2 into O2P and C7 of DT into whatever it shold go) enough and then I can jsut proceed with pdb2gmx and charmm36m into my simulations, or is there anything additional I should know/do?

Thanks in advance

[jalemkul]

@MKozic

• Is it true that recent updates have lost those “pre installed” DEOX patches?
• If so, why, and is there a specific reason?
• Is there any way to apply DEOX patches in GROMACS?

The DA, DC, DG, and DT residues are included in na.rtp in the force field port. I don’t know why you assert that they aren’t there.

@MKozic

Is just renaming entries (e.g. OP1 into O1P and OP2 into O2P and C7 of DT into whatever it shold go) enough and then I can jsut proceed with pdb2gmx and charmm36m into my simulations, or is there anything additional I should know/do?

C7 becomes C5M. In addition to the phosphate oxygens, this should be the only renaming required if the input coordinates are otherwise using standard nomenclature.

[MKozic]

Thank you for correcting me,

Kind regards,