How to solve 'Unknown bond_atomtype error'?

User discussions
1

GROMACS version: 2019.4
GROMACS modification: Yes

Dear GROMACS users,
I’ve simulated the protein-ligand complex system with charmm36-jul2017 force field.
I generated the ligand topology/parameter files from the cgenff and cgenff_charmm2gmx_py3_nx.py file. and there is no problem until I used the ‘gmx grompp’. Even the ‘gmx pdb2gmx’ didn’t raise error.

But when I issued the command line below:

gmx_mpi grompp -f MDP/ions.mdp -c COMPLEX/complex_solv.gro -p TOP/topol.top -o COMPLEX/ions.tpr

Fatal error came up

ERROR 1 [file lig.prm, line 12]:
Unknown bond_atomtype CG2R61

I attached my lig.prm file generated by cgenff_charmm2gmx_py3_nx.py

[ bondtypes ]
;      i        j  func           b0           kb

[ angletypes ]
;      i        j        k  func       theta0       ktheta          ub0          kub

[ dihedraltypes ]
;      i        j        k        l  func         phi0         kphi  mult
CG2R61    CG321    CG321    NG311     9     0.000000     0.167360     3
CG321    CG321    NG311    SG3O2     9     0.000000     0.418400     1
CG321    CG321    NG311    SG3O2     9     0.000000     2.928800     2
CG321    CG321    NG311    SG3O2     9     0.000000     0.418400     3
CG321    CG321    NG311     HGP1     9     0.000000     0.418400     3

[ dihedraltypes ]
; 'improper' dihedrals
;      i        j        k        l  func         phi0         kphi

Any ideas? Thank you for reply in advance

2

CG2R61 is a standard CGenFF atom type and is defined in ffnonbonded.itp. There should be no problem using this atom type unless you have edited the parent force field in some way or have otherwise not #included it in your topology.

Note that there is a newer (March 2019) version of the CHARMM36 port that includes parameter updates. That is not relevant to this particular issue (CG2R61 is in both versions), but FYI.

3

you’re right. The wrong force field name was wrote in my topology file. It’s all my fault. Thank you for your reply :)
And I did actually run MD simulation with CHARMM36 2019 force field, but when I used the g_mmpbsa tool with CHARMM36 2019, ‘Cannot allocate memory’ error rised. So I reduced the trajectory file size by using ‘gmx trjconv’ and it did not solve the ‘Cannot allocate memory’ error.
Therefore because I assumed the reason is that CHARMM36 2019 force filed is not compatitable with g_mmpbsa, I’m trying to run with CHARMM36 2017 force field

4

I doubt that will solve the problem. The force field versions are mostly the same, and there is no reason to think that there is some specific issue with the CHARMM force field and g_mmpbsa. That is not an officially supported program, so there is no way to know if it is properly updated and maintained.

5

Thank you for your quickly reply.
It is right. So I have tried to reinstall with another version of g_mmpbsa.

Thank you again