Regarding comm_grps in mdp files for Membrane protein simulation

User discussions
1

GROMACS version:2022.5
GROMACS modification: No

Hi, I am doing a membrane protein simulation based on the system prepared using CHARMM GUI. I was going through one of the equilibration.mdp files and came across a question. The mdp file has tc_grps = SOLU MEMB SOLV (SOLU (protein) and Membrane grouped separately) and comm_grps = SOLU_MEMB SOLV (Protein and Membrane grouped together). This is little confusing since for tc_grps also it should be SOLU_MEMB and SOLV just like the comm_grps. Infact the Membrane Protein: KALP15 in DPPC tutorial written by Dr. Lemkul also describes the same thing. Can some expert please help me with this question.

Thanks in advance

Best regards

2

Can some one answer this query please

Thanks in advance

3

Nowadays one can in nearly all cases uses a single group for both temperature coupling and center of mass removal. Multiple groups is a remnant for old times when integration was not sufficiently accurate.

4

Hi Dr. Hess,

Many thanks answering my question. So if I understand correctly, I can have tc_grps = SOLU_MEMB_SOLV and comm_grps = SOLU_MEMB_SOLV.

Also, I have a follow-up question. I already ran the simulations with coupling protein and membrane separately and water and ions separately as shown below:
tc_grps = SOLU_MEMB SOLV and comm_grps = SOLU_MEMB SOLV

Is this approach still accurate.

Thanks for answering my question

best regards

5

I suggest to use “system”, as then you are sure you are not excluding any atoms.