Weeks-Chandler-Anderson Potential

GROMACS version: 2020.2
GROMACS modification: No
Here post your question

Dear GROMACS users,

Can you please tell me if it is possible to use the non-tabular potential of the Weeks-Chandler-Anderson Potential in last versions of Gromacs?

Best regards,
Yel21

HI,
if you have only one sigma you can use L-J potential with an ad-hoc defined cut-off.
If you have more sigma values than you have to use tabulated potential. Tabulated potential are not implemented in 2020, in place use 2019).
Best regards
Alessandra

Dear Alessandra,

thank you for your reply.

I have only one type of beads in my simulation. But, however, I do not clear understand how to use shift function in Gromacs, that suggested in eq. (21) http://manual.gromacs.org/documentation/2019/reference-manual/functions/nonbonded-interactions.html.

Could you please explain more precisely, what did you mean when wrote “ad-hoc defined cut-off”? How can shift and cut L-J potential using only mdp file parameters to obtain WCA potential?

I wanted to implement in Gromacs the Kremer – Grest model for polymer modeling.

Best regards,
Yel21

Hi,

My suggestion was to set in mdp the following option
vdwtype = cutoff
rvdw = [2^1/6 * (your sigma value)]

while the LJ potential parameter has to set in the topology file

For the bond interaction in Kremer - Grest model, you can use FENE bond stretching potential (that is implemented in GROMACS)

Best regards
Alessandra

Dear Alessandra,

Thank you for your suggestion.

I tried to find in gromacs 2020.2 manual any restrictions of tabulated potentials and did not find any words about this. Could you please point out me where was written that gromacs 2020.2 does not support tabulated potentials?

Best regards,
Yel21

Hi,
You are right from the manual, it is not clear, but the mdp option for ‘user-specified potential functions’ is currently unsupported in 2020,
as you can see in http://manual.gromacs.org/documentation/current/user-guide/mdp-options.html. It may come back in 2021, that was the reason why I suggest to use 2019.
Best regards
Alessandra