Follow-up: For future reference, I was able to answer my own question by coming up with the following procedure.
1) Edit the.fingerprint file generated by MCPB in the first step (-s 1) for the standard (or small) model to have the same residues as the fingerprint file for the large model. Also replace the standard model pdb with the large model pdb (not sure if this is needed).
2) Run step 2 (-s 2)
3) Revert the fingerprint file for the standard model back to the original version. (If this is not done, MCPB complains of mismatch residue names in the database when assigning charges.)
4) Run step 3 and continue as normal from hereon out.
This work-around leverages the fact that the bonded parameters and the charges are derived independently by MCPB!
Using this procedure, I got force constants and charges on the metal centers (I have two metal ions) that fit the guidelines given in the MCPB tutorial!
Best,
Matthew
> On Oct 23, 2022, at 6:26 PM, Matthew Guberman-Pfeffer <matthew.guberman-pfeffer.uconn.edu> wrote:
>
> Dear AMBER community,
>
> The “small model” built by MCPB for my system undergoes a rearrangement upon optimization that is not consistent with the crystal structure and does not happen if I fully optimize the "large model” built by MCPB.
>
> Is it possible to have the small model be the same as the large model for parameterization by MCPB? Or, to enlarge the small model to include the extra groups found in the large model?
>
> I’ve tried additional_resids, but this keyword only seems to add the side chains of the additional residues. In an optimization, those side chains will just fly away. Is there a way to include the backbone of the additional residues? (By additional, I’m thinking of including the preceding and proceeding residue to the one that is coordinated.)
>
> Best,
> Matthew
>
>
>
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 24 2022 - 08:30:02 PDT
