Dear ChemShell Developers,
I am trying to run hybrid QM/MM calculations with a polarizable MM region on a protein system, following the Tcl-ChemShell manual section here:
However, DL_POLY appears unable to read the CHARMM Drude protein force field. The issue seems to be that DL_POLY supports only up to three cosine terms for a periodic torsional potential, whereas the CHARMM Drude protein force field can contain six cosine terms for a single periodic torsion. For example:
OD30CL CD2O3B CD32C CD32A 0.218 1 180.00
OD30CL CD2O3B CD32C CD32A 0.303 2 180.00
OD30CL CD2O3B CD32C CD32A 1.647 3 0.00
OD30CL CD2O3B CD32C CD32A 0.983 4 180.00
OD30CL CD2O3B CD32C CD32A 3.430 5 180.00
OD30CL CD2O3B CD32C CD32A 0.720 6 180.00
When I truncate each periodic torsional potential to three of the six cosine terms, DL_POLY runs successfully.
Is this a known limitation of the DL_POLY interface in ChemShell? If so, is there a recommended workaround that preserves the full CHARMM Drude torsional potential?
As a related but separate issue, DL_POLY also appears to lack support for CMAP corrections, which prevents that part of the CHARMM force field, Drude or additive, from being used during simulations or optimizations. I would also appreciate any guidance on whether there is a workaround for this limitation.
Best,
Daniel