Building topology for medium-sized non-biological molecules

User discussions
1

GROMACS version: 2021
GROMACS modification: No
Dear community!

I’m trying to model polymer host-guest complexes, which consists of polymer chains (1000+ atoms) and small (~200 atoms) guest molecules. In the literature I’ve found that OPLS or GAFF forcefields are used for that task.

At the moment I’m stuck with building topology for polymer chains. I’ve prepared it’s geometry as .xyz and .pdb files. I’m trying to use acpype to produce .itp files. The only problem is it takes forever to build topology for not that big molecule (~350 atoms) and doesn’t work with OPLS forcefield.

So, what’s the proper workflow and tools for polymer systems regarding topology building?

2

You should not attempt to parametrize the polymer as a whole, as you will likely get buildup of partial charges that are geometry-dependent. The appropriate approach is to define model compounds that mimic monomeric units and any additional molecules that encompass all relevant linkages so you can generate appropriate bonded terms between the internal monomers, end units, etc. Then you should write .rtp entries for each and produce the topology via pdb2gmx.