CLI tools: expected behaviour when copying system.top for non-OPLS forcefields?

[a-ws-m]

Hi, I've been trying to use MDPOW to compute partition coefficients with the CHARMM forcefield. The first thing I've noticed is that the system.top file that is copied by mdpow-equilibrium (from here) has the oplsaa.ff directory and files hardcoded. Is this intended behaviour?

[a-ws-m]

I'm drafting a PR that improves the way non-OPLS force fields are handled and I think I've discovered that this isn't expected behaviour, see here. I'll include a fix for this in the PR.

[orbeckst]

Ok, thanks!

I am not checking this repo regularly so please feel free to ping me on a PR with @orbeckst .

[orbeckst]

@VOD555 you've been doing simulations with different FF.

Have we documentation on how we do this normally?

[VOD555]

This line will replace the path in the template topology file with the force field directory defined at here using the keyword forcefield.

Currently, we only support 3 specific force fields (they are in the documentation ). If one wants to use other force fields, it's better to run the equilibrium by himself/herself and then use the mdpow-fep or fep function to import the output of the equilibrium simulation and set the fep simulations.

[a-ws-m]

Actually I'm mistaken, sorry! I hadn't realised there was a forcefield option for the YML file, as it doesn't show up in the benzene.yml example. I'm still making that PR though as I've got some changes that should make using arbitrary forcefields easier.

[orbeckst]

Our documentation does need some work... I typically look at templates/runinput.yml to see what's there but I also didn't remember that there's a forcefield: key.

Do you want to mark your own comment above as an answer?