search

choderalab / perses

Experiments with expanded ensembles to explore chemical space
http://perses.readthedocs.io
MIT License
178 stars 50 forks source link

[CLI] no entry point for non-tip3p solvation model #1192

Closed IAlibay closed 1 year ago

IAlibay commented 1 year ago

As far as I can tell, there's no way to pass an entry that isn't 'tip3p' to _solvate_system in perses.app.relative_setup.py:

https://github.com/choderalab/perses/blob/8e3e423a388553fadd9d3f9c0342ffe3e3b4e4c0/perses/app/relative_setup.py#L815

This leads to a template matching ValueError when trying to use tip4pew (i.e. amber/tip4pew_standard.xml).

ijpulidos commented 1 year ago

It should be easy to pass this from the YAML file and pass this when we solvate the system. Maybe some solvate_water_model option in this file.

ijpulidos commented 1 year ago

The way to do this is adding yet another keyword argument to the RelativeFEPSetup object. Which we already know it's not an ideal way to do this, but I think it's easy enough for us to move forward with this and to keep in mind in the future that this is something we want to let users choose once we do the API refactor in the 0.11 release.