HJ – There’s an interaction between peptide N and alpha carbon that causes stereo to not get defined. This is causing parameter assignment to fail because the protein FF expects specific stereo. CBayly thinks that the FF should not have stereo marks.
JW – It’s possible that a protein consisting of mixed chiralities of AAs. But that’s extremely unlikely, so I’d be happy to remove the stereo marks from the FF.
HJ – There’s a paper where people looked at mixed-chirality proteins and found that it didn’t have a major influence on parameters.
JW – I’d leave that decision up to the science team, but from my eprsepctive I’m happy to remove the stereo marks from the FF to make it easier for users. How urgent is it for you to have this ina release? To make the release I’ll just edit the text file to not have any @ signs.
HJ – For now I’m working with a local copy so this isn’t a rush, but in the long run I’d like to be able to use an official elease.
JW – Ok, I’ll add this to the next release.
HJ – Update on interchange adoption / summary of meeting from Monday?
MT – Testing is on track. Everything seems like it will act identically except vsites (which were very buggy to begin with), and small changes in long-range electrostatics/vdw (but these changes are generally agreed to be fixes)
HJ – Is there a rough timeline for this?
JW – It’s tied to the 0.11 toolkit release. The timeline for the 0.11 release is to expect the alpha within a month (pip installing), and then the “full” conda package within 3 months.
HJ – When will I be able to use a “flask” containing solvent, ions, small molecules, proteins, and nucleic acids
JW – The hardest part of this will be nucleic acids, since those will NEED to be parameterized by an external tool.
HJ – I plan to get the nucleic acid parameters in OpenMM system format
MT – OpenMM importer could be expected roughly in Q3 2022 - Approximately september.
HJ – Will I need to update OpenMM version to new namespace?
MT – Yes, the 0.11 toolkit release will only work with openmm 7.6 or higher (so no simtk imports, only openmm imports)
HJ – Are there major changes I should be aware of in OMM 7.6 and above?
MT – No, the changelog was really small, it was almost entirely a import path change.
JW – I was just working on the 0.11 code the other day and I think I made good choices about sending residue info to/from OpenEye, but I’d really like your input once it’s available for you to test.