General updates | JW – IA told me that he’s temporarily put the SFE (+interchange integration?) stuff on pause while they work on replacing RG. This makes sense to me, just wanted to make sure you and I are on the same page. MT – SFE and interchange integration are different things but highly intertwined, so I would assume they’re both on pause. I think there was a weird merge issue and some code that I remember writing isn’t there, leading to some errors. I think the remaining work is really straightforward but I don’t want to plunge ahead until IA has attention to give to it. JW – Awesome. Agree with your decisionj.
I still need to organize biopolymer people meeting - there’s probably more they could coordinate on (polymer loading tricks, residue nomenclature, slow/painful steps, etc). I’ll send out a scheduling poll with you as a “nice to have”. MT – Who are the scientists here? JW – Hannah T, Patrick F, Anika F, maybe Chapin.
MT – Interchange 0.3.27 is released: https://docs.openforcefield.org/projects/interchange/en/stable/releasehistory.html#id1 Re: The packmol thing - HT’s solution was a bit byzantine and went through OpenBabel. My code had a block where it tries RDKit and falls back to mdtraj, and I started considering falling back to openmm as a third option. And then I realized that all we need is the coords, so I just wrote a dumb parser using python+numpy to pull those. DSlochower had a bug that I fixed last month but hadn’t released, so I released it (interchange.from_openmm/combine performs a cutoff check, and this had an overlooked corner case when the openmm component came from certain pathways. Fixed now.)
Big MDTraj release is coming this week. Improves support for py3.12, some bugfixes. Will pull trigger Thursday/Monday. Would like to take June 26-July 5 as vacation
|