2024-12-18 Thompson/Wagner Check-in meeting notes

General updates

• JW

  • Nothing major to work on urgently, I expect some stuff will have broken over the holidays and be waiting for us in early Jan.

    • MT – Agree, I think there will be stuff to fix and that will be a good time to catch back up.

  • MT – Also, heads-up - re the c-f python_min stuff – I have things almost ready to go for our feedstocks, and could plausibly click the big green button, but decided to avoid possible deployment drama over the holidays. So I’ll handle this in Jan.

    • JW – Perfect, thanks

• MT

  • Anything major from POSE meeting yesterday that was important? I skimmed notes and didn’t see much.

    • JW – AWS action will have a major release, mostly to simplify things like specifying machine image ID+region.

    • JW – Also they were pushing the idea of having non-expiring GHA workflows, or AWS workflows that are inherently non-expiring. I didn’t think we’d find much use in this.

      • MT – This has happened to us before - 90 days is a somewhat low bar for inactivity.

    • JW – They were also looking at doing some work to explore+document paths from OpenFold prediction to OpenFF+OpenFE inputs.

      • MT – Does OpenFold not produce standard outputs that we can load? Shouldn’t this be free?

      • JW – I think they’d look into doing more of the structure prep - proton assignment, possibly ligand docking, solvation, etc.

    • JW – Also, met with OpenFE and OpenFF and chodera this morning to discuss if we could consolidate ask to new openmm ecosystem developer. Current thinking is that they won’t be interested in ommffs ever handling vsites or smirnoff proteins, so the ball is in OpenFE’s court right now to request support for combination of specific ommffs-derived components with SMIRNOFF components.

    •  

Trello

https://trello.com/b/dzvFZnv4/infrastructure