2023-08-22 OpenFF QCA Working Group Meeting Notes

JW : openff-qcsubmit compatibility with QCFractal next

• working on openff-qcsubmit today; having trouble with pip-installing QCFractal

• pip install qcfractal[services]

• BP – if you’re installing from within git repo, then possible you have stale directories that are confusing pip; empty dirs should be cleared

• openff-qcsubmit 0.50 is out

  • fixes broken retrieval of Industry Benchmark sets

  • also gives compability openff-toolkit 0.14.1 and up

BP : updates from MolSSI

• BP – Not much, might just move next branch to main, but worried it’ll break a lot of workflows.

• DD – Since the old default branch was master , it shouldn’t be super disruptive to move next to main

• JW – Could also be nice to make a slack+twitter announcement

  • (General) – It’d also be good to put it in the readme,

DD – Anything more we can provide for site review?

• BP – Could use higher res version of SPICE logo (actually OPenMM logo)

• PE – I’ll send a high-res version of OpenMM logo to BP

CI – Will there be a hdf5 view option for the new version? Or more persistent storage locally?

• BP – Absolutely. No hdf5 now, but looking to do some local sqlite. This would be in the “views” classes in the codebases.

• CI – Yeah, with a poor internet connection, I would really like local storage.

• CI – Will there be a way to get a static copy of some big chunk of data? Would be useful for storage on zenodo.

• BP – There are some possibilities for doing this using views later on.

• CI – Great - I’ll articulate thoughts/needs in Zenodo.
  
  

New datasets

• qca-dataset-submission:

  • OpenFF Optimization Diverse Fragments with Iodine (w/ ESPs)

    • LW – Have SMILES that can be made ready to go quickly

    • LW – general guidelines as to dataset construction – one big dataset vs. many small?

      • LW – I have a large combined set of diverse fragments now – would it be useful to split it out into separate I, Br, S datasets?

      • DD – I don’t know what performance looks like for new vs. old QCF. We had problems in the past where things got slow around 100k mols.

      • BP – Should be a lot better than that now, we won’t know until we hit the limit.

      • DD – So I’d recommend crafting thesedatasets along the lines that make sense to you, LW, and then we’ll see which bottlenecks we hit.

    • PB – How bigwill this dataset be?

      • LW – Haven’t deduplicated ESP datasets yet. We’re starying from 200k total with 1.3k iodine, 5k with bromine, 15k? with sulfur. Would be helpful to have them split out by these groups.

      • DD – Do you feel ready to start the PR to qca-dataset-submission?

      • LW – I was thinking this could be good for Brent or Lexie

      • DD – Ok, they can look to previous submissions or guidance, and PB and I will be available for questions on slack.

  • OpenFF Optimization Diverse Fragments with Bromine (w/ ESPs)

  • OpenFF Optimization Hypervalent Sulfurs (w/ ESPs)

  • OpenFF DNA

  • OpenFF Protein Fitting (Chapin Cavender)

• spice-dataset:

  • SPICE v1.3 - will be prepared by @David Dotson ; will attempt to perform without openff-qcsubmit using find_existing kwargs with new PortalClient

  • SPICE v2.0 - in preparation:

    • PE – Waiting on manager setup instructions

    • DD – Right, we also need to get the managers running, but we also need to get datasets submitted.

  • qca-dataset-submission-like automation on SPICE repo for handling execution?

MolSSI QCArchive user group

• user questions / issues / feature request

• server instance statuses

  • QCArchive Legacy

    • v0.15.8.1

  • QCArchive OpenFF

    • v0.50.0b13

    • currently retains everything from Legacy

    • will lose ML datasets at some point

  • QCArchive ML

    • v0.50.0b13

    • mostly only contains ML datasets

  • QCArchive Validation

    • v0.50.0b13

    • MolSSI internal projects

  • QCArchive Demo

    • test instance

• compute resources statuses

• call for new users

Resourcing updates from stakeholders

QCFractal development

• QCFractal v0.50.0 - imminent

  • milestone for completion?

  • milestone for follow-up release?

Additional business

• MolSSI QCArchive Working Group start date: 8/29

• (General) – …

• BP – Will be quite busy next week, I’ll start this Sept 5.