Throughput status | OpenFF Protein Capped 3-mer Backbones v1.0 Opts: 299557 → 310894 → 311229 TDs: 19 → 20 → 22 (General) – Should we throw more compute at this? DD – It looks like there are about as many jobs as there are workers, so I don’t think that throwing more compute at it will help. So I’ll just keep these going on Lilac.
OpenFF Protein Capped 1-mer Sidechains v1.2 OpenFF Protein Dipeptide 2-D TorsionDrive v2.1 RNA Trinucleotide Single Point Dataset v1.0 57299 (remaining 24362) DD – In the future, it’d be good to not use pickle files here. This PR had source PDBs (50MB) and also pickle files (50 MB), and the script for turning the PDB into mols isn’t documented. PB – Should I ask for another PR here to add the code for the PDB-to-pickle process? DD – That could be good, but make sure that he makes a new branch, since the old branch was squash-merged. Otherwise there will be a merge conflict. PB – Can we keep building off the same branch to remove the file size footprint? DD – No, since the squash commit is already in the main branch, there’s no easy way to lower the file size. I’d recommend not worrying too much about this PR, but let them know that in the future they should include the conversion code and avoid bulky pickle files (pickle isn’t a great long-term storage format, SDF may be better in the long run) PB –
|