JW – Is PDB2PQR unambiguously good? Compared to PDBFixer/reduce/PROTOSS/etc?
JM – Uses ProPKA under the hood.
JW – That’s good, I’ve heard of that.
JW – Protein with heme?
JW – Will this run on colab? The notebook you have now is running on your beefy machine
JM – Should be fine on colab, the slow steps are CPU limited.
JM – Tried OpenFold, huge weights file that needs to be loaded, very involved. Most of the workshop complexity would end up being getting it installed/weights downloaded/inference to run.
JW – Would be good to have a way to help people understand WHEN our tools can be used. Like a green/yellow/red system, where we say “if your input has green characteristics, it’s fine. If it has yellow, then you may need to use other tools/open a text editor/cut out some things. If it’s red, then this isn’t suitable for our ecosystem.”
SMIRNOFF workshop
JM – Looking at some good candidate systems. Like CRISPR inhibitor (protein+NA+ligand).
Signups
JM – Draft form -
Dates?
Feb 20, 28, Mar 12 @ 3 PM Los Angeles/10 AM Canberra
Apr 10, 17, 24 @ 2 PM Berlin/1 PM London/8 AM New York
We will confirm times at our next check in
To dos
(high) Have draft notebooks for all 3 workshops that at least run for Jan 17
(high) Have google signup forms for workshops ready for Jan 17
(medium) Fix openff-docs CI/figure out what’s up with deployments
(medium) Fix openff-docs env cache issue
(low) Reach out to science team for additional vignette ideas
(low) Propagate PDB-bond loading options into nglview