2021-04-14 Pulido/Wagner Check in

Contacting other developers for requirements

• IP messaged Hannah and Dom on the OpenFF Slack, turns out Hannah isn’t on perses any more (Jeff’s mistake).

  • IP will contact Dominic and Ivy on #perses channel on chodera lab slack, include JC if possible

• JW pinged MT on slack to ask him what would be the best way to gather requirements.

• Should we present to MoSDeF developers tomorrow? Let’s decide based on how proposed API looks.

API mock-up

• IP – Can we do custom errors in pydantic? Or are they all validationError? This will prevent more useful error types from propagating up through the stack.

• How to handle atom order in cases where atoms are added or removed? How would the cheminformatics molecule match up with atom-order expectations from PDB in the first place?

• If we mutate one residue into another, where do new coordinates come from?

  • JW – Let’s make this a plugin interface. Then we can make a really bad initial solution, and invite users to make a better one later.

• Let’s not build an atom seleciton language initially – Just make a to_mdtraj method and let people pass queries to there

•