Kyle Beauchamp

1. I agree that a huge issue with flat text approaches is the curation of metadata I agree that XML is nice, but I do think there is value in...

I think one way to achieve this might be to make Molecule a subclass or MixIn class for Trajectory. That way you can have something that is a Trajectory plus...

That would work too.

FWIW, I manually inspected the bed file in IGV (ftp://ftp.broadinstitute.org/pub/ExAC_release/release0.3.1/cnv/exac-final.autosome-1pct-sq60-qc-prot-coding.cnv.bed) and it seems to have data at NEB.

OK, thanks. So all platforms will automatically find water clusters by the triangular bond pattern and constraint them analytically, with no tolerance parameter? How do you feel about having this...

One other logical place to discuss this is here: http://docs.openmm.org/6.2.0/userguide/theory.html#integrators

Ping @peastman I can add this information to the docs once you decide which location you prefer

Either SGTM. Github also has an "Archive + read-only" feature in the settings page, so we could possibly add "DEPRECATED" to the name and then archive it.

Really, I'm not sure if that's part of the recommended pipeline for most applications anyway.

E.g. SCRE is not currently slated for porting to MSMB3