July Status Update

Well, I missed May and June updates, but I’ll definitely keep sending out development updates.

Since the April update, there are some general community updates:

Upcoming Releases

We’re still in need of a 1.91 release soon. There have been over 335 commits since Dec. 2016 when 1.90 was released.

Roadmap

As a reminder, we’re trying to make progress on features for 2.0. A summary is available on GitHub Projects

In particular, I’m personally working to finish up the Python commands/workflows and auto-optimize force field support. This should be done “real soon now” and I’ll open discussion topics as they finish.

I’ve also hired a student who’s working to add efficient molecular surfaces (e.g., Van der Waals, Solvent Accessible, Solvent Excluded) and some of this is up as a pull request

Help Needed

I’d like to know if others are willing to help - I have entered a pile of possible tasks into GitHub - many are “good first issues” and can be tackled fairly easily.

More importantly, please feel free to submit suggestions for discussion on the forum or GitHub. The survey indicated over 80% of Avogadro users have never even submitted a bug report. (But were willing to fill out a 5-10 minute survey…)

The initial pull request for force field optimization is here:

I’ll be working to integrate Open Babel and (optionally) RDKit force fields. The current code includes a Lennard-Jones force field and several optimization methods:

  • Steepest descent
  • Conjugate gradients
  • L-BFGS

Right now, L-BFGS will be the default optimizer. A new contributor will work on the “FIRE” optimizer, based on concepts from molecular dynamics.

Extending the method, you just need to write an interface that sets up a molecule and calculates energies at a given geometry. Obviously if gradients are available, those will be retrieved too.