search

openforcefield / openff-benchmark

Comparison benchmarks between public force fields and Open Force Field Initiative force fields
MIT License
11 stars 2 forks source link

Make Boron-containing molecules fail out of workflow in coverage report #67

Closed j-wags closed 3 years ago

j-wags commented 3 years ago

Bill Swope reported that a boron-containing molecule was failing in the MM energy evaluation step. I don't think this should have made it through the coverage reporting step. This issue can be closed once the coverage reporting step filters out molecules containing Boron (or other missing parameters).

j-wags commented 3 years ago

@jthorton was unable to reproduce this issue. When Josh tried, the boron-containing molecule was filtered out by the coverage report step when it ran create_openmm_system. Josh notes that his test was for a single conformer, it's possible that a multi-conformer molecule would have a different behavior.

dotsdl commented 3 years ago

Ian Craig also reported that silicon-containing molecules made it through to optimization as well. I'm not sure if we expect this?

j-wags commented 3 years ago

Oof, nope. We shouldn't have parameters for silicon either (Cmd-F #14 in openff-1.3.0 returns 0 hits). So let's make sure our instructions for running the SMARTS filter will remove boron, silicon, and iodine-containing molecules.