Speed

[julistanley]

The Chimera pipeline is damn slow.

Ideas for eventual optimization:

• Features at a radius of N Features at a radius of N-1. Keep track of lower radius features so that you don't have to compute as much.

[julistanley]

It's still not fast, but removing excessive printing and logging in 4a7c1c1 made the pipeline very noticeably faster (maybe ~30s for one protein instead of 2-3 minutes)