Comments from Jane while debugging a hung MolProbity run:
It seems very likely that Mike's script for generating flipkins is flaking out on 2-character chainIDs. Can you tell if it actually did the flips OK and it was just the flipkins that failed?
Maybe we should add a test for 2-character chainIDs (and perhaps for upper-vs-lowercase chainIDs) on input, and either provide a way to disambiguate them internally, or (temporarily) turn off flipkin generation for cases that would cause problems.
Comments from Jane while debugging a hung MolProbity run:
It seems very likely that Mike's script for generating flipkins is flaking out on 2-character chainIDs. Can you tell if it actually did the flips OK and it was just the flipkins that failed?
Maybe we should add a test for 2-character chainIDs (and perhaps for upper-vs-lowercase chainIDs) on input, and either provide a way to disambiguate them internally, or (temporarily) turn off flipkin generation for cases that would cause problems.