Thanks for fixing that Bernhard! I agree that maybe setting the B-factor based on adjacent residues might be a bit risky, so Eleanor's suggestion of setting it based on the average B for the molecule or chain sounds like a good interim solution. That value is unlikely to be wrong, at least after the first couple of cycles of refinement, and seems likely to be closer to the "true" B-factor than an arbitrary value.
It would also only need to be updated at the time the user first loads the molecule - probably no need to update on the fly as the average B won't change much On a completely unrelated note, as of the last couple of nightlies I have noticed that I am now unable to resize my Coot window (on Mac, OS X 10.10.3). Not a big deal, but if anyone has any suggestions on how to fix I would be most grateful. Cheers, Oliver.