The locked portion of the protein seems erroneously arranged: there are sheets that seem odd in appearance and sections of loops with 90 degree angles and completely straight sections crossing from one side of the protein and connecting to the other while overlapping other loops.
Additionally attempting to rotate a sheet segment will rotate the entire loop as if it is a solid part of the sheet, disconnecting the cut end from the locked portion.
The antibody protein (the locked region that cradles the CSP loop) has been trimmed in this puzzle, which causes this region to look funny.
The full antibody protein is much larger, but in this puzzle we only need to model the portion close to the CSP loop. By trimmed a few hundred residues from the antibody protein, this puzzle should be run a little easier and be more responsive. Rest assured, those weird segments that seem to crisscross the frozen region are just visual artifacts. This does not represent the actual protein structure and should not affect scoring.
Your second comment is a little concerning. Do you mean that the locked CSP loop moves? Can you share your solution using the Share with Scientists feature?
crashing when running mutate scripts
That's not good. Any chance you could post the log.txt file from a crashed client?
Is the crash easily reproducible? Can you describe exact steps to reproduce so we can try to track down the bug on our end?
1507b is acting funky; sometimes it is hard to select
It seems to be a problem when you have the recipe output window open
Definitely the Show Output box is acting funky.
I am on MAC and using devprev