[phenixbb] parallel structure comparison

[Nathaniel Echols]

On Thu, Apr 14, 2011 at 11:48 AM, Kendall Nettles <knettles at scripps.edu> wrote:
> In this weeks build, when I launch coot from the structure comparison results, no models or maps are loaded. It works in the last major release. Also, in the major release version the ligands are stripped out. Is this a bug? If not, I would like to request that the ligands be left in the structures.

I'm not sure what would be causing the problem with Coot - I'd need to
see the console output.  It appears to be working properly with a
previous result of mine, but that's not a guarantee that something
didn't go wrong while it was running.  Unfortunately I just found an
unrelated bug in map generation that was recently introduced and
completely crashes the program, so it's going to take more debugging
anyway.

The ligands should only be removed if you check "Superpose processed
structures with reference model", which pulls out individual chains
based on sequence alignment.  If you don't superpose the models, the
models loaded in Coot should be complete.  Eventually I'd like to have
an option keep ligands associated with whatever chain they appear to
be bound to, but this is less easy than it sounds, especially since
some structures like HIV protease share ligands between multiple
chains.

-Nat