On Jul 20, 2011, at 2:09 PM, Sierk, Michael wrote:
Eric,
Thanks for the script! It works but I have a couple of issues still:
1) The second sequence in the alignment is not automatically being associated with the second structure. (It looks like the first sequence/structure association is done automatically, but not the second.) I can do it manually with Structure->Associations (with zero mismatches), and then the script works. Is there a way to create the second association via either the Chimera command line or using python? Is that association supposed to occur automatically as well?
AFAIK the second sequence should associate. If you send me the alignment and the structures (or tell me their PDB codes if they're standard PDB entries) I can investigate (send them off-list!).
2) What I am trying to do is load in two molecules and 3 different sequence alignments of the two molecules. In my Chimera script I make 2 copies of the 2nd molecule (using "combine") and color them differently. I would like to then do the Structure->Match process for each of the 3 alignments. Does the "findMAV" function in the mavMatch.py script return all instances of MAVs or just one? I know the titles so could I perform 3 different mav.match statements, one for each MAV instance?
Well, this is a more complicated scenario than what the script I wrote really handles -- since you have multiple MAV instances and only want certain open structures to associate with each instance despite the fact that those structures are identical. So it needs two scripts! Here's the approach: In your Chimera script: 1) Open the three alignments. 2) Open the first Python script (no-auto-assoc.py) to turn off auto- association in the MAV instances. 3) Open your two structures, "reference" structure first (more specifically, in a lower model number than the matching structures) 4) Make the two copies, or just open the second structure two more times. 5) Run the second Python script (assoc-and-match.py) to associate the reference structure with the first sequence in all the alignments, and one copy of the matching structure with the second sequence in each alignment, and then match them. --Eric