[Chimera-users] selection (from distance) among PDB files after docking

[Elaine Meng]

Hi Damien,
First consider which commands you would use on each file and try them out in the command line on an example file.  If you only wanted CA-CA distances, each residue pair would require one command, something like

dist :246.b at ca :131.c at ca
<http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/distance.html>

That automatically sends the distance to the Reply Log.  However, it sounds like you might want to measure all pairwise distances from all atoms in one residue to all atoms in the other residue.   It could be done with a lot of distance commands naming the atoms, but one way to do it in only a single Chimera command is to stretch the function of "findclash" and force it to report all pairwise distances, something like

findclash :83.a test :100.b overlap -100 hb 0 log true
<http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/findclash.html>

The large negative overlap means to just consider everything a contact and report it even if large as 100 angstroms, hb 0 means no special treatment of H-bonds, and log true means to send all the pairwise distance information to the Reply Log.  This will be several lines of output.  To do something fancier and only output the smallest distance would require Python, which is beyond my skills.  

So let's say that for each structure, you chose to run a few "findclash" commands, one for each pair of residues.  Next consider that you will need to loop through structures.  Here is general information on how to use Python to loop through structures:  
<http://www.cgl.ucsf.edu/chimera/docs/ProgrammersGuide/basicPrimer.html>

That example also includes putting a "processing filename" message in the Reply Log.  So if you followed that general procedure, the Reply Log would have

processing file1
[a bunch of lines of findclash output including distances]

processing file2
[a bunch of lines of findclash output including distances]

... etc.

That would still require some "manual" looking through the Reply Log.  I hope this helps,
Elaine
----------
Elaine C. Meng, Ph.D. 
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco

On Jan 23, 2011, at 10:48 AM, Damien Larivière wrote:

> Dear all,
> I have numerous PDB files resulting from a docking run.
> 
> From crosslinking experiments, we know here that one way to select 
> among all the solutions is to identify the PDB files for which the 
> distance between some residues of interest is lower than a specific value:
> 
> distance between Res 246 in chain B and Res 131 chain C < 17 A
> distance between Res 162 in chain B and Res 135 chain C < 24 A
> 
> The two constraints must be respected.
> 
> So, is it possible to write a script which:
> 
> - opens one by one the pdb files,
> 
> -calculates the two distances for the current pdb file,
> 
> - send in the reply log the name-number of this PDB file if it respects 
> the two constraints?
> 
> Then, I would just have a look to the reply log in order to know the pdb 
> files to study in depth.
> 
> Many thanks for your help
> Damien