[Chimera-users] findclash "batch" calculations

[Elaine Meng]

On Feb 4, 2011, at 2:51 AM, Julien wrote:

> Hi Elaine,
> I would like to find clashes between a ligand and a protein. I have around 100 000 poses of this ligands.
> The information to get the poses are at the moment saved in a pdb file. Indeed three pseudo atoms are used to place the ligand in the correct position using the command match.
> 
> The workflow would be the followin:
> open ligand
> open protein
> open pseudo.pdb
> 
> match ligand position1
> findclash
> match ligand position2
> ...
> 
> First, this is very slow, even in batch mode. Maybe you know a better way to do that.
> Second chimera does not open more than 9999 residues, the pseudo pdb have nearly 100 000 residues. Is there a way to get around that.
> 
> I guess the routine findclash calculates all distances brutally and compare them to a threshold.
> many greetings
> Julien

Hi Julien,
Findclash does calculate distances between atom centers, but the cutoff parameter is compared to "overlap" between the VDW spheres -- in other words, smaller atoms must be closer together than larger atoms to be considered overlapping or clashing.  The output information (optional) includes both the center-center distances and the overlaps, however.  You can limit the calculation to only certain atoms. There are also options to control cutoff and other settings.  Details are here:
<http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/findclash.html>

If you are really just using "findclash" without arguments, that would look at the whole structure and take more time than necessary.  For example, if your ligand is model #0 and the protein model #1, one way to limit the calculation to only clashes between them and send info to the Reply Log is: findclash #0 test #1 log true

Otherwise, I'm not an expert on any large-scale calculations or how you might optimize the procedure for large sets of atoms, so I'm CC-ing this question to the chimera-users list.  (It is generally better to send Chimera questions to this list rather than to me personally.)  The other users or developers may have good ideas.

Best,
Elaine
-----
Elaine C. Meng, Ph.D.                       
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco