[Chimera-users] ligand comformation clustering

[Ben Keshet]

Hi Maurice, if you use DOCK for the docking, you can cluster the 
conformations within DOCK 
(http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm). The default 
value is 2A, I think, but you can change that. I am not too familiar 
with scripting in Chimera, but I think that if you simply write all the 
commands you want to perform, save them as cmd file, you should be able 
to run it within Chimera 
(http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/indexcommand.html#cmdfile). 
You could also write a python code to run more systematically in Chimera 
(I am not familiar with that, see the Chimera Programmer's Guide). 
Otherwise, you can relatively easily write a simple code that 
systematically reads ligand files, calculates rmsd, and clusters based 
on given threshold (e.g. in Matlab, python of even excel).

Hope this helps,
Ben


On 5/5/2010 9:39 PM, Maurice Ho wrote:
> Hi there,
>
> How to cluster the different conformations of a ligand after a docking 
> experiment according to their relative orientations in the binding 
> pocket?  Any software can help to do so? I've tried the Ensemble 
> Cluster function in Chimera but it seems clustering the different 
> conformations according to the relative positions of the atoms of the 
> ligand itself.  I know that the command rmsd #0.1:UNK #0.2:UNK (UNK is 
> the ligand) could be a way to illustrate how distant in average 
> between different conformations.  How could I make use of this command 
> in script so that I could have all the possible comparisons and 
> cluster them in a sensible way?
>
> Best regards,
> Maurice Ho
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
>