[Chimera-users] showing buried pocket

[Elaine Meng]

> On Apr 17, 2008, at 8:59 AM, Gareth Young wrote:
> So I've been trying to make an 'inverted surface'. Basically I have  
> a cavity inside a protein which is almost completely surrounded and  
> so it is hard to illustrate it to others without removing parts of  
> the protein (which I could of course do). Does Chimera have a  
> function to mark the inside of a cavity with a surface, then  
> allowing the user to remove the protein and just leave the surface  
> so as to illustrate the size and shape of the cavity in relation to  
> the protein.

Hi Gareth,
Here is the man page on fetching CASTp data from Chimera:
http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/castp.html

Some of what I describe below is newish, so you would need to get a  
daily build of Chimera to follow along:
http://www.cgl.ucsf.edu/chimera/alpha-downloads.html

Although not all protein PDB  entries are in the CASTp database, you  
can upload a structure to the CASTp server directly (not from Chimera):
http://sts-fw.bioengr.uic.edu/castp/calculation.php
At least the last time I tried it, if you choose the "E-mail only"  
output option, one of the files it mails you contains only the atoms  
that form each pocket.  You could look at the text of that file, find  
the pocket of interest, and then only specify those atoms in the  
"surface" command in Chimera.  (It wouldn't work to just open a file  
with those atoms and surface them, because the surface would  
completely enclose the atoms, not just the "real" surface side.)   
This process is much more tedious than the direct fetch, however.

Even if you surface only the pocket atoms, there are other issues.   
One problem is that Chimera  only turns on/off surface patches per  
atom, and sometimes the same atom has surface on one side that is  
part of the pocket, but also surface on another side that is part of  
some different pocket or the protein exterior; when you show the  
surface for all "pocket atoms" there may be some extra bits.   
Example: use the command "open castp:2gbp" and in the resulting  
dialog, make sure the Surface option is turned on and click on the  
second pocket in the list.  It may take a few seconds to compute the  
surface. That pocket is completely enclosed, but some of the same  
atoms have additional surface.  Extra bits can be clipped away to  
some extent, but you are limited by the planarity of the clipping plane.

Another possibility in Chimera is to use the Surfnet tool.  I find it  
often makes too many small blobs and requires experimentation.   
Example: open 2gbp, delete ligand (to make the pocket empty), choose  
Tools... Surface/Binding Analysis... Surfnet - Selected Atoms.  Enter  
Midas Selection:
     :154 z<5
(I know residue 154 is next to the pocket and include other residues  
within 5 angstroms), set the color if you want to, click Apply.    
That gives one extra small blob.  Actually I tried 8 angstroms first  
and that gave several extra small blobs.  You can close (delete) the  
entire Surfnet results in the Model Panel (under Favorites).  You can  
Ctrl-click the Surfnet surface to select it.  You still can't control  
the different blobs independently until you split that surface,  
however (after selecting the surface, use command "ac Sc").  Then you  
can Ctrl-click an individual blob, open the Selection Inspector  
(Actions... Inspect), and change the blob's color, type of display  
(mesh/dot/filled) and lighting, and whether it is displayed at all -  
thus you can hide the extra blobs.

In fact, several separate programs have been developed to show pocket  
or tunnel surface or void volume:

Hole  (can't find a current web page)
Mole http://mole.chemi.muni.cz/
Caver  http://loschmidt.chemi.muni.cz/caver/
Hollow  http://hollow.sourceforge.net/    (you may enjoy the Munch  
painting in this page!)

I haven't had the activation energy to try any of these myself, and  
unfortunately none have been specifically adapted to work with  
Chimera (or vice versa).  Caver has a Pymol plugin.
I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
UCSF Computer Graphics Lab and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
                      http://www.cgl.ucsf.edu/home/meng/index.html




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20080418/2a2bd9dc/attachment.html>