[Chimera-users] Request for advice on Protein Structure Alignment

[Elaine Meng]

On May 26, 2009, at 8:46 AM, Madhura Rale wrote:

> Dear Dr. Meng,
> I am a post-doctoral researcher working in the the Lab of Prof. W.  
> D. Fessner, Technical University of Darmstadt, Germany. We work in  
> the area of Biocatalysis and I am an organic chemist by training.
>
> We are currently looking for an online software / tool to compare  
> the active sites of two proteins whose structures are already known.  
> Our work focuses mainly on the utility of these protein catalysts  
> for chemoenzymatic syntheses.
>
> My query is that, if the structures of these two proteins has  
> already been determined by XRD, is it possible to align their active  
> site sequences to highlight the similarities of the substrate  
> binding sites? Would the tools mentioned on your website (list of  
> online resources for structure superposition) be useful for this  
> purpose and if so, would it be possible for you to advice on which  
> would be the best one and how one could go about it?
> Our need is to more or less obtain a pictorial comparison of the two  
> active sites to emphasise on the substrate specificites and  
> preferences of both the catalysts. I have no experience in such  
> determinations and I hope that my query is not too unappropriate.
>
> Thank you. Hoping to be excused for my ignorance and hoping to hear  
> from you soon,
> With warm regards,
> Madhura Rale.

Dear Dr. Rale,
It is a good question -- Unfortunately, there is no simple answer to  
what is the best method for comparing known structures or active sites  
-- it generally depends on the specific structures you are comparing!   
If the two proteins are rather closely related, it is an "easy" case  
and probably most of the methods will do a fine job.  If they are  
distantly related, it may be that some methods will work and others  
will not, and which programs are successful will be different for  
different input proteins.

It also depends what you already know, and what you are trying to  
prove.  If you already know that in structure A, residues 10, 12, and  
15 correspond to residues 48, 50, and 52 in structure B, then you  
would simply use any molecular modeling program to superimpose A and B  
using just those residues.  Or, if you already have a sequence  
alignment of the two proteins, some programs can use that alignment to  
superimpose the 3D structures.  I would just use Chimera to match  
specific residues, or use a sequence alignment, or even to guess the  
sequence alignment and then use that to superimpose the structures.  I  
work with Chimera and familiar with it, and know it can do these  
things.  Chimera is not a web server, but can be downloaded free for  
noncommercial use (Mac, Windows, Linux):
<http://www.cgl.ucsf.edu/chimera/index.html>

Discussion of how to superimpose structures in Chimera:
<http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/superposition.html>
Structure comparison tutorial:
<http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/tutorials/ 
squalene.html>
If you try using Chimera and run into difficulties, you can contact me  
for more specific help.

My "online resources" web page <http://www.cgl.ucsf.edu/home/meng/grpmt/structalign.html 
 > is separate from Chimera and is specifically limited to:
(a) Web servers and databases (there are other programs, e.g. Chimera,  
that you could download and then use on your own computer)
(b) comparisons based on overall structures (there is another large  
set of programs that instead focus on binding or active site patterns)

Thus if you are more interested in comparing the active sites and not  
so much the overall structure, a general molecular modeling program  
may be more useful than the servers in that page, but if your proteins  
are an "easy" case, probably any of the above would be fine.  The  
active site methods mentioned above are mostly for finding other  
proteins with a similar active site starting with your one structure  
of interest (not superimposing two proteins you already know about),  
so they are probably not very relevant to your situation.

I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
                     http://www.cgl.ucsf.edu/home/meng/index.html