[Chimera-users] Is it possible?

[Elaine Meng]

Dear Ahir,
We're glad you like Chimera! Your question is reasonable but there  
isn't a simple answer.  First I'll explain how the superimposed  
structures are interpreted by Chimera, and then talk about the  
sequence alignment part.

How the structures are handled in Chimera depends on the format of the  
PDB file created by the superposition server:

(a) Some servers separate the structures with MODEL and ENDMDL lines,  
and this makes each one a separate model with separate listing in the  
Chimera Model Panel.  This is more convenient than the other  
possibilities.
(b) Some servers put the structures into one model but give them  
separate chain IDs, for example making one structure chain A and the  
other chain B.  You can still separately control them in Chimera, but  
not as easily as separate models.
(c) Some servers create incorrect PDB that have duplicate atom and  
residue numbers without using MODEL/ENDMDL or different chain IDs.   
These will display in Chimera, but it is very difficult to do anything  
else because you cannot uniquely specify the atoms or residues.

I list several superposition servers and discuss their outputs in this  
page:
<http://www.cgl.ucsf.edu/home/meng/grpmt/structalign.html>

If the server does not output a sequence alignment, it is not possible  
to back-calculate from the superposition that exact sequence  
alignment.  Why?  Because different sequence alignments (with  
different numbers of paired residues) can give essentially the same  
superposition, and we do not know the details of the fitting procedure  
used by the server:  whether alpha-carbons or all backbone atoms were  
fitted, iteration cutoffs that could weight certain positions in the  
alignment more heavily than others, etc.

That said, however, you can use the "Match->Align" tool in Chimera  
(under Tools... Structure Comparison) to create a sequence alignment  
consistent with your structural superposition.  It uses CA-CA  
distances and a user-specified cutoff for how far apart the residues  
can be in space, yet end up in the same column of the output sequence  
alignment.
<http://www.cgl.ucsf.edu/chimera/current/docs/ContributedSoftware/matchalign/matchalign.html 
 >

If the server does give a sequence alignment but not in a standard  
format, you could manually edit it (not pleasant, but often required  
in bioinformatics!).  Previously the  Mammoth server allowed  
downloading the sequence alignment in a convenient format, but I see  
it has moved to a different URL and behaves differently now.  You  
could try contacting the authors and see if they are willing to  
include the sequence alignment in the outputs.

Finally, you may want to try Chimera's superposition tool:  
"MatchMaker" (under Tools... Structure Comparison) or command  
"matchmaker":
<http://www.cgl.ucsf.edu/chimera/current/docs/ContributedSoftware/matchmaker/matchmaker.html 
 >

These Chimera tools are also used in the "Superpositions and  
Alignments" tutorial:
<http://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/tutorials/alignments.html 
 >

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



On Feb 4, 2009, at 2:55 AM, Ahir Utsav Pushpanath wrote:

> Hello Dear Sir/Madam,
>                   I am very fond of chimera, and everytime I have
> looked at scripts for doign things, I find that chimera has them  
> already
> inbuilt as features. There is one particular problem that I am sure
> chimera has the capability of doing but I am struggling to do it.
>
> 1) I want to use structural superposition servers such as Mammoth  
> etc to
> produce a structural alignment of two proteins, 2nadA and 1ybaB.
> However, some of these servers do not produce a sequence alignment
> output in a format compatible with chimera (i.e such as .aln or  
> aligned
> fasta). But they do produce a .pdb file showing the superposition.
>
> 2) If I load the .pdb onto chimera, sometimes in "model list"  
> window, i
> see the two proteins, wheras in others I only see 1 model(yet there is
> clearly 2 models there)
>
> 3) I was wondering whether you can load the superposition onto  
> chimera,
> and see the sequence alignment that gives rise to that structural
> superposition, so that I may save it as a .aln format using multialign
> viewer. How do I do this?
>
> Apologies if this is a very straightforward scenario.
>
> Thanks
> Ahir
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users