Hi AJ/Dan<br><br>Indeed the code we used in the lab for this is now public as Dan mentions. Initially it started as a python project and this is the code that Dan is talking about. Later we continued it as a pure java project and this is what you can find now at <a href="http://www.bioinformatics.org/groups/?group_id=1014">http://www.bioinformatics.org/groups/?group_id=1014</a><br>
<br>If you like java surely you are welcomed to give it a go and even contribute if you feel like it! The project is about analysing protein structures and integrating different bioinformatics tools. In particular we have a lot of code that deals with contact maps a.k.a. Residue Interaction Graphs. We have implemented a fast algorithm for the computation of the contact maps that you can use like this:<br>
<br> Pdb pdb = new PdbfilePdb("/path/to/mypdbfile.pdb");<br> RIGraph graph = pdb.getGraph("Ca",8);<br><br>That will get you a graph object that contains all the contacts for residues whose Ca are <8A apart. <br>
<br>If you don't really want to get into all that you can always use the genGraph command-line program that we provide. See a mini how-to I wrote on this list some time ago:<br><br><a href="http://www.bioinformatics.org/pipermail/owl-devel/2010-March/000005.html">http://www.bioinformatics.org/pipermail/owl-devel/2010-March/000005.html</a><br>
<br>With that you'll be able to get contact map files (simple text file with some headers and 2 columns representing the i residue number and j residue number) for contact types Ca or Cb and any distance cut-off.<br><br>
Hope it helps<br><br>Jose<br><br><br><br><div class="gmail_quote">On 9 April 2010 16:31, Dan Bolser <span dir="ltr"><<a href="mailto:dan.bolser@gmail.com">dan.bolser@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi AJ,<br>
<br>
I hope your PhD is going well! I bet Cambridge is nice at the moment;<br>
are the magnolia trees out yet?<br>
<br>
I'm glad that the paper was of interest to you.<br>
<br>
To compile the data for the analysis I did, I used the tools and<br>
databases developed by colleagues while I was in the Structural<br>
Proteomics group in the Otto-Warburg-Laboratory at the Max Planck<br>
Institute for Molecular Genetics. The good news is that the group has<br>
decided to make those tools publicly available via an open source<br>
project at Bioinformatics.Org:<br>
<br>
<a href="http://www.bioinformatics.org/groups/?group_id=1014" target="_blank">http://www.bioinformatics.org/groups/?group_id=1014</a><br>
<br>
<br>
I have CC'ed this reply the open source mailing list for that group,<br>
where you can ask any questions about what you will need to get up and<br>
running. I have included my processing pipeline notes, and some basic<br>
scripts for your reference (see 'sample_files.tar.gz').<br>
<br>
JOSE: my python scripts say "from pypdb import Pdb"?<br>
<br>
<br>
Unfortunately it is often very hard to find simple software<br>
encapsulations of empirical potentials for protein scoring. This was<br>
the reason why I couldn't rigorously compare the potentials I<br>
developed to others in the literature.<br>
<br>
<br>
GREAT NEWS! I was just collecting the set of files for you to look at<br>
when I discovered I *have* created a 'stand alone' Perl program for<br>
calculating scores! (Terrible memory!)<br>
<br>
See the attached 'score_any_structure-2010.04.09.tar.gz'<br>
<br>
<br>
Let me know how you get on.<br>
<br>
All the best,<br>
Dan.<br>
<br>
<br>
P.S. Will you ask Madan when he will get NCI back up? It's been down for years!<br>
<br>
<a href="http://www.mrc-lmb.cam.ac.uk/genomes/nci/" target="_blank">http://www.mrc-lmb.cam.ac.uk/genomes/nci/</a><br>
<br>
<br>
<br>
On 9 April 2010 14:26, A.J.Venkatakrishnan<br>
<<a href="mailto:venkatakrishnan.a.j@gmail.com">venkatakrishnan.a.j@gmail.com</a>> wrote:<br>
> Dear Dan,<br>
> Trust you are doing well. I am doing good, I started my PhD at the LMB under<br>
> Madan Babu and I am six months into it.<br>
> I recently came across your paper titled "Residue contact-count potentials<br>
> are as effective as residue-residue contact-type potentials for ranking<br>
> protein decoys" and it was a very interesting read. On a related note, I<br>
> would like to ask you for some advice.<br>
> I would like to compare the non covalent interaction energy between a given<br>
> pair of residues in different conformational states of a protein. I was<br>
> wondering if there is a software/program using which I could use to compute<br>
> residue-residue interaction potential given a protein structure.<br>
> Best,<br>
> AJ<br>
> --<br>
> --------------------------------------------<br>
> A.J.Venkatakrishnan<br>
> Ph.D. Student<br>
> St. Johns College<br>
> MRC Lab of Molecular Biology<br>
> University of Cambridge<br>
> --------------------------------------------<br>
><br>
<br>_______________________________________________<br>
Owl-devel mailing list<br>
<a href="mailto:Owl-devel@bioinformatics.org">Owl-devel@bioinformatics.org</a><br>
<a href="http://www.bioinformatics.org/mailman/listinfo/owl-devel" target="_blank">http://www.bioinformatics.org/mailman/listinfo/owl-devel</a><br>
<br></blockquote></div><br>