Hi everyone,

Here's a short summary of the situation I wrote for the project status update. Briefly, it tells what has been done until now, why it has been done (no, it is not just for the fun of running huge calculations!) , and where we are going (hopefully!).

Once again I would like to thank all of you for taking part to this project.

Best regards,
Sophie

During the first phase of the project, cross-docking calculations for a
database of over 150 proteins were done with the help of WCG. That is, we
modeled every possible pair of proteins within the database in order to see
how they could interact together.

The data resulting from these calculation is now going to be analyzed in terms
of binding energy, protein-protein interface areas and interface residues to
see wether we can determine parameters which can help detect the experimental
partners (that is proteins which actually belong to the same protein-protein
complex in vivo) within the database. The protein interfaces that have been
generated during phase 1 will be analyzed to see if our computations can be
of use for the prediction of protein interaction sites (thus confirming
preliminary results that were obtained on a reduced dataset).

We will also use the results from the cross-docking calculations to test the
program (called JET) developed by the group of A. Carbone for the prediction
of protein binding sites from evolutionary information.

Should the results be satisfying (in terms of binding sites and binding partner
prediction) we will turn to the second phase of the project. In phase 2, we
will include JET in the cross-docking program. The information brought by JET
concerning the binding site of the protein will permit to make much more
efficient (and therefore much faster) docking calculations so that we will be
able to work on a larger protein dataset (up to 4000 proteins)

The long term goal of this project, is to develop a performing tool that
permits to detect within an extended protein database (such as the Protein
Data Bank, which comprises over 40 000 proteins) potential interaction
partners for specific proteins which are known to be involved in
neuromuscular diseases.