Working group: "Computational Biology"

2011, the 29th of March

Linda Dib

Title: co-evolution in non-divergent or small sets of protein sequences.


Co-evolving residues in a protein structure correspond to groups of residues whose mutations have been implied simultaneously during evolution within different species, and this is due to several possible reasons involving the three-dimensional shape of the protein: functional interactions, conformational changes and folding dynamics.
Co-evolution signals have been detected on a few divergent protein families while families of conserved protein sequences remain untractable by current methods. A large scale investigation of residue networks can only be made with the development of refined methods treating conserved sequences as well. We propose a new combinatorial approach to overcome this difficulty. We highlight the existence of a structure that organizes networks of co-evolving residues, and extract, from this structure, important insights on the interaction between different parts of the protein during the folding process. BIS is the first attempt to highlight a structure among co-evolution signals and to extract functional insights from this structure. These results point out the complexity of the evolutionary process but also the possibility to detect the different interplay within signals coming from functional pressures.
BIS helps to highlight sectors within a protein structure, that is contiguous regions in the three- dimensional structure that appear to have co-evolved together. Sectors do not necessarily correspond to do- mains and identify new evolutionary units to study protein structures. We analyzed the MukB protein and highlighted two sectors envelopping the Walker- A motif.