Formation of specific protein-protein complexes is an essential requisite for most of biological processes, and constitutes one of the most relevant unsolved theoretical problems. Although several docking methods for abinitio prediction of protein-protein association have been reported, their systematic application at proteomic scale is still challenging.
Difficulties arise from both the definition of the scoring function and the complexity of conformational changes upon binding. An important step to solve this problem is the use of (1) docking simulations by Monte-Carlo techniques (pseudo-Brownian movements and Biased Probability minimization) and (2) soft potentials pre-calculated on a 3-D grid.
Based on this strategy, the candidate has developed, during his research work at The Scripps Research Institute (La Jolla, USA) and the University of Cambridge (UK), an accurate and efficient docking algorithm and several computational tools to predict protein-binding surfaces.
The application of these methods to large databases of structures (generated by structural genomics projects and by homology modeling), together with the integration of biophysical data and genomic information, will facilitate the structural characterization of protein-protein complexes at a proteomic scale.
The proposed project will have the following specific objectives:
- development and optimization of computational predictive methods for their high-throughput application to PDB structures and homology models (generated at different resolution levels);
- creation of a comparative database of protein-protein interactions in order to understand the structural and sequence factors that determine complex formation in proteins;
- biophysical and computational characterization of protein-protein interactions of biological and/or therapeutical interest; and (4) application of the developed computational methods to the design of selective inhibitors of protein-protein interactions.
Fields of science
Call for proposal
See other projects for this call