Characteritzation of protein-protein interfaces and identification of transient cavities for its modulation

Protein-protein interactions (PPIs) play an essential role in many biological processes, including disease conditions. Strategies to modulate PPIs with small molecules have therefore attracted increasing interest over the last few years, where successful PPI inhibitors have been reported into transient cavities from previously flat PPIfs. Recent studies emphasize on hot-spots (those residues contribute for most of the energy of binding) as promising targets for the modulation of PPI. PyDock is the only computational method that uses docking to predict PPIfs and hot-spots (HS) residues. Using Normalized Interface Propensity (NIP) values derived from rigid-body protein docking simulation, we are able to predict the PPIfs and HS residues without any prior structural knowledge of the complex. We benchmarked the protocol in a small set of protein-protein complexes for which both structural data and PPI inhibitors are known. We present an approach aimed at identifying HS and transient pockets from predicted PPIfs in order to find potential small molecules capable of modulating PPIs. The method uses pyDock to identify PPIfs and HS and molecular dynamics (MD) techniques to describe the possible fluctuations of the interacting proteins in order to suggest transient pockets. Afterwards, we evaluated the validity of predicted HS and pockets for in silico drug design by using ligand docking. We present a strategy based on MD and NIP which allows to identify cavities as potentially good targets to bind inhibitors when there is no information at all about the protein-protein complex structure