Experimental determination of protein structures, protein-protein complexes, is a highly time-consuming task. Molecular Docking provides a faster solution to identify protein-protein interaction sites. Docking is the key to rational drug design: it is used to find inhibitors for specific target proteins and thus to design new drugs.
The proposed framework is summarized in the block diagram below. The input is the PDB files of the interacting proteins, which is used to generate the Solvent Excluded Surface (SES). Then, a set of critical points is extracted from the surface. For each critical point, an Extended Surface Patch (ESP) is created, which spreads over a wider s
urface area around that point. Each ESP that corresponds to a convex (or concave) elementary patch of the receptor protein is matched w
ith all ESPs that correspond to concave (or convex) elementary patches of the ligand protein. The pairs of ESPs ranked as most complementary are given as input to the final step of the algorithm, where the candidate poses are scored, using a distance transform grid.
Experiments were performed on Docking Benchmark 2.0. It consists of 84 test cases: 63 Rigid-body, 13 Medium, and 8 Difficult. The dataset can be found here (http://zlab.umassmed.edu/zdock/benchmark.shtml).
A. Axenopoulos, P. Daras, G. Papadopoulos, E. Houstis, “SP-Dock: Protein-Protein Docking using Shape and Physicochemical Complementarity”, IEEE/ACM Transactions on Computational Biology and Bioinformatics, accepted for publication.
A. Axenopoulos, P. Daras, G. Papadopoulos, E. Houstis, “A Shape Descriptor for Fast Complementarity Matching in Molecular Docking”, IEEE/ACM Transactions on Computational Biology and Bioinformatics, DOI: 10.1109/TCBB.2011.72, 2011.
A. Axenopoulos, P. Daras, G. Papadopoulos, E. Houstis, “3D Protein-Protein Docking using Shape Complementarity and Fast Alignment”, IEEE int Conference on Image Processing, ICIP 2011, Sep 11-14, Brussels.