| The problem of determining a relative transformation for a pair of proteins and their conformations that form a stable complex, reproducible in nature, is known as protein-protein docking. In this paper we describe an improved version of our FFT (Fast Fourier Transform) based rigid-body protein-protein docking algorithm F2Dock (F2 = Fast Fourier) which includes improved shape-complementarity and electrostatics functions as well as a new on-the-fly affinity function based on interface propensity. The current version uses uniform FFT, but exploits the sparsity of FFT grids for faster execution, and also restricts its search within a narrow band around the larger molecule. A proximity clustering phase then penalizes docking poses that are structurally very similar to poses with better scores. It also includes efficient on-the-fly filters based on Lennard-Jones potential, steric clashes, interface propensity, residue-residue contact preferences, antibody active sites, and glycine richness at the interface for enzymes. A solvation energy based reranking program called GB-rerank can also be utilized. Several F2Dock filters and GB-rerank use octree-based fast multipole type approximation schemes and can be tuned to obtain useful speed-accuracy tradeoffs, and have been implemented as multithreaded programs for faster execution on multicore machines. Our molecular visualization software TexMol serves as a front-end to F2Dock in a client-server mode of execution. We have calibrated F2Dock based on an extensive experimental study on a list of benchmark complexes and conclude that it works very well in practice. |