Tribe-PSO: A novel global optimization algorithm and its application in molecular docking

One of the goals of researches on molecular docking is to develop robust algorithms with powerful ability in global optimization because the primary purpose of molecular docking is to find out the most stable conformation of receptor-ligand complexes. Inspired by the principles of Hierarchical Fair Competition, we propose Tribe-PSO, a multi-layered and multi-phased hybrid particle swarm optimization model, in this article. With benchmarks on several widely used multimodal test functions, Tribe-PSO proved to be a powerful global optimization algorithm. Tribe-PSO is then incorporated with the scoring functions in AutoDock 3.05 to perform molecular docking simulations of 100 receptor-ligand complexes from the Protein Data Bank. The simulation results show that the docking procedures using Tribe-PSO method generally find out the stable conformation as well as the native-like complexes more efficiently and accurately than GA-LS combination which is used in AutoDock 3.05 as the optimizer. All the tests demonstrate that Tribe-PSO is a promising search algorithm for molecular docking problems. (c) 2005 Elsevier B.V. All rights reserved.