An Efficient Approach for Flexible Docking Base on Particle Swarm Optimization

Abstract

Molecular docking is an important tool in identifying potential drug candidates. The molecular dock- ing problem is to find a good conformation for docking ligand to a large receptor molecule. It can be formulated as a parameter optimization problem consisting of a scoring function and a global optimization method. Based on a variant of Particle Swarm Optimization (PSO) named Fully Informed Particle Swarm (FIPS) and the semiempirical free energy force field in AutoDock 4.0, a new approach to flexible docking method called FIPSDock was implemented. The search ability and docking accuracy of FIPSDock were evaluated by multiple redocking experiments, and the results of which demonstrate that FIPS is more suitable than Lamarckian Genetic Algorithm (LGA) for the force field of AutoDock. FIPSDock is superior to AutoDock and SODOCK which was also proposed by improving AutoDock with PSO in term of obtaining a lower binding energy, a better docked conformation, convergence speed and robustness. Compared with the four currently widely used methods-GOLD, DOCK, FlexX and AutoDock, FIPS- Dock is more accurate. Thus, FIPSDock is an efficient and accurate docking method and its promising prospects can be expected in the application to virtual screening.