Docking of small ligand molecules in protein active sites is a very important and challenging problem in the structure-based drug design field. In this work we propose a Differential Evolution algorithm in conjunction with a multi-solution strategy for the flexible ligand docking problem. The proposed algorithm is evaluated on five highly flexible HIV-1 protease ligands, with known three-dimensional structures, having up to 19 conformational degrees of freedom. The docking results and comparison with classic Differential Evolution algorithm indicate that the incorporation of a multi-solution strategy in Differential Evolution algorithms is very promising and can significantly improve molecular docking accuracy. © 2012 Springer-Verlag.
CITATION STYLE
De Magalhães, C. S., Carlos, C. H., Almeida, D. M., & Dardenne, L. E. (2012). Improving differential evolution accuracy for flexible ligand docking using a multi-solution strategy. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7435 LNCS, pp. 688–698). https://doi.org/10.1007/978-3-642-32639-4_82
Mendeley helps you to discover research relevant for your work.