In this work we propose a hybrid ant colony optimisation algorithm as an alternative search engine in the GOLD protein-ligand docking framework [4]. The approach treats the placement of a ligand molecule in the protein's binding site as a discrete assignment problem and a geometric point fitting procedure generates protein-ligand complex conformations from this representation. As in PLANTS [5,6], we combine this approach with a local search in the continuous search space of the objective function. Continuous solutions are finally reassigned to approximate solutions of the discrete assignment problem resulting in a high-performing optimisation approach. We discuss certain aspects of the hybridisation strategy including the integration of heuristic information into the search process and compare the performance to the genetic algorithm currently used in GOLD. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Korb, O., & Cole, J. (2010). Ant colony optimisation for ligand docking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6234 LNCS, pp. 72–83). https://doi.org/10.1007/978-3-642-15461-4_7
Mendeley helps you to discover research relevant for your work.