人工蜂コロニーアルゴリズムを用いた タンパク質―リガンドの結合予測と評価

抄録

Protein-ligand docking is one of the most significant issues in structure-based drug design (SBDD). Generally, this docking is considered as an optimization problem which specifies the energetically stable conformation of the ligand at the binding site. However, it is very difficult to identify the correct pose because of many optimization parameters with high correlation. In previous studies, it has been reported that popular docking programs can identify the correct docking pose with an accuracy of only about 60%. In this work, we attempted to apply the Artificial Bee Colony (ABC) algorithm to docking. ABC is an optimization algorithm based on the intelligent behavior of a honey bee swarm, which has higher global search ability than other algorithms. The performance of the ABC for docking was evaluated for 85 protein-ligand complexes of Astex diverse set using AutoDock Energy Function 4.2 as a scoring function. In comparison with three novel docking algorithms, namely SODOCK, PSO@AutoDock and AutoDock default (LGA; Lamarckian Genetic Algorithm), ABC provides the highest success rate of 72.9% (Table 1). The results reveal that the ABC might be more suitable for docking than others in particular for dealing with highly flexible ligands (Figure 1).