Link to full page (citation export, more details):
Automated docking of highly flexible ligands by genetic algorithms: A critical assessment
Full Text PDF:
Journal: J Comput Chem
Year: 2004
Volume: 25
Issue: 3
Pages: 412-422
DOI: 10.1002/jcc.10384
Type of Publication: Journal Article
Keywords:
Algorithms; Binding Sites; Crystallography, X-Ray; Estrogen Receptor beta; HIV Protease Inhibitors; Humans; Ligands; Models, Molecular; Protein Binding; Protein Conformation; Receptors, Estrogen; Thrombin
Abstract:
An improved version of the fragment-based flexible ligand docking approach SEED-FFLD is tested on inhibitors of human immunodeficiency virus type 1 protease, human α-thrombin and the estrogen receptor β. The docking results indicate that it is possible to correctly reproduce the binding mode of inhibitors with more than ten rotatable bonds if the strain in their covalent geometry upon binding is not large. A high degree of convergence towards a unique binding mode in multiple runs of the genetic algorithm is proposed as a necessary condition for successful docking.