Protocol

Computational Docking of Biomolecular Complexes with AutoDock

Adapted from Protein-Protein Interactions (eds. Golemis and Adams). CSHL Press, Cold Spring Harbor, NY, USA, 2005.

INTRODUCTION

Computer prediction of the interaction between enzymes and small molecules has now advanced to the point that it allows accurate prediction of bound conformations and binding constants. For instance, the program AutoDock allows consistent computational docking of flexible ligands with about a dozen torsional degrees of freedom, and the empirical free-energy force field provides predicted energies that are accurate to within ~2 kcal/mol, or an ~30-fold difference in binding constants. Thus, these methods can easily separate compounds with micromolar and nanomolar binding constants from those with millimolar binding constants, and can often rank molecules with finer differences in affinity. Computational docking methods can be used to screen a variety of possible compounds, searching for new compounds with specific binding properties or testing a range of modifications of an existing compound. The approach has been successful in numerous cases, most notably, the discovery of human immunodeficiency virus (HIV) protease inhibitors. This protocol presents a detailed outline and advice for use of AutoDock and its graphical interface, AutoDock Tools, to analyze biomolecular complexes using computational docking. The first step is to prepare the coordinate files for the docking molecule and the target molecule. The second step is the calculation of the affinity grid for the target molecule. In the third step, the docking molecule is docked with the affinity grid, and, finally, the results are analyzed.