The biological activities of proteins depend on specific molecular rec
ognition and binding. Computational methods for predicting binding mod
es can facilitate the discovery and design of ligands and yield inform
ation on the factors governing complementarity. The DOCK suite of prog
rams has been applied to several systems; here, the degree of orientat
ional sampling required to reproduce and identify known binding modes,
with and without rigid-body energy minimization, is investigated for
four complexes. There is a tradeoff between sampling and minimization.
The known binding modes can be identified with intensive sampling alo
ne (10,000 to 20,000 orientations generated per system) or with modera
te sampling combined with minimization. Optimization improves energies
significantly, particularly when steric clashes are present, and brin
gs many orientations closer to the experimentally observed position. W
hether or not minimization is performed, however, sampling must be suf
ficient to find at least one structure in the vicinity of the presumed
true binding mode. Hybrid approaches combining docking and minimizati
on are promising and will become more viable with the use of faster al
gorithms and the judicious selection of fewer orientations for minimiz
ation. (C) 1993 Wiley-Liss, Inc.