Statistical analysis of computational docking of large compound data basesto distinct protein binding sites

The results of 16 docking simulations with rigid receptor sites and flexibl e ligands (similar to 60,000 compounds in each case) are statistically anal yzed and compared. Different combinations of binding sites, scoring functio ns, and compound collections are used in these calculations. The docking sc ores are not randomly distributed over the scoring range; they follow Gauss ian distributions (regardless of the binding sites), scoring functions, or screened compounds. If the docking sites are small, the Gaussian distributi ons are positively skewed. Peaks of the Gaussian distributions are populate d with compounds having similar scores but different sizes and binding mode s. These findings have implications for compound selection via computationa l docking. (C) 1999 John Wiley & Sons, Inc.