Virtual ligand screening by combined use of two grid-based docking methods, FLOG and ICM

COMP 86

Vladimir N. Maiorov, Molecular Systems, Merck Research Laboratories, Merck & Co., Inc, 126 E. Lincoln Ave., Rahway, NJ 07065 and Robert P. Sheridan, Molecular Systems, Merck Research Laboratories, Rahway, NJ 07065.
Flexible docking is a routine part of a modern structure-based lead discovery process. There are a variety of docking methods available to a modeler in a typical industrial environment to screen large corporate databases. How should these tools be optimally used to improve the selection of candidate molecules from the viewpoint of screening speed, software cost, and quality of the results? Many commercial docking programs are available, but the cost of the multiple licenses to do docking calculations simultaneously on multiple CPUs (software cost factor) and relatively long time required to get qualitative results (speed factor) do not allow one to use them for “virtual screening”. A combination of two grid-based docking methods, ‘fast-and-approximate’ in-house FLOG and ‘slow-and-accurate’ commercial ICM is presented as an example of the solution. Several hundreds of compounds ranked best by FLOG from a whole database are further carefully docked and re-ranked by ICM calculations. Validation tests with PDB protein structures and MDDR compounds are given.
 

Docking and Scoring
9:00 AM-12:20 PM, Tuesday, August 24, 2004 Pennsylvania Convention Center -- 109B, Oral

Division of Computers in Chemistry

The 228th ACS National Meeting, in Philadelphia, PA, August 22-26, 2004