The software aims to provide a solution to screening large compound libraries, which remains an expensive and time-consuming task. Computational high-throughput screening can enrich the fraction of suitable compounds in a screening collection, thereby reducing the cost of biological testing in lead discovery.
Surflex-Dock offers benefits in virtual high-throughput screening. It uses an updated and re-parameterised empirical scoring function (based on the Hammerhead docking system) with additional negative training data and a search engine that relies on a surface-based molecular similarity method.
"The software combines a scoring function with a patented search engine. The combination has been shown to yield excellent results in terms of docking accuracy, and distinctively superior results in terms of screening enrichment," said Ajay Jain, a faculty member at the University of California San Francisco Cancer Research Institute and also developer of the Surflex-Dock program.
Two aspects determine the quality of a docking method: docking accuracy and screening enrichment.
Docking accuracy measures the likelihood that a method will correctly identify and recognise the true binding mode of a ligand bound to a target protein.
Screening enrichment measures the relative improvement in the identification of true binding ligands using a docking method versus random screening.
Surflex-Dock addresses the protein ligand docking problem, demonstrating a novel approach that is faster, more accurate, and is five to ten-fold more specific than competing methods in detecting true binding ligands from non-binding ligands in computational screening applications. The docking search engine makes use of a patented technology.