<title>Author Summary</title> <p>As an integral part of drug development, high-throughput virtual screening is a widely used tool that could in principle significantly reduce the cost and time to discovery of new pharmaceuticals. In practice, virtual screening algorithms suffer from a number of limitations. The high sensitivity of all-atom ligand docking approaches to the quality of the target receptor structure restricts the selection of drug targets to those for which high-quality X-ray structures are available. Furthermore, the predicted binding affinity is typically strongly correlated with the molecular weight of the ligand, independent of whether or not it really binds. To address these significant problems, we developed FINDSITE<sup>LHM</sup>, a novel threading-based approach that employs structural information extracted from weakly related proteins to perform rapid ligand docking and ranking that is very much in the spirit of homology modeling of protein structures. Particularly for low-quality modeled receptor structures, FINDSITE<sup>LHM</sup> outperforms classical all-atom ligand docking approaches in terms of the accuracy of ligand binding pose prediction and requires considerably less CPU time. As an attractive alternative to classical molecular docking, FINDSITE<sup>LHM</sup> offers the possibility of rapid structure-based virtual screening at the proteome level to improve and speed up the discovery of new biopharmaceuticals.</p>
Search
Reviews
[Write a review of this article]
Find related articles from these CiteULike users
