Theory of docking scores and its application to a customizable scoring function

Abstract

In general, the docking scoring tends to have a size dependence related to the ranking of compounds. In this paper, we describe a novel method of parameter optimization for docking scores which reduce the size dependence and can efficiently discriminate active compounds from chemical databases. This method is based on a simplified theoretical model of docking scores which enables us to utilize large amounts of data of known active and inactive compounds for a particular target without requiring large computational resources or a complicated procedure. This method is useful for making scoring functions for the identification of novel scaffolds using the knowledge of active compounds for a particular target or a customized scoring function for an interesting family of drug targets.

Keywords:

• docking score
• virtual screening
• protein–ligand interaction
• Z-score

Acknowledgments

This work was supported by the National Institute of Biomedical Innovation and the Drug Discovery Program of the Pharma Valley Center.