3D enantioselective descriptors for ligand-based computer-aided drug design

Abstract

A novel three dimensional stereoselective descriptor for use in ligand-based computer-aided drug design (CADD) is presented in this thesis. CADD is an important tool for drug discovery and ligand-based CADD involves the computational analysis and representation of known active compounds for the prediction of novel active compounds. Several approaches are available for ligand-based CADD including quantitative structure-activity relationship (QSAR) which seeks to describe molecules numerically using descriptors. Stereochemistry is an important determinant of a molecule’s biological activity but presents a challenge for QSAR applications. Presented is the novel 3D-QSAR descriptor termed “EMAS” (Enantio-selective Molecular ASymmetry) that is capable of distinguishing between enantiomers and describing stereochemistry in a physically meaningful way. The descriptor aims to measure the deviation from an overall symmetric shape of the molecule. EMAS showed good predictability when tested with a dataset of thirty-one steroids commonly used to benchmark stereochemistry descriptors (r2=0.89, q2= 0.78). Additionally, EMAS improved enrichment of 4.38 versus 3.94 without EMAS in a simulated virtual high-throughput screening (vHTS) for inhibitors and substrates of cytochrome P450.