Using Extended-Connectivity Fingerprints with Laplacian-Modified Bayesian Analysis in High-Throughput Screening Follow-Up

Abstract

This article describes the use of a combination of extended-connectivity fingerprints (ECFPs) and Laplacian-modified Bayesian analysis in a study of the inhibition of Escherichia coli dihydrofolate reductase. The McMaster High-Throughput Screening Lab at McMaster University proposed a competition to predict the hits in a separate test set of 50,000 compounds. Although the problem seemed best approached with 3D methods, the authors show that 2D methods offer surprisingly competitive results with a low computational cost.