pkCSM: Predicting Small-Molecule Pharmacokinetic and Toxicity Properties Using Graph-Based Signatures

TLDR

Drug development suffers high attrition due to poor pharmacokinetic and safety properties. The study aims to use computational methods to reduce drug development risks. The authors developed pkCSM, a graph‑based signature model and freely available web server, to predict central ADMET properties. pkCSM matches or exceeds the performance of existing ADMET prediction methods.

Abstract

Drug development has a high attrition rate, with poor pharmacokinetic and safety properties a significant hurdle. Computational approaches may help minimize these risks. We have developed a novel approach (pkCSM) which uses graph-based signatures to develop predictive models of central ADMET properties for drug development. pkCSM performs as well or better than current methods. A freely accessible web server (http://structure.bioc.cam.ac.uk/pkcsm), which retains no information submitted to it, provides an integrated platform to rapidly evaluate pharmacokinetic and toxicity properties.