In silico prediction of the inhibition activity of drug-metabolizing enzymes in rat

Abstract

In this study, we aimed to develop a new In silico method for predicting the inhibitory activity of chemical substances against drug-metabolizing enzymes using chemical structure information. We used In vitro experimental values of 218 substances selected from the HESS database as learning data. All models by random forest showed ROC-AUC achieved 0.8 or more. We have developed high-efficiency and high-performance In silico prediction models for the inhibitory activity of chemical substances to drug-metabolizing enzymes.