News & Publications


Learn more about Prous Institute and our progress in drug discovery, technology, industry partnerships, collaborations and scientific articles.

Prous Institute BioEpisteme® delivers encouraging findings from CYP3A4 and CYP2D6 model validation, presented at Society of Toxicology Annual Meeting in San Francisco

San Francisco, March 14, 2012 - Prous Institute for Biomedical Research has reported encouraging statistically significant results for cross-validation and external validation of cytochrome P450 3A4 and 2D6 inhibitor models created using its BioEpisteme® predictive software. The findings are a result of an FDA-approved research collaboration and are being presented jointly with the FDA Center for Drug Evaluation and Research (CDER) at the Annual Meeting of the Society of Toxicology in San Francisco. BioEpisteme®, a quantitative structure-activity relationship (QSAR) software package that creates customized drug prediction models against fundamental research endpoints, was made available to the FDA/CDER Office of Pharmaceutical Science via a research collaboration agreement with Prous Institute. Under the collaboration, BioEpisteme® was used to build in silico predictive QSAR models for the inhibition of CYP3A4 and CYP2D6. Cross-validation of the predictive performance of the models showed concordance of 87.8% for CYP3A4 and 74.7% for CYP2D6, and sensitivity of 94.7% and 88.3% respectively. External validation using test sets of known inhibitors for each enzyme confirmed the predictive performance of the models. Such models may therefore be used as a preliminary approach to predict whether an investigational drug is an inhibitor of metabolizing enzymes before further studies are undertaken. The presentation will form part of the Society of Toxicology platform session: “Bioinformatics: Signature Applications and Network Analyses” on Wednesday, March 14, 10:33-10:51. (A.J. Dixit et al. Society of Toxicology 51st Ann. Meeting (San Francisco, March 11-15, 2012), Abst. 1793).)     


To learn more about of BioEpisteme® prediction software, please visit