
Lewis has spent over a decade developing and applying machine learning methods in drug discovery. Before joining Apheris, he founded a molecular property prediction platform and spent six years at AstraZeneca, where he worked on predictive modelling across chemistry, ADMET, and pharmacology. Lewis holds a PhD in Chemistry from the University of Cambridge and has contributed to major collaborative initiatives in the field, including the MELLODDY federated learning consortium and the JUMP–Cell Painting program, helping to shape how industry approaches shared model development.

ADMET liabilities drive 40–45% of clinical attrition, and models break on novel chemistry which is precisely where chemists need them most. Pre-competitive federated networks address this through complementary data collaboration. A walk through the evidence behind the Apheris ADMET Network.

Improved ADMET predictions come from complementary chemistry, not sheer data volume. A scientific study of public–proprietary integration shows why diversity, balance, and harmonization matter for reliability, calibration, and broader model applicability.

This blog explores state-of-the-art science in ADMET prediction, showing how federated learning enables pharma companies to collaboratively train models on diverse data, achieving higher accuracy and broader applicability without compromising data privacy.
© 2026 Apheris AI GmbH. All rights reserved.