Read the article: https://www.techtimes.com/articles/327885/20260923/five-pharma-rivals-ran-federated-drug-discovery-ai-20000-secret-structures-it-won.htm
AbbVie, Astex, Bristol Myers Squibb, Johnson & Johnson and Takeda trained AISB-1-Fed on more than 20,000 proprietary protein-ligand structures, with only model weights leaving each company.

Read the article: https://www.techtimes.com/articles/327885/20260923/five-pharma-rivals-ran-federated-drug-discovery-ai-20000-secret-structures-it-won.htm
Introducing Apheris Foundry: federated drug discovery models, customized to your data and accessible to your agents
Affinity scores should depend on how a compound binds. ApherisAffinity uses structurally validated training data, a geometry-aware affinity head and disrupted-binding-site negatives, so its predictions respond to target changes and rank compounds in a real discovery series better than Boltz‑2 and Nesso‑1.

Apheris federated networks for drug discovery and European healthcare studies can now run on T-Systems’ sovereign infrastructure, with patient data staying inside each hospital.
NEWSLETTER
A quaterly briefing on operationalizing and customizing co-folding, ADMET and antibody developability models. Written by Robin Röhm, co-founder and CEO at Apheris.

Robin Röhm
Co-Founder & CEO
© 2026 Apheris AI GmbH. All rights reserved.