Apheris

Apheris is a life sciences AI platform that delivers superior drug discovery models through federated data networks. By enabling secure, local deployment of AI models and powering industry-wide federated collaborations, Apheris allows pharmaceutical organizations to benefit from cross-company data without ever exposing proprietary information. The platform is purpose-built for drug discovery teams who need to understand model reliability, benchmark performance on internal datasets, and adapt models as their programs evolve.

Apheris powers the industry's largest federated data networks spanning multiple pharmaceutical organizations, and provides the technology layer for collaborations involving leading biopharma companies including AbbVie, Astex, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, Genentech, Johnson & Johnson, Sanofi, and Takeda.

Federated Data Networks

• AI Structural Biology (AISB) Network: An unprecedented collaboration among major biopharma organizations aimed at transforming AI drug discovery. State-of-the-art AI models are trained and evaluated on unique data from multiple companies without exposing proprietary information, with Apheris providing the core technology layer.
• ADMET Network: Enables drug discovery teams to design more informative batches and make better use of experimental capacity within Design-Make-Test-Analyze (DMTA) cycles.
• Antibody Developability Network: Supports antibody research and development with purpose-built datasets and federated AI training to advance developability assessments.

Key Products and Applications

• ApherisFold: A co-folding model application that can be used independently of any federated data network. Teams can run co-folding models locally, benchmark model versions on in-house data, and fine-tune models on proprietary data entirely within their own environment. Federated networks further extend fine-tuning capabilities across organizations.
• Secure Local Inference: Models are deployed and run within the customer's own environment, ensuring that data, queries, and outputs remain in-house at all times.
• In-House Benchmarking: Teams can compare model versions against curated internal datasets to evaluate reliability and understand model behavior on their own data.
• Model Customization: Models can be fine-tuned for specific targets and chemotypes, with results accessible via a graphical user interface (GUI) or API for integration into existing discovery workflows.

Engagement Models and Deployment

• Local Application Use: Organizations can deploy and run models in their own environment for secure inference, benchmarking, and customization without reliance on external infrastructure.
• Join a Federated Data Network: Companies can participate in industry-wide data networks to improve model performance using cross-company data — without sharing the underlying proprietary data.
• Build a Custom Network: Organizations can create their own federated collaboration across partners, internal sites, or disparate data sources using Apheris' federated infrastructure.
• API and GUI Integration: Results and model outputs can be integrated into existing drug discovery workflows through both programmatic and graphical interfaces.

Design Philosophy and Compliance

• Apheris is designed specifically for the realities of drug discovery teams, addressing the need to understand when models are reliable, how they behave on proprietary data, and how to adapt them as programs evolve.
• The platform is described as certifiably secure, ensuring that participation in federated networks does not require exposure of sensitive or proprietary organizational data.
• All federated model training and evaluation occurs without sharing underlying data between participating organizations, preserving data confidentiality across the network.

Apheris is trusted by leading pharmaceutical companies and partners who recognize the transformative potential of federated AI in accelerating drug discovery. By combining secure local deployment with access to models trained across industry-wide datasets, Apheris enables organizations to develop better medicines faster while maintaining full control over their proprietary information.