The AI Structural Biology (AISB) Network, powered by Apheris GmbH, has expanded its Federated OpenFold3 Initiative with the addition of Astex Pharmaceuticals, Bristol Myers Squibb, and Takeda. The new partners join founding members AbbVie and Johnson & Johnson to fine-tune OpenFold3 on proprietary structural datasets from multiple pharmaceutical companies while preserving confidentiality and intellectual property.

The initiative, developed in collaboration with Mohammed AlQuraishi’s Lab at Columbia University, pools thousands of experimentally determined protein–small molecule structures. By combining these datasets using Apheris’ federated computing platform, OpenFold3 is being trained to improve predictions of protein–ligand interactions, a critical step in small molecule drug discovery. Importantly, the federated approach ensures that sensitive data never leaves the contributing companies’ servers.

Paul Mortenson, senior director, computational chemistry and informatics at Astex, said: “By contributing data from our small molecule–protein structures to a federated effort, we can help ensure models like OpenFold3 better reflect the challenges medicinal chemists face every day, while keeping proprietary science protected.”

Payal Sheth, vice president, discovery biotherapeutics & lead discovery and optimization at Bristol Myers Squibb, added that federated learning enables collaboration on a scale no single company could achieve. “These efforts strengthen our ability to accelerate the development of transformational therapies for patients,” she said.

Hans Bitter, head of computational science and data strategy at Takeda, noted: “We are excited to be joining the AISB Network and providing additional data to support the refinement of OpenFold3. This promising prediction tool, focused on identifying and predicting binding affinities of small molecule–protein and antibody–antigen interactions, could be transformative in how we discover and develop small molecule therapeutics.”

Robin Röhm, co-founder and CEO of Apheris, emphasised the broader implications of the initiative: “In drug discovery, no single company has enough data to solve the hardest problems alone. Federated learning allows the industry to overcome this barrier. With the AISB Network, we’re showing that it is possible to combine the power of pharma datasets without ever moving or exposing them. The AISB Network will run several initiatives across both small and large molecules, and together these efforts are changing how AI models for drug discovery are developed.”

The Federated OpenFold3 Initiative represents one of the most ambitious collaborative efforts in AI-driven drug discovery, aiming to accelerate the design of small molecule and biologic therapeutics while preserving the privacy and security of proprietary data.