Astex Pharmaceuticals, Bristol Myers Squibb and Takeda have agreed to pool data to support work on an artificial intelligence model, joining AbbVie and Johnson & Johnson on the starry list of participants in the program.
Biopharma companies are collectively sitting on a vast trove of data. Pooling resources to build a bigger, more diverse data set could theoretically yield drug discovery models that are beyond what any one company could build in isolation.
Astex, BMS and Takeda are the latest companies to bet on that vision. The trio agreed to contribute to the Federated OpenFold3 Initiative, a program that launched in March with the support of AbbVie and J&J. The Columbia University lab of Mohammed AlQuraishi, Ph.D., is developing OpenFold3.
Each partner is contributing data from thousands of experimentally determined protein-small molecule structures. Hans Bitter, Ph.D., head of computational science and data strategy at Takeda, discussed what he hopes to get out of the initiative.
“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,” Bitter said in a statement.
The AI Structural Biology Network is behind the initiative. Using Apheris’ federated computing platform, the project is designed to enable collaboration without moving or exposing sensitive data. Paul Mortenson, Ph.D., senior director of computational chemistry and informatics at Astex, said in a statement that the system will improve medicinal chemistry models “while keeping proprietary science protected.”