Antibody Developability Consortium launched to advance AI-powered drug discovery

Ginkgo Datapoints and Apheris have launched the Antibody Developability Consortium, a new industry collaboration aimed at improving the prediction of manufacturability and developability risks in antibody candidates. The initiative seeks to address limitations in current predictive models, which are often hindered by small, fragmented, or inconsistent datasets.

By building a large-scale, standardized, and diverse dataset of approximately 10,000 antibodies, the consortium intends to create more robust AI models. Founding members include AbbVie, argenx, Lundbeck, and Takeda. These companies will contribute proprietary antibody sequences, while Ginkgo Datapoints will supplement the dataset with publicly available sources to ensure breadth and diversity.

The consortium utilizes privacy-preserving federated data networks, allowing members to access higher-performing models without compromising their proprietary data. This approach aims to identify promising antibody candidates earlier in the development process, potentially reducing investment risks and accelerating the delivery of new therapies to patients.