Ginkgo Datapoints, an offering of Ginkgo Bioworks, and Apheris GmbH have announced that the Antibody Developability Consortium has officially launched with its founding members. The founding members include AbbVie, argenx, Lundbeck, and Takeda, and the consortium remains open to additional pharmaceutical and biotech companies. The initiative is designed to help companies predict manufacturability and developability risks earlier by building the largest standardized antibody developability dataset in the field.
Addressing a Critical Drug Development Challenge
Developability challenges can prevent otherwise promising antibody candidates from progressing efficiently toward patients. Existing predictive models have been limited by small, fragmented, and inconsistent datasets, while even large internal datasets are constrained in sequence diversity. The consortium was created to address this gap by unifying efforts to produce a standardized, purpose-built, diverse dataset and corresponding AI models at scale.
How the Consortium Operates
Each founding member will contribute proprietary antibody sequences, and Ginkgo Datapoints will fill any remaining capacity from publicly available sources to reach 10,000 antibodies in total. Using Apheris infrastructure, members can train, benchmark, and refine AI models on the full consortium dataset without exposing raw proprietary sequences to other members. Members can apply resulting models and their own fine-tuned derivatives internally while retaining ownership of the proprietary sequences and assay data they contribute.
Scientific Leadership and Data Protection
Ginkgo Datapoints leads the scientific design and execution of the consortium, including sequence selection, antibody production, and high-throughput wet-lab characterization across core developability endpoints. Ginkgo also trains a foundation antibody developability model within Apheris secure environment. Apheris delivers the foundation model into each member's environment for fine-tuning on proprietary data and protects the confidentiality of contributed sequences.
Goals and Timeline
The consortium's initial dataset is targeted for delivery to members by early 2027. Over time, the initiative will explore the addition of more complex antibody formats to enable new drug classes and other key properties that support emerging therapeutic approaches. These capabilities could support early predictions of which drug candidates will succeed or fail during clinical development.
Leadership Perspectives
Robin Röhm, CEO and co-founder of Apheris, said that for AI to impact developability decisions, it has to perform on a pharma's own molecules. Rich Cohen of Ginkgo Datapoints emphasized the lab data generation scale and published modeling track record needed to lead the initiative. Athena Hadjixenofontos of AbbVie noted that federated infrastructure enables participants to contribute data while keeping proprietary sequences private.
Member Perspectives on Collaboration
Erwin Pannecoucke of argenx said predictive developability models can significantly accelerate discovery and development, and the consortium's extensive antibody dataset lets partners learn together. Allan Jensen of Lundbeck said that in complex therapeutic areas such as CNS, the ability to select well-behaved candidates with superior developability properties is essential. Yves Fomekong Nanfack of Takeda said pooling standardized developability data across the industry can create stronger predictive models than any one company could build alone.
The consortium represents an important step forward in building predictive models for antibody developability by creating datasets designed for machine learning. By pooling standardized data while preserving privacy, members can strengthen candidate selection, reduce avoidable development time, and improve investment decisions. This collaborative model could meaningfully accelerate antibody discovery and help advance new medicines for patients.