Boston — Ginkgo Datapoints, an offering of Ginkgo Bioworks, and Apheris GmbH announced the launch of the Antibody Developability Consortium with founding members AbbVie, argenx, Lundbeck and Takeda.
The consortium is designed to help pharmaceutical and biotechnology companies identify antibody manufacturability and developability risks earlier by building a large standardized dataset and related artificial intelligence models.
Antibody developability refers to the biophysical properties that determine whether a candidate can be manufactured, formulated and advanced into a clinical product. The companies said current predictive models are often limited by small, fragmented or inconsistent datasets.
Under the consortium, founding members will contribute proprietary antibody sequences, while Ginkgo Datapoints will supplement the dataset with publicly available sequences to reach a total of 10,000 antibodies.
Using Apheris’ federated infrastructure, participating companies will be able to train, benchmark and refine AI models using the consortium dataset without exposing their proprietary sequences to other members.
The consortium is structured so members can apply models trained on the full dataset and further fine-tune them using their own proprietary data while retaining ownership of the sequences and assay data they contribute.
Ginkgo Datapoints will lead the scientific design and execution of the initiative, including sequence selection, antibody production and high-throughput laboratory characterization across key developability measures.
Ginkgo will also train a foundation antibody developability model on the resulting dataset within Apheris’ secure environment. Apheris will provide the infrastructure that allows participating companies to use and refine the model within their own environments.
Charlotte Deane, Professor of Structural Bioinformatics at the University of Oxford, and Peter Tessier, Professor of Pharmaceutical Sciences and Chemical Engineering at the University of Michigan, will provide independent scientific oversight.
The initial dataset is expected to be delivered to consortium members by early 2027. The group also plans to explore more complex antibody formats and additional properties that could help predict which drug candidates are more likely to succeed or fail during development.
“For AI to impact developability decisions in a drug program, it has to perform on a pharma’s own molecules,” said Robin Röhm, CEO and co-founder of Apheris. “The Antibody Developability Consortium delivers the largest standardized antibody dataset and the foundation model trained on it.”
Rich Cohen, Senior Director at Ginkgo Datapoints, said the initiative combines Ginkgo’s large-scale laboratory data generation capabilities with sequence diversity and predictive modeling.
“We are building the largest, most standardized antibody developability dataset the industry has ever seen, along with the predictive models trained on it,” Cohen said.
Representatives from AbbVie, argenx, Lundbeck and Takeda said the consortium could help improve predictive modeling, support earlier drug development decisions and accelerate antibody discovery.
