Elix and LINC Launch First Federated AI Drug Discovery Platform

Elix, Inc. (CEO: Shinya Yuki / Headquarters: Tokyo, hereinafter “Elix”), an AI drug discovery company with the mission of "Rethinking Drug Discovery” and the Life Intelligence Consortium (Representative Director: Yasushi Okuno / Headquarters: Osaka, hereinafter “LINC”) are pleased to announce that for the first time in the world we have commercialized an AI drug discovery platform that incorporates multiple AI models trained using federated learning on data provided by 16 pharmaceutical companies.

The key to AI drug discovery lies in high-quality and sufficiently large datasets. Diverse and abundant data are indispensable for building superior AI models; however, pharmaceutical companies are generally limited to utilizing their own proprietary data and public datasets, resulting in significant data shortages that have posed major challenges to progress.

Federated learning technology provides a solution to this challenge. Elix, in partnership with the Department of Biomedical Data Intelligence, Graduate School of Medicine, Kyoto University, developed the federated learning library kMoL*¹, enabling multiple companies to collaboratively develop a suite of AI models without disclosing their confidential data externally. 16 pharmaceutical companies participated in building these federated learning-based models, which are now implemented on Elix Discovery™, Elix’s proprietary AI drug discovery platform.

By introducing Elix Discovery™, users can immediately leverage these newly developed models, and several pharmaceutical companies have already adopted the platform. The number of adoptions is expected to expand further, and Elix Discovery™ is on track to become the de-facto standard for AI drug discovery in Japan. Moreover, this initiative marks the world’s first commercialization of an AI drug discovery platform in partnership with numerous pharmaceutical companies utilizing federated learning.

The development of federated learning-based AI models*² was advanced through “Development of a Next-generation Drug Discovery AI through Industry-academia Collaboration” (DAIIA), an industry-academia collaborative program under the Project Promoting Support for Drug Discovery led by the Japan Agency for Medical Research and Development (AMED). Launched in FY2020 with the aim of establishing a drug discovery support infrastructure leveraging AI, the project involved 17 pharmaceutical companies, research institutes such as RIKEN, Kyoto University, Nagoya University, along with about 10 IT companies with AI expertise. The project concluded at the end of March 2025.

To ensure the continued operation and advancement of the innovative models and mechanisms cultivated in DAIIA, Elix—which already operates its own AI drug discovery platform—and LINC—a consortium supporting industry-academia collaboration in AI for life sciences with participation from many DAIIA member companies, have joined forces to commence the world’s first commercialization of an AI drug discovery platform serving models pre-trained with federated data, subsequently starting from April 2025. Through this initiative, we expect to see further adoptions of these technologies in real-world drug discovery settings.

Initially, the primary users will be the pharmaceutical companies that participated in DAIIA. However, as more companies join, the pool of available data will expand, further improving the accuracy and usability of the AI models for all users. We also plan to actively open the platform to companies that were not part of DAIIA. If you are interested in this initiative, please reach out to us via the contact information below.