Elix and LINC Unveil AI Drug Platform Using Pharma Data

Elix Inc. and the Life Intelligence Consortium (LINC), a general incorporated association are pleased to announce that they have commercialized the world’s first AI drug discovery platform equipped with multiple AI models that have been federated and trained using data held by 16 pharmaceutical companies.

The key to AI drug discovery is high-quality and sufficient data. Diverse and abundant data is essential to building an excellent AI model, but pharmaceutical companies can basically only use their own data and limited public data, and the lack of data has been a major barrier.

Federated learning technology has solved this problem. Elix developed the federated learning library kMoL together with the Department of Big Data Medical Sciences, Graduate School of Medicine, Kyoto University, enabling each company to jointly develop AI models without disclosing confidential data to outside parties. Sixteen pharmaceutical companies are participating in the development of AI models using federated learning, and the developed models are installed on Elix Discovery™, an AI drug discovery platform developed and operated by Elix.

Also Read: Actair® Promotion in Japan Transitions to CEOLIA Pharma

Elix Discovery™ makes it possible to utilize the developed models, and several pharmaceutical companies have already begun adopting it. The number of companies adopting it is expected to increase in the future, and we believe it will become the de facto standard for AI drug discovery platforms in Japan. In addition, this is the world’s first case of using federated learning to commercialize an AI drug discovery platform in collaboration with multiple pharmaceutical companies.

The development of AI models using federated learning has been promoted as part of the Japan Agency for Medical Research and Development (AMED)’s Drug Discovery Support Promotion Project and Industry-Academia Collaboration Development of Next-Generation Drug Discovery AI (DAIIA). This project was launched in fiscal 2020 with the aim of building a drug discovery support platform using AI technology, and involved 17 pharmaceutical companies, as well as research institutes such as the RIKEN, Kyoto University, and Nagoya University, and about 10 IT companies with AI technology. The project expired at the end of March 2025.

In order to continue to operate and develop the innovative model and mechanisms cultivated in DAIIA, Elix, which already operates its own AI drug discovery platform, and LINC, which supports industry-academia collaboration related to AI in the life sciences field and has many of the DAIIA member companies involved, will cooperate to begin the world’s first commercialization from April 2025. It is expected that this commercialization will further advance the adoption of the technology in drug discovery sites in the future.

At the start of commercialization, the main users will be pharmaceutical companies participating in DAIIA, but the more participating companies there are, the more data will be provided, and the more accurate and convenient the AI ​​models will be for all user companies.

SOURCE: PRTimes