Research on Automated Optimization of Cell Manufacturing Protocols for Regenerative Medicine Using Our AI Published in eLife
Epistra Inc., as part of a joint research group with RIKEN and Robotic Biology Institute Inc., has developed a robot-AI system that autonomously conducts trial-and-error exploration of cell culture conditions, and has successfully improved cell culture protocols used in regenerative medicine.
Research Overview
In this research, we conducted proof-of-concept experiments for automated protocol optimization by combining a robot capable of high-precision life science experimental operations with our artificial intelligence software “Epistra Accelerate.” Through three rounds of experiments (144 conditions) conducted over six months, we achieved the highest evaluation score to date (91%).
Results
The research results demonstrate that robotic and AI-driven exploration alone, without expert intervention, can identify conditions that produce results of quality equivalent to those achieved by experts, proving the feasibility of autonomous experimentation in the field of cell culture.
Related Articles
Won Grand Prix in the 7th IP BASE AWARD Startup Category Hosted by the Japan Patent Office
On March 3, 2026, Epistra Inc. won the Grand Prix in the startup category at the '7th IP BASE AWARD' ceremony hosted by the Japan Patent Office.
Jikkenigaku Special Issue 'AI and Data-Driven Drug Discovery Research' Published — CTO Tsuzuki Authors Chapter on AI-Driven Drug Target Discovery
In 'Jikkenigaku Special Issue: AI and Data-Driven Drug Discovery Research' published by Yodosha, our CTO Taku Tsuzuki authored a chapter on AI-driven knowledge mining for drug target discovery.
Speaking Engagement: Delivered Keynote at the 10th Cell Manufacturing Kotozukuri Symposium
On March 4, 2026, CEO Yosuke Ozawa delivered a keynote speech on optimizing cell manufacturing processes using AI and robotics at the '10th Cell Manufacturing Kotozukuri Symposium' hosted by Osaka University.