RevolKa Director & CSO Prof. Mitsuo Umetsu (Tohoku-Univ.) and Team Announce New AI-Driven Method toAccelerate Discovery of Highly Functional Antibodies
Preprint Published on bioRxiv Highlights Application of Massive Sequence Data to Machine Learning for Enhanced Molecular Design
RevolKa Ltd. announced today that a research group led by Prof. Mitsuo Umetsu, Director & Chief Scientific Officer (CSO) of RevolKa and Professor at the Graduate School of Engineering, Tohoku University, has published a preprint on bioRxiv, the repository for the life sciences. The paper describes a novel AI-powered approach that accelerates the discovery of high-affinity antibodies—those capable of binding strongly to specific target molecules.
- Research Overview
The study reports an efficient method for discovering and designing high-performance antibody candidates. By leveraging the vast amounts of sequence data (deep sequencing data) generated during the selection process of VHH antibody libraries, the team successfully trained machine
learning models to optimize both training data selection and search space design, leading to a more streamlined and precise discovery process.
- Future Outlook and Applications at RevolKa
RevolKa intends to integrate the insights and methodologies derived from this study into its proprietary antibody and protein optimization technologies. This integration will further advance RevolKa’s AI-driven molecular design platform, enabling faster, more efficient, and groundbreaking research and development for therapeutic solutions.
Note: This preprint is a preliminary report of work that has not been certified by peer review (it reports new medical/biological research that has yet to be evaluated and so should not be used to guide clinical practice).
