- Evogene (EVGN, Financial) completes version 1.0 of its AI foundation model for small molecule design, achieving 90% precision.
- The model was developed with Google Cloud and trained on a dataset of 38 billion molecular structures.
- Version 2.0, focusing on multi-parameter optimization, is already in development.
Evogene Ltd. (EVGN), a leader in computational biology, has announced the completion of version 1.0 of its generative AI foundation model for small molecule design. Developed in collaboration with Google Cloud, the model enhances the capabilities of Evogene's ChemPass AI by addressing multiple complex product criteria simultaneously in both pharmaceutical and agricultural applications.
The proprietary foundation model showcases a remarkable 90% precision in novel molecule designs, significantly outperforming traditional GPT AI models, which achieve approximately 29% precision. This model is built on a comprehensive dataset comprising 38 billion molecular structures and is designed to facilitate the creation of potent, synthesizable, and patentable molecules, thereby ensuring robust intellectual property portfolios.
In addition to the successful completion of version 1.0, Evogene is already working on version 2.0 of its AI model. This new version aims to offer enhanced flexibility for multi-parameter optimization by incorporating predefined, customized parameters tailored for therapeutic contexts or specific agricultural requirements.
Ofer Haviv, President and CEO of Evogene, stated that this achievement opens up new frontiers for ChemPass AI, emphasizing the importance of generating novel molecules that not only perform optimally but also expand new intellectual property spaces. Boaz Maoz, Managing Director at Google Cloud Israel, expressed satisfaction in collaborating on this innovative endeavor, highlighting the synergy between advanced AI infrastructure and deep scientific insights.
Evogene invites continued collaboration with partners across the pharmaceutical and agriculture industries interested in leveraging ChemPass AI for next-generation product development.
