Evogene Ltd. has introduced a groundbreaking generative AI model for small-molecule design, enhancing its ChemPass AI platform in collaboration with Google Cloud. Unveiled on June 10, 2025, this model addresses key challenges in pharmaceuticals and agriculture by enabling the concurrent optimization of multiple complex criteria such as efficacy, toxicity, and stability during compound discovery. Unlike traditional sequential screening, which is slow and often leads to failure, this AI-driven approach allows researchers to design effective molecules while minimizing costly late-stage setbacks.
ChemPass AI’s foundation model, termed ChemPass-GPT, leverages a massive dataset of 40 billion molecular structures and employs transformer neural networks to generate novel, chemically valid compounds. This model excels in multi-objective optimization, effectively balancing various properties throughout the design process—significantly enhancing the likelihood of success in clinical trials and creating innovative, patentable products in drug development and crop protection. Overall, Evogene’s technology is poised to revolutionize R&D by significantly accelerating and improving the discovery process across life sciences.
Source link
