为千亿酶缺口定制生物钥匙！中国团队首创AI零样本酶设计方法

Core Insights - The recent breakthrough in AI enzyme design by MoleculeMind and Hong Kong Polytechnic University has been recognized at the ICML 2025 conference, marking a significant advancement in the field of AI enzyme design [1] - The traditional methods of enzyme discovery and optimization are time-consuming and costly, with a success rate of less than 1%, highlighting the urgent need for innovative solutions in the biomanufacturing sector [2] - The introduction of the SENZ method, which utilizes substrate structure similarity for enzyme design, represents a novel approach that could revolutionize enzyme generation [3][5] Industry Overview - Enzymes are crucial for the development of the trillion-dollar bio-economy, impacting sectors such as biomedicine, green chemistry, and environmental degradation [1] - The lack of ideal biocatalysts is a major barrier to scaling production in the biomanufacturing industry, leading to annual capacity losses exceeding $100 billion in pharmaceuticals, chemicals, and agriculture [2] - AI protein design has emerged as a promising solution to generate precise catalysts by learning from existing enzyme structure-function relationships, although it faces challenges with novel synthetic molecules due to limited training data [2] Company Developments - MoleculeMind has developed the SENZ method, which integrates biological data retrieval and generative AI to create enzymes without direct catalytic data, thus addressing a critical challenge in enzyme generation [3][5] - The SENZ method has demonstrated superior performance compared to traditional enzyme design methods, potentially providing tailored solutions for complex drug synthesis and environmental remediation [6] - MoleculeMind is expanding its capabilities in "on-demand design" across various fields, including antibodies, vaccines, and industrial enzymes, aiming to provide innovative biological solutions for health, environmental, and sustainability challenges [7]