大年初一，浙大发顶刊论文：用人工智能为膜蛋白造“装甲”

Core Viewpoint - Zhejiang University published a paper titled "De novo Design of GPCR Exoframe Modulators" in Nature, showcasing the practical capabilities of artificial intelligence in membrane protein design and functional reconstruction, providing a new technical pathway for understanding and utilizing the regulatory potential of G protein-coupled receptor exoframe modulators (GEM) [1][3]. Group 1 - The research team, led by interdisciplinary doctoral student Cheng Shizhuo, includes members from Zhejiang University School of Medicine and the School of Computer Science, indicating a collaborative approach to tackle GPCR-related challenges [3][5]. - GPCRs, as the largest "signal receivers" in the human body, are crucial for regulating key life processes such as sensation, emotion, cardiovascular function, and metabolism [3][5]. - The study proposes an "exoskeleton" approach to receptor regulation, where GEM acts on the transmembrane domain of GPCRs, altering their conformational state to finely tune signal output, differing from traditional drugs that occupy receptor pockets [5][6]. Group 2 - The research team discovered that the AI-designed GEM modules can be fused with GPCRs, expanding the classic seven transmembrane helices to nine or even thirteen, thereby regulating downstream signaling pathways and their corresponding functions [5][6]. - The interdisciplinary training of Cheng Shizhuo involves attending multiple group meetings weekly to integrate medical knowledge and computer skills, transforming biological genetic codes into logical computational languages [5][6]. - The team aims to provide new therapeutic strategies for GPCR functional disorders, such as certain Parkinson-like syndromes, which have long lacked effective intervention methods [8].