Wang Xing-You, Chen Yang, Li Yu-Fan, Wei Chao-Yang, Liu Meng-Ya, Yuan Chen-Xing, Zheng Yao-Yu, Qin Mo-Han, Sheng Yu-Feng, Tong Xiao-Chu, Zheng Ming-Yue, Li Xu-Tong
Drug Discovery and Design Center, State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica Chinese Academy of Sciences, Shanghai, 201203, China.
University of Chinese Academy of Sciences, Beijing, 100049, China.
Acta Pharmacol Sin. 2025 Jun 17. doi: 10.1038/s41401-025-01591-x.
The discovery of active compounds for novel, underexplored targets is essential for advancing innovative therapeutics across a wide range of diseases. Recent advancements in artificial intelligence (AI) are revolutionizing active compound discovery by dramatically enhancing the efficiency, accuracy, and scalability previously challenged by traditional methods. This review provides a comprehensive overview of AI-driven methodologies for active compound discovery, with a particular focus on their application to novel targets. Initially, we explore how AI overcomes traditional bottlenecks in molecular design, enabling precise protein perception through high-accuracy protein structure prediction and enhanced docking precision. Building upon these target-focused capabilities, AI-driven approaches also advance ligand exploration, effectively bridging biological and chemical spaces through sophisticated data transfer techniques that maximize the utility of available activity data. By assessing overall cellular or organismal responses, AI plays a pivotal role in decoding complex biological systems, driving phenotypic drug discovery (PDD) through multi-modal data integration. Finally, we discuss how AI is addressing challenges associated with targeting previously undruggable proteins, exemplified by the development of protein degraders. By synthesizing these cutting-edge advancements, this review serves as a valuable resource for researchers seeking to leverage AI in the discovery of next-generation therapeutics.
发现针对新型、未充分探索靶点的活性化合物对于推进针对多种疾病的创新疗法至关重要。人工智能（AI）的最新进展正在彻底改变活性化合物的发现方式，极大地提高了传统方法以前面临挑战的效率、准确性和可扩展性。本综述全面概述了用于活性化合物发现的人工智能驱动方法，特别关注其在新型靶点上的应用。首先，我们探讨人工智能如何克服分子设计中的传统瓶颈，通过高精度蛋白质结构预测和提高对接精度实现精确的蛋白质感知。基于这些以靶点为重点的能力，人工智能驱动的方法还推进了配体探索，通过复杂的数据转移技术有效地弥合生物和化学空间，从而最大限度地利用可用的活性数据。通过评估整体细胞或机体反应，人工智能在解码复杂生物系统中发挥关键作用，通过多模态数据整合推动表型药物发现（PDD）。最后，我们讨论人工智能如何应对与靶向以前难以成药的蛋白质相关的挑战，以蛋白质降解剂的开发为例。通过综合这些前沿进展，本综述为寻求在发现下一代疗法中利用人工智能的研究人员提供了宝贵资源。
