Li Yue, Yi Jiacai, Li Hui, Li Kun, Kang Fenghua, Deng Youchao, Wu Chengkun, Fu Xiangzheng, Jiang Dejun, Cao Dongsheng
Xiangya School of Pharmaceutical Sciences, Central South University Changsha 410013 Hunan P.R. China
College of Computer, National University of Defense Technology Changsha 410073 Hunan China.
Chem Sci. 2025 Aug 19. doi: 10.1039/d5sc05395a.
Structure-based molecular docking, a cornerstone of computational drug design, is undergoing a paradigm shift fueled by deep learning (DL) innovations. However, the rapid proliferation of DL-driven docking methods has created uncharted challenges in translating predictions to biomedical reality. Here, we delve into the performance and prospects of traditional methods and state-of-the-art DL docking paradigms-encompassing generative diffusion models, regression-based architectures, and hybrid frameworks-across five critical dimensions: pose prediction accuracy, physical plausibility, interaction recovery, virtual screening (VS) efficacy, and generalization across diverse protein-ligand landscapes. We reveal that generative diffusion models achieve superior pose accuracy, while hybrid methods offer the best balance. Regression models, however, often fail to product physically valid poses, and most DL methods exhibit high steric tolerance. Furthermore, our analysis reveals significant challenges in generalization, particularly when encountering novel protein binding pockets, limiting the current applicability of DL methods. Finally, we explore failure mechanisms from a model perspective and propose optimization strategies, offering actionable insights to guide docking tool selection and advance robust, generalizable DL frameworks for molecular docking.
基于结构的分子对接作为计算药物设计的基石，正经历着由深度学习（DL）创新推动的范式转变。然而，由DL驱动的对接方法的迅速扩散在将预测转化为生物医学现实方面带来了未知的挑战。在此，我们深入研究传统方法和最新的DL对接范式（包括生成扩散模型、基于回归的架构和混合框架）在五个关键维度上的性能和前景：姿态预测准确性、物理合理性、相互作用恢复、虚拟筛选（VS）功效以及在不同蛋白质-配体格局中的泛化能力。我们发现生成扩散模型实现了卓越的姿态准确性，而混合方法提供了最佳平衡。然而，回归模型常常无法生成物理上有效的姿态，并且大多数DL方法表现出较高的空间耐受性。此外，我们的分析揭示了泛化方面的重大挑战，特别是在遇到新型蛋白质结合口袋时，这限制了DL方法当前的适用性。最后，我们从模型角度探索失败机制并提出优化策略，提供可操作的见解以指导对接工具的选择，并推进用于分子对接的强大、可泛化的DL框架。
