arXiv cs.AI· Dingshuo Chen, Hao Yang, Kuangqi Zhou, Zhixun Li, Qiang Liu, Xiangxiang Zeng, Shu Wu, Yaosen Min·· 3 小时前AI 评分39
BindMol：基于热点感知与探索性偏好的药物发现生成框架
Binding Mode Matters: Hotspot-Aware Drug Discovery via Explorative Preferences
AI 导读
研究者提出 BindMol，一种基于定制化多目标强化学习算法的生成式药物发现框架，将靶向药物设计建模为多目标探索任务，每个目标对应增强与特定热点的相互作用。通过在训练中引入探索性偏好，BindMol 能发现兼具高亲和力、结构新颖性和多样化结合模式的分子，并在靶向药物发现与多性质优化任务中得到验证。该工作将发表于 NeurIPS 2026。
正文 · 原文
Abstract:Discovering hit molecules requires not just high binding affinity, but also the identification of diverse binding modes, which are critical for experimental assays. Existing generative approaches have predominantly relied on optimizing a scalar docking score, obscuring the distinct contributions of key binding determinants. To this end, we introduce a paradigm shift by formulating target-based drug design as a multi-objective exploration task, where each objective explicitly corresponds to enhancing interactions with a specific hotspot. Here, we introduce BindMol, a novel generative framework driven by a customized multi-objective reinforcement learning algorithm. By incorporating explorative preferences during training, our approach efficiently uncovers molecules with diverse and desirable binding profiles. Empirical results demonstrate that BindMol facilitates the discovery of high-affinity compounds characterized by both structural novelty and diverse binding modes. Validated across target-based drug discovery and multi-property optimization tasks, our approach provides a versatile paradigm for goal-oriented drug discovery.
| Comments: | To appear at NeurIPS 2026, 26 pages |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.14956 [cs.LG] |
| (or arXiv:2611.14956v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.14956 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Dingshuo Chen [view email]
[v1]
Wed, 11 Nov 2026 03:03:08 UTC (25,321 KB)
来源：arXiv cs.AI · arxiv.org