arXiv 2508.154802025-11-24cs.LG

在双曲空间中学习蛋白质-配体结合

Learning Protein-Ligand Binding in Hyperbolic Space

Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China（人工智能产业研究院（AIR），清华大学，北京，中国）；University of Electronic Science and Technology of China, Chengdu, China（电子科技大学，成都，中国）；Department of Computer Science and Technology, Tsinghua University, Beijing, China（计算机科学与技术系，清华大学，北京，中国）；Beijing Academy of Artificial Intelligence, Beijing, China（北京人工智能研究院，北京，中国）

Jianhui Wang, Wenyu Zhu, Bowen Gao, Xin Hong, Ya-Qin Zhang, Wei-Ying Ma, Yanyan Lan

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

HypSeek在双曲空间中学习蛋白质-配体结合，通过双曲几何提升虚拟筛选和亲和力排名性能。

摘要：

蛋白质-配体结合预测是虚拟筛选和亲和力排名两项药物发现中的基本任务。尽管最近的检索基于方法将配体和蛋白质口袋嵌入到欧几里得空间中进行基于相似性的搜索，但欧几里得嵌入的几何结构往往无法捕捉分子相互作用中固有的层次结构和细粒度亲和力变化。在本工作中，我们提出了HypSeek，一种将配体、蛋白质口袋和序列嵌入到洛伦兹模型双曲空间中的双曲表示学习框架。通过利用双曲空间的指数几何和负曲率，HypSeek能够产生具有表达力和亲和力敏感性的嵌入，可以有效建模全局活性和细微功能差异，尤其是在挑战性情况下，如活性悬崖，其中结构相似的配体表现出显著的亲和力差距。我们的模型将虚拟筛选和亲和力排名统一在一个框架中，引入了一种蛋白质引导的三塔架构以增强表示结构。HypSeek在DUD-E上提高了虚拟筛选的早期富集率，从42.63提高到51.44（+20.7%），并在JACS上提高了亲和力排名的相关性，从0.5774提高到0.7239（+25.4%），证明了双曲几何在两项任务中的优势，并突显了其作为蛋白质-配体建模强大归纳偏置的潜力。

英文摘要：

Protein-ligand binding prediction is central to virtual screening and affinity ranking, two fundamental tasks in drug discovery. While recent retrieval-based methods embed ligands and protein pockets into Euclidean space for similarity-based search, the geometry of Euclidean embeddings often fails to capture the hierarchical structure and fine-grained affinity variations intrinsic to molecular interactions. In this work, we propose HypSeek, a hyperbolic representation learning framework that embeds ligands, protein pockets, and sequences into Lorentz-model hyperbolic space. By leveraging the exponential geometry and negative curvature of hyperbolic space, HypSeek enables expressive, affinity-sensitive embeddings that can effectively model both global activity and subtle functional differences-particularly in challenging cases such as activity cliffs, where structurally similar ligands exhibit large affinity gaps. Our mode unifies virtual screening and affinity ranking in a single framework, introducing a protein-guided three-tower architecture to enhance representational structure. HypSeek improves early enrichment in virtual screening on DUD-E from 42.63 to 51.44 (+20.7%) and affinity ranking correlation on JACS from 0.5774 to 0.7239 (+25.4%), demonstrating the benefits of hyperbolic geometry across both tasks and highlighting its potential as a powerful inductive bias for protein-ligand modeling.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。