[论文] Pyrite: Building and Benchmarking Protein-Ligand Pose Ranking Agents
研究领域: ML 作者: Megha Srivastava, Michael H. Lim, Allison M. Keys, Pranav Murugan, Nate Gruver 发布时间: 2026-11-10 arXiv: 2611.14511
论文概要
研究领域: ML 作者: Megha Srivastava, Michael H. Lim, Allison M. Keys, Pranav Murugan, Nate Gruver 发布时间: 2026-11-10 arXiv: 2611.14511
中文摘要
预测配体如何结合靶点是制药小分子发现的核心。现代共折叠模型可生成多样候选构象,但可靠排序仍是挑战。我们提出 Pyrite:由 1500 对构象组成、覆盖多种蛋白-配体复合物的成对构象排序基准。配备 3D 可视化与物理评分的专家化学家表现接近随机,凸显解读现有化学信号的难度。为探索自动化排序,我们提出 PyriteRep——分子几何、蛋白-配体相互作用与理化评分的结构化表示,并构建两个 LLM 智能体:PyriteReasoner(文本推理直接比较构象)与 PyriteFeaturizer(生成可执行代码为逻辑回归构造特征,可融合 LLM 子调用特征)。二者准确率达 67% 与 74%,超过人类专家。LLM 引导的特征发现总体优于直接推理;GRPO 提升开源权重模型,使其匹敌甚至超越闭源前沿模型。结果确立了 Pyrite 作为构象排序能力的挑战性测试,凸显 LLM 引导的特征发现是理解和组合计算化学工具输出的有前景方向。
原文摘要
Predicting how a ligand binds to its target is central to small-molecule drug discovery. Modern cofolding models can generate diverse candidate poses, but reliably ranking these candidates remains challenging. We introduce Pyrite, a pairwise pose-ranking benchmark comprising 1,500 pose pairs across diverse protein-ligand complexes. Expert chemists equipped with 3D visualization tools and physics-based scores perform near chance, highlighting the difficulty of interpreting available chemical signals. To investigate automated ranking, we introduce PyriteRep, a structured representation of molecular geometry, protein-ligand interactions, and physicochemical scores, and develop two LLM-based agents: PyriteReasoner, which directly compares poses through text-based reasoning, and PyriteFeaturize...
*自动采集于 2026-11-12*
#论文 #arXiv #ML #小凯