arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2505.09262physics.chem-phcs.AIcs.CVcs.LG
收藏

EDBench:用于分子建模的大规模电子密度数据

EDBench: Large-Scale Electron Density Data for Molecular Modeling

College of Computer Science and Electronic Engineering, Hunan University(湖南大学计算机科学与电子工程学院);College of Life Sciences, East China Normal University(华东师范大学生命科学学院);College of Materials Science and Engineering, Hunan University(湖南大学材料科学与工程学院);University of Science and Technology of China(中国科学技术大学);AIMS Lab, The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州)AIMS实验室);The Hong Kong University of Science and Technology(香港科技大学);Eastern Institute of Technology(技术研究所)

Hongxin Xiang, Ke Li, Mingquan Liu, Zhixiang Cheng, Bin Yao, Wenjie Du, Jun Xia, Li Zeng, Xin Jin, Xiangxiang Zeng

AI总结:

针对现有分子力场模型忽视电子密度的问题,本文提出大规模电子密度数据集EDBench,覆盖330万分子,并设计预测、检索、生成基准任务,验证了学习方法可高效高精度计算电子密度,降低计算成本。

摘要:

现有的分子机器学习力场(MLFFs)通常侧重于学习原子、分子和简单的量子化学性质(如能量和力),但忽略了电子密度(ED)ρ(r)在准确理解分子力场(MFFs)中的重要性。电子密度描述了在原子或分子周围特定位置找到电子的概率,根据Hohenberg-Kohn定理,它唯一决定了相互作用多粒子系统的所有基态性质(如能量、分子结构等)。然而,电子密度的计算依赖于耗时的第一性原理密度泛函理论(DFT),这导致缺乏大规模电子密度数据,并限制了其在MLFFs中的应用。在本文中,我们介绍了EDBench,一个大规模、高质量的电子密度数据集,旨在推动电子尺度上的基于学习的研究。EDBench基于PCQM4Mv2构建,提供了准确的电子密度数据,覆盖330万个分子。为了全面评估模型理解和利用电子信息的能力,我们设计了一套以电子密度为中心的基准任务套件,涵盖预测、检索和生成。我们对几种最先进方法的评估表明,从EDBench学习不仅可行,而且能达到高精度。此外,我们展示了基于学习的方法能够以相当的精度高效计算电子密度,同时相对于传统DFT计算显著降低计算成本。EDBench的所有数据和基准将免费提供,为电子密度驱动的药物发现和材料科学奠定坚实基础。

英文摘要:

Existing molecular machine learning force fields (MLFFs) generally focus on the learning of atoms, molecules, and simple quantum chemical properties (such as energy and force), but ignore the importance of electron density (ED) $ρ(r)$ in accurately understanding molecular force fields (MFFs). ED describes the probability of finding electrons at specific locations around atoms or molecules, which uniquely determines all ground state properties (such as energy, molecular structure, etc.) of interactive multi-particle systems according to the Hohenberg-Kohn theorem. However, the calculation of ED relies on the time-consuming first-principles density functional theory (DFT) which leads to the lack of large-scale ED data and limits its application in MLFFs. In this paper, we introduce EDBench, a large-scale, high-quality dataset of ED designed to advance learning-based research at the electronic scale. Built upon the PCQM4Mv2, EDBench provides accurate ED data, covering 3.3 million molecules. To comprehensively evaluate the ability of models to understand and utilize electronic information, we design a suite of ED-centric benchmark tasks spanning prediction, retrieval, and generation. Our evaluation on several state-of-the-art methods demonstrates that learning from EDBench is not only feasible but also achieves high accuracy. Moreover, we show that learning-based method can efficiently calculate ED with comparable precision while significantly reducing the computational cost relative to traditional DFT calculations. All data and benchmarks from EDBench will be freely available, laying a robust foundation for ED-driven drug discovery and materials science.

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。

↑