Spherical Message Passing for 3D Molecular Graphs
Spherical Message Passing for 3D Molecular Graphs
复制标题
DOI：
--
复制
发表时间：
2021-02
期刊：
影响因子：
--
通讯作者：
Yi Liu;Limei Wang;Meng Liu;Xuan Zhang;Bora Oztekin;Shuiwang Ji
中科院分区：
文献类型：
--
作者：
Yi Liu;Limei Wang;Meng Liu;Xuan Zhang;Bora Oztekin;Shuiwang Ji
我们考虑3D分子图的表示学习，其中每个原子与3D中的空间位置相关联。这是一个未充分探索的研究领域，目前缺乏一个原则性的消息传递框架。在这项工作中，我们进行分析，在球坐标系（SCS）的三维图形结构的完整识别。基于这样的观察，我们提出了球形消息传递（SMP）作为一种新的和强大的三维分子学习计划。SMP大大降低了训练的复杂性，使其能够在大规模分子上有效地执行。此外，SMP能够区分几乎所有的分子结构，未被发现的情况在实践中可能不存在。基于有意义的基于物理的3D信息表示，我们进一步提出了用于3D分子学习的SphereNet。实验结果表明，在SphereNet中使用有意义的3D信息可以显著提高预测任务的性能。我们的研究结果还证明了SphereNet在能力、效率和可扩展性方面的优势。我们的代码作为DIG库的一部分公开提供（https：//github.com/divelab/DIG）。
We consider representation learning of 3D molecular graphs in which each atom is associated with a spatial position in 3D. This is an under-explored area of research, and a principled message passing framework is currently lacking. In this work, we conduct analyses in the spherical coordinate system (SCS) for the complete identification of 3D graph structures. Based on such observations, we propose the spherical message passing (SMP) as a novel and powerful scheme for 3D molecular learning. SMP dramatically reduces training complexity, enabling it to perform efficiently on large-scale molecules. In addition, SMP is capable of distinguishing almost all molecular structures, and the uncovered cases may not exist in practice. Based on meaningful physically-based representations of 3D information, we further propose the SphereNet for 3D molecular learning. Experimental results demonstrate that the use of meaningful 3D information in SphereNet leads to significant performance improvements in prediction tasks. Our results also demonstrate the advantages of SphereNet in terms of capability, efficiency, and scalability. Our code is publicly available as part of the DIG library (https://github.com/divelab/DIG).