Cartesian message passing neural networks for directional properties: Fast and transferable atomic multipoles
Cartesian message passing neural networks for directional properties: Fast and transferable atomic multipoles
复制标题
DOI：
10.1063/5.0050444
复制
发表时间：
2021-06-14
影响因子：
4.4
通讯作者：
Sherrill, C. David
中科院分区：
文献类型：
--
作者：
Glick, Zachary L.;Koutsoukas, Alexios;Sherrill, C. David
消息传递神经网络(MPNN)框架是一种很有前途的原子性质建模工具，但直到最近，它还与笛卡尔张量等定向性质不兼容。我们提出了一种改进的笛卡尔MPNN(CMPNN)，它适用于预测原子为中心的多极子，这是从头算力场的重要组成部分。在一个新开发的包含46 623个化学结构和相应的高质量原子多极子的数据集上展示了该模型的有效性，该数据集存储在公开可用的分子科学软件研究所QCArchive服务器中。我们表明，CMPNN准确地预测了原子中心的电荷、偶极和四极，并且预测的原子多极的误差对多极-多极的静电能量的影响可以忽略不计。CMPNN足够精确，可以模拟分子电子结构的构象依赖关系。这开启了在整个模拟过程中动态重新计算原子多极的可能性，在模拟中，它们可能表现出强烈的构象依赖性。由AIP出版公司独家授权出版。
The message passing neural network (MPNN) framework is a promising tool for modeling atomic properties but is, until recently, incompatible with directional properties, such as Cartesian tensors. We propose a modified Cartesian MPNN (CMPNN) suitable for predicting atom-centered multipoles, an essential component of ab initio force fields. The efficacy of this model is demonstrated on a newly developed dataset consisting of 46 623 chemical structures and corresponding high-quality atomic multipoles, which was deposited into the publicly available Molecular Sciences Software Institute QCArchive server. We show that the CMPNN accurately predicts atom-centered charges, dipoles, and quadrupoles and that errors in the predicted atomic multipoles have a negligible effect on multipole-multipole electrostatic energies. The CMPNN is accurate enough to model conformational dependencies of a molecule's electronic structure. This opens up the possibility of recomputing atomic multipoles on the fly throughout a simulation in which they might exhibit strong conformational dependence. Published under an exclusive license by AIP Publishing.