Department of Chemistry, Norwegian University of Science and Technology, Høgskoleringen 5, 7491 Trondheim, Norway.
Department of Chemistry, University of Helsinki, P.O. Box 55, FI-00014 Helsinki, Finland.
J Chem Theory Comput. 2021 Oct 12;17(10):6193-6202. doi: 10.1021/acs.jctc.1c00458. Epub 2021 Sep 24.
We propose to analyze molecular dynamics (MD) output a supervised machine learning (ML) algorithm, the decision tree. The approach aims to identify the predominant geometric features which correlate with trajectories that transition between two arbitrarily defined states. The data-driven algorithm aims to identify these features without the bias of human "chemical intuition". We demonstrate the method by analyzing the proton exchange reactions in formic acid solvated in small water clusters. The simulations were performed with MD combined with a method to efficiently sample the rare event, path sampling. Our ML analysis identified relevant geometric variables involved in the proton transfer reaction and how they may change as the number of solvating water molecules changes.
我们建议分析分子动力学（MD）输出，使用监督机器学习（ML）算法，决策树。该方法旨在识别与在两个任意定义状态之间跃迁的轨迹相关的主要几何特征。该数据驱动算法旨在在没有人为“化学直觉”偏见的情况下识别这些特征。我们通过分析甲酸在小水簇中的溶剂化质子交换反应来演示该方法。模拟是通过 MD 与一种有效采样稀有事件的方法（路径采样）相结合进行的。我们的 ML 分析确定了质子转移反应中涉及的相关几何变量，以及随着溶剂水分子数量的变化它们如何变化。
