Machine-Learning-Assisted Free Energy Simulation of Solution-Phase and Enzyme Reactions.
Machine-Learning-Assisted Free Energy Simulation of Solution-Phase and Enzyme Reactions.
复制标题
机器学习辅助的溶液和酶反应的自由能模拟。
DOI：
10.1021/acs.jctc.1c00565
复制
发表时间：
2021-09-14
影响因子：
5.5
通讯作者：
Shao, Yihan
中科院分区：
文献类型：
--
作者：
Pan, Xiaoliang;Yang, Junjie;Van, Richard;Epifanovsky, Evgeny;Ho, Junming;Huang, Jing;Pu, Jingzhi;Mei, Ye;Nam, Kwangho;Shao, Yihan
尽管最近在开发用于生物分子模拟的机器学习潜能（MLP）方面取得了进展，但在开发用于酶促反应的稳定且准确的MLP方面的努力有限。在这里，我们报告了一个协议，用于在从头算量子力学和分子力学（ai-QM/MM）的准确性水平上进行溶液相和酶反应的机器学习辅助自由能模拟。在我们的协议中，MLP被构建为再现ai-QM/MM能量以及QM（反应性）和MM（溶剂/酶）原子上的力。作为替代策略，训练增量机器学习势（ΔMLP）以再现ai-QM/MM与半经验（se）QM/MM能量和力之间的差异。为了考虑MLP和ΔMLP中凝聚相环境的影响，将分子系统的DeePMD表示扩展到每个QM原子上的外部静电势和场。以Menshutkin反应和分支酸盐反应为例，我们证明了所发展的MLP和ΔMLP再现了ai-QM/MM反应的能量和力，对于沿着反应途径的典型构型，误差平均小于1.0 kcal/mol/m2。对于这两个反应，基于MLP/Δ MLP的模拟产生的自由能分布与参考ai-QM/MM结果的差异小于1.0 kcal/mol，但仅以分数计算成本。
Despite recent advances in the development of machine learning potentials (MLPs) for biomolecular simulations, there has been limited effort in developing stable and accurate MLPs for enzymatic reactions. Here, we report a protocol for performing machine learning assisted free energy simulation of solution-phase and enzyme reactions at an ab initio quantum mechanical and molecular mechanical (ai-QM/MM) level of accuracy. Within our protocol, the MLP is built to reproduce the ai-QM/MM energy as well as forces on both QM (reactive) and MM (solvent/enzyme) atoms. As an alternative strategy, a delta machine learning potential (ΔMLP) is trained to reproduce the differences between ai-QM/MM and semi-empirical (se) QM/MM energy and forces. To account for the effect of the condensed–phase environment in both MLP and ΔMLP, the DeePMD representation of a molecular system is extended to incorporate external electrostatic potential and field on each QM atom. Using the Menshutkin and chorismate mutase reactions as examples, we show that the developed MLP and ΔMLP reproduce the ai-QM/MM energy and forces with an error on average less than 1.0 kcal/mol and 1.0 kcal/mol/Å for representative configurations along the reaction pathway. For both reactions, MLP/ΔMLP-based simulations yielded free energy profiles that differed by less than 1.0 kcal/mol from the reference ai-QM/MM results, but only at a fractional computational cost.
登录
查看更多内容
DOI：
10.1073/pnas.90.18.8600
发表时间：
1993-09-15
影响因子：
11.1
作者：
CHOOK, YM;KE, HM;LIPSCOMB, WN
通讯作者：
LIPSCOMB, WN
影响因子：
4.4
作者：
Fukunishi, H;Watanabe, O;Takada, S
通讯作者：
Takada, S
影响因子：
1.7
作者：
Boresch, Stefan;Woodcock, H. Lee
通讯作者：
Woodcock, H. Lee
影响因子：
8.6
作者：
Behler, Joerg;Parrinello, Michele
通讯作者：
Parrinello, Michele
影响因子：
5.5
作者：
Boeselt, Lennard;Thuerlemann, Moritz;Riniker, Sereina
通讯作者：
Riniker, Sereina