Department of Mathematics and Computer Science, Freie Universität Berlin, Arnimallee 12, 14195 Berlin, Germany.
Department of Physics and Astronomy, Rice University, Houston, Texas 77005, United States.
J Phys Chem Lett. 2023 May 4;14(17):3970-3979. doi: 10.1021/acs.jpclett.3c00444. Epub 2023 Apr 20.
Machine-learned coarse-grained (CG) models have the potential for simulating large molecular complexes beyond what is possible with atomistic molecular dynamics. However, training accurate CG models remains a challenge. A widely used methodology for learning bottom-up CG force fields maps forces from all-atom molecular dynamics to the CG representation and matches them with a CG force field on average. We show that there is flexibility in how to map all-atom forces to the CG representation and that the most commonly used mapping methods are statistically inefficient and potentially even incorrect in the presence of constraints in the all-atom simulation. We define an optimization statement for force mappings and demonstrate that substantially improved CG force fields can be learned from the same simulation data when using optimized force maps. The method is demonstrated on the miniproteins chignolin and tryptophan cage and published as open-source code.
基于机器学习的粗粒化（CG）模型在模拟大型分子复合物方面具有很大的潜力，而这是原子分子动力学所无法实现的。然而，训练准确的 CG 模型仍然是一个挑战。一种广泛使用的学习自下而上 CG 力场的方法是将所有原子分子动力学的力映射到 CG 表示，并平均匹配 CG 力场。我们表明，将所有原子力映射到 CG 表示的方法具有一定的灵活性，并且最常用的映射方法在原子模拟存在约束的情况下在统计上效率低下，甚至可能是不正确的。我们定义了一种力映射的优化语句，并证明当使用优化的力映射时，可以从相同的模拟数据中学习到改进的 CG 力场。该方法在 miniproteins chignolin 和 tryptophan cage 上进行了演示，并作为开源代码发布。
