Highly accurate force fields are a mandatory requirement to generate predictive simulations. Here we present the path for the construction of machine learned molecular force fields by discussing the hierarchical pathway from generating the dataset of reference calculations to the construction of the machine learning model, and the validation of the physics generated by the model. We will use the symmetrized gradient-domain machine learning (sGDML) framework due to its ability to reconstruct complex high-dimensional potential-energy surfaces (PES) with high precision even when using just a few hundreds of molecular conformations for training. The data efficiency of the sGDML model allows using reference atomic forces computed with high-level wavefunction-based approaches, such as the $gold$ $standard$ coupled cluster method with single, double, and perturbative triple excitations (CCSD(T)). We demonstrate that the flexible nature of the sGDML framework captures local and non-local electronic interactions (e.g. H-bonding, lone pairs, steric repulsion, changes in hybridization states (e.g. $sp^2 \rightleftharpoons sp^3$), $n\toπ^*$ interactions, and proton transfer) without imposing any restriction on the nature of interatomic potentials. The analysis of sGDML models trained for different molecular structures at different levels of theory (e.g. density functional theory and CCSD(T)) provides empirical evidence that a higher level of theory generates a smoother PES. Additionally, a careful analysis of molecular dynamics simulations yields new qualitative insights into dynamics and vibrational spectroscopy of small molecules close to spectroscopic accuracy.
From Energy-Force Weighting to Primal-Dual Optimization of Machine-Learned Interatomic Potentials
School of Mathematical Sciences, Institute of Natural Sciences and MOE-LSC, Shanghai Jiao Tong University(上海交通大学数学科学学院、自然科学研究院和教育部线性科学中心);Department of Mathematics, National University of Singapore(新加坡国立大学数学系)
Teachers that teach the irrelevant: Pre-training machine learned interaction potentials with classical force fields for robust molecular dynamics simulations
University of California, Berkeley(加州大学伯克利分校);Lawrence Berkeley National Laboratory(劳伦斯伯克利国家实验室)
ORION: Unifying Top-Down and Bottom-Up Chemical Space Sampling for a Universal Organic Force Field
Future Technology School, Shenzhen Technology University(深圳技术大学未来技术学院);College of Applied Sciences, Shenzhen University(深圳大学应用科学学院);Heilongjiang Provincial Key Laboratory of Oil & Gas Chemical Technology, Northeast Petroleum University(东北石油大学油气化学技术黑龙江省重点实验室);Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University(清华大学化学工程系绿色化学反应工程与技术北京市重点实验室);School of Science, Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳)理学院);School of Energy and Power Engineering, Dalian University of Technology(大连理工大学能源与动力工程学院);Department of Chemical and Materials Engineering, University of Alberta(阿尔伯塔大学化学与材料工程系)
Developing Machine-Learned Potentials for Coarse-Grained Molecular Simulations: Challenges and Pitfalls
Institute of Informatics and Telecommunications(信息学与电信研究所);Institute of Nanoscience and Nanotechnology(纳米科学与纳米技术研究所);National Centre for Scientific Research “Demokritos”(“德谟克利特”国家科学研究中心);SciFY P.N.P.C.(SciFY公共非营利公司);School of Chemical Engineering(化学工程学院);National Technical University of Athens(雅典国立技术大学)
Generating High-Precision Force Fields for Molecular Dynamics Simulations to Study Chemical Reaction Mechanisms using Molecular Configuration Transformer
Peking University(北京大学);Shenzhen Bay Laboratory(深圳湾实验室);Changping Laboratory(昌平实验室)
Electrostatic Phenomenology Benchmarks for Machine-Learned Interatomic Potentials in Electrochemistry: Beyond the Energy-Force Metric
Fritz-Haber-Institut der Max-Planck-Gesellschaft(马克斯·普朗克学会弗里茨·哈伯研究所);Department of Chemistry and Biochemistry, University of South Carolina(南卡罗来纳大学化学与生物化学系);Department of Chemical Engineering, University of California, Davis(加州大学戴维斯分校化学工程系);Department of Materials, Imperial College London(帝国理工学院材料系);Department of Chemistry, Imperial College London(帝国理工学院化学系);Department of Materials Science and Engineering, University of Florida(佛罗里达大学材料科学与工程系);Department of Chemistry, Korea University(高丽大学化学系);Max Planck Institute for Sustainable Materials(马克斯·普朗克可持续材料研究所)
Broad chemical transferability in structure-based coarse-graining
与自顶向下的粗粒度(CG)模型相比,自底向上的方法能够提供更高的结构保真度。这种保真度来自与更高分辨率参考的紧密联系,使CG模型具有化学特异性。不幸的是,化学特异性可能与化合物筛选策略不一致,这需要可转移的参数化。在这里,我们提出了一种方法来调和自下而上,结构保持CG模型与化学转移。我们考虑了3,441个C $_7$ O $_2$小分子异构体的自底向上CG参数化。我们的方法结合了原子表示,无监督学习和大规模扩展集成力匹配参数化。我们首先确定了一个子集的19个代表性的分子,最大限度地编码的所有气相构象的局部环境。从均质散装液体和各种二元混合物中获得19个代表性分子之间的参考相互作用。一个扩展的合奏参数化所有703个状态点,导致CG模型,这是基于结构和化学转移。值得注意的是,由此产生的力场在结构上平均比单一状态点等效物更精确。对扩展系综进行平均可以作为平均力正则化器,消除对单个状态点过于特定的力和结构相关性。我们的方法旨在通过一组CG珠类型实现可转移性,这些CG珠类型可用于轻松构建新分子,同时保留基于结构的参数化的优点。
Machine-learned molecular mechanics force field for the simulation of protein-ligand systems and beyond
Memorial Sloan Kettering Cancer Center(纪念斯隆凯特琳癌症中心);Asahi Kasei Pharma Corporation(旭化成制药株式会社);University of California, Irvine(加州大学欧文分校);University of California San Diego(加州大学圣迭戈分校);University of Colorado Boulder(科罗拉多大学博尔德分校);Open Molecular Software Foundation(开放分子软件基金会)
Accurate Machine Learned Quantum-Mechanical Force Fields for Biomolecular Simulations
Google Research(谷歌研究院);Technische Universität Berlin(柏林工业大学);University of Luxembourg(卢森堡大学);Korea University(高丽大学);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所);Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所)
Algorithmic Differentiation for Automated Modeling of Machine Learned Force Fields
Technische Universität Berlin(柏林工业大学);Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所);Korea University(高丽大学);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所);Google Research(谷歌研究院)
Revealing the Physico-Chemical Basis of Organic Solid-Solid Wetting Deposition: Casimir-Like Forces, Hydrophobic Collapse, and the Role of the Zeta Potential
What Rate Theories Compute: Chemical kinetics from Arrhenius to machine-learned potentials
Center for Multidimensional Carbon Materials, Institute for Basic Science (IBS)(多维碳材料中心,韩国基础科学研究所);Department of Chemistry, Ulsan National Institute of Science and Technology (UNIST)(蔚山科学技术院化学系)
SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects
Technische Universität Berlin(柏林工业大学);Korea University(高丽大学);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所);BIFOLD – Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所);Google Research(谷歌研究院)
From Peptides to Nanostructures: A Euclidean Transformer for Fast and Stable Machine Learned Force Fields
TU Berlin(柏林工业大学);BIFOLD, Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所);Google DeepMind(谷歌DeepMind);Korea University(高丽大学);Max Planck Institut für Informatik(马克斯·普朗克计算机科学研究所)