We have developed a novel differential equation solver software called PND based on the physics-informed neural network for molecular dynamics simulators. Based on automatic differentiation technique provided by Pytorch, our software allows users to flexibly implement equation of atom motions, initial and boundary conditions, and conservation laws as loss function to train the network. PND comes with a parallel molecular dynamics (MD) engine in order for users to examine and optimize loss function design, and different conservation laws and boundary conditions, and hyperparameters, thereby accelerate the PINN-based development for molecular applications.
Quantitative Analysis of Molecular Transport in the Extracellular Space Using Physics-Informed Neural Network
Tsinghua University（清华大学）；Peking University Health Science Center（北京大学医学部）；Fudan University（复旦大学）；Xiamen University（厦门大学）；Peking University Third Hospital（北京大学第三医院）
Sliced Denoising: A Physics-Informed Molecular Pre-Training Method
Academy of Mathematics and Systems Science（数学与系统科学研究院）；Chinese Academy of Sciences（中国科学院）；Institute for AI Industry Research (AIR)（智能产业研究院（AIR））；Tsinghua University（清华大学）
Physics-informed neural networks have been widely applied to learn general parametric solutions of differential equations. Here, we propose a neural network to discover parametric eigenvalue and eigenfunction surfaces of quantum systems. We apply our method to solve the hydrogen molecular ion. This is an ab-initio deep learning method that solves the Schrodinger equation with the Coulomb potential yielding realistic wavefunctions that include a cusp at the ion positions. The neural solutions are continuous and differentiable functions of the interatomic distance and their derivatives are analy…
Time-dependent Partial Differential Equations with given initial conditions are considered in this paper. New differentiation techniques of the unknown solution with respect to time variable are proposed. It is shown that the proposed techniques allow to generate accurate higher order derivatives simultaneously for a set of spatial points. The calculated derivatives can then be used for data-driven solution in different ways. An application for Physics Informed Neural Networks by the well-known DeepXDE software solution in Python under Tensorflow background framework has been presented for thr…
MultiPUFFIN: A Multimodal Domain-Constrained Foundation Model for Molecular Property Prediction of Small Molecules
Department of Chemical Engineering, Norwegian University of Science and Technology (NTNU)（挪威科学与技术大学化学工程系）；Faculty of Industrial Engineering, KU Leuven（鲁文大学工业工程学院）；University of Surrey（萨里大学）
MIRRA: Fast, High-Fidelity Molecular Generation with Chemistry-Informed Discrete Flow Matching
Courant Institute of Mathematical Sciences, New York University（纽约大学柯朗数学科学研究所）；Tandon School of Engineering, New York University（纽约大学坦登工程学院）；Department of Physics, New York University（纽约大学物理系）
Molecular Insights into Yb(III) Speciation in Sulfate-Bearing Hydrothermal Fluids from X-ray Absorption Spectra Informed by ab initio Molecular Dynamics
Structure-Informed Bayesian Inference of Anomalous Transport and Hidden Molecular Trapping in Amorphous Media
Laboratory for Disordered Systems, Phystech School of Applied Mathematics and Computer Science, Moscow Institute of Physics and Technology（莫斯科物理技术学院应用数学与计算机科学普列特赫学院无序系统实验室）；Center for Computational Physics, Landau School for Physics and Research, Moscow Institute of Physics and Technology（莫斯科物理技术学院朗道物理与研究中心计算物理中心）
Lost in Translation: Simulation-Informed Bayesian Inference Improves Understanding of Molecular Motion From Neutron Scattering
University of Bristol（布里斯托大学）；School of Chemistry, University of Bristol（布里斯托大学化学学院）；Centre for Computational Chemistry, School of Chemistry, University of Bristol（布里斯托大学化学学院计算化学中心）；University of Oxford（牛津大学）；Department of Materials, University of Oxford（牛津大学材料系）；The Faraday Institution（法拉第研究所）；Rutherford Appleton Laboratory（拉瑟福德·阿普尔顿实验室）；ISIS Pulsed Neutron and Muon Facility, Rutherford Appleton Laboratory（拉瑟福德·阿普尔顿实验室中子与缪子源设施）；University of Stavanger（斯塔万格大学）；Department of Mathematics and Physics, University of Stavanger（斯塔万格大学数学与物理系）；University of Bath（巴斯大学）；Department of Chemistry, University of Bath（巴斯大学化学系）；Diamond Light Source（钻石光源）
A software package, called DFTBaby, is published, which provides the electronic structure needed for running non-adiabatic molecular dynamics simulations at the level of charge-consistent tight-binding DFT. A long-range correction is incorporated to avoid spurious charge transfer states. Excited state energies, their analytic gradients and scalar non-adiabatic couplings are computed using tight-binding TD-DFT. These quantities are fed into a molecular dynamics code, which integrates Newton's equations of motion for the nuclei together with the electronic Schrödinger equation. Non-adiabatic eff…
A systematic comparison of hard- and soft-constrained physics-informed molecular machine learning with the Clapeyron equation
RWTH Aachen University（亚琛工业大学）；JARA Center for Simulation and Data Science (CSD)（JARA 模拟与数据科学中心）；Institute of Climate and Energy Systems ICE-1: Energy Systems Engineering, Forschungszentrum Jülich GmbH（于利希研究中心气候与能源系统研究所）
VIANA: character Value-enhanced Intensity Assessment via domain-informed Neural Architecture
LSRE-LCM, ALiCE, Faculty of Engineering, University of Porto（波尔图大学工程学院LSRE-LCM、ALiCE）；Department of Chemical Engineering, Norwegian University of Science and Technology（挪威科技大学化学工程系）；Corporate Fragrance R&D, The Procter & Gamble Company（宝洁公司企业香精研发部）；Heritage Science Laboratory, Faculty of Chemistry and Chemical Technology, University of Ljubljana（卢布尔雅那大学化学与化学技术学院遗产科学实验室）
Interpretable Machine Learning for Quantum-Informed Property Predictions in Artificial Sensing Materials
Institute for Materials Science（材料科学研究所）；Max Bergmann Center for Biomaterials（生物材料中心）；TUD Dresden University of Technology（德累斯顿技术大学）；Institute of Physical Chemistry（物理化学研究所）；Friedrich Schiller University Jena（耶拿弗里德里希·施莱尔大学）；Dresden Center for Computational Materials Science (DCMS)（德累斯顿计算材料科学中心）；Cluster of Excellence CARE（卓越中心CARE）；Cluster of Excellence CeTI（卓越中心CeTI）
On the Interplay of Subset Selection and Informed Graph Neural Networks
Fraunhofer SCAI（弗劳恩霍夫算法与科学计算研究所）；Universität Bonn（波恩大学）；University of Tromsø（特罗姆瑟大学）；Ludwig-Maximilians-Universität München（慕尼黑大学）；Technische Universität Braunschweig（布伦瑞克工业大学）
The use of variational nuclear motion programs to compute line lists of transition frequencies and intensities is now a standard procedure. The ExoMol project has used this technique to generate line lists for studies of hot bodies such as the atmospheres of exoplanets and cool stars. The resulting line list can be huge: many contain 10 billion or more transitions. This software update considers changes made to our programs during the course of the project to allow for such calculations. This update considers three programs: {\sc Duo} which computed vibronic spectra for diatomics, {\sc DVR3D} …
A cluster dynamics (CD) model has been developed to investigate the nucleation and growth of point defect clusters, i.e., interstitial prismatic loops and nanoscale and sub-nanoscale voids, in ThO2 during irradiation by energetic particles. The model considers cluster off-stoichiometry due to the asymmetry of point defect generation on the O and Th sublattices under irradiation, as well as the point defect diffusivities and the defect binding energies to clusters. The energies were established using detailed molecular dynamics simulations considering the statistical variability of cluster conf…
Halogen bonding has emerged as an important noncovalent interaction in a myriad of applications, including drug design, supramolecular assembly, and catalysis. Current understanding of the halogen bond is informed by electronic structure calculations on isolated molecules and/or crystal structures that are not readily transferable to liquids and disordered phases. To address this issue, we present a first-principles simulation-based approach for quantifying halogen bonds in molecular systems rooted in an understanding of nuclei-nuclei and electron-nuclei spatial correlations. We then demonstra…
