skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Machine learning predictions of molecular properties: Accurate many-body potentials and nonlocality in chemical space

Journal Article · · Journal of Physical Chemistry Letters
 [1];  [2];  [3];  [2];  [4];  [5];  [1]
  1. Max-Planck-Gesellschaft, Berlin (Germany)
  2. Techincal Univ. of Berlin, Berlin (Germany)
  3. Univ. of Basel, Basel (Switzerland)
  4. Univ. of Basel, Basel (Switzerland); Argonne National Lab. (ANL), Argonne, IL (United States)
  5. Techincal Univ. of Berlin, Berlin (Germany); Korea Univ., Seoul (Korea)
+ Show Author Affiliations

Simultaneously accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient toward rational compound design in chemical and pharmaceutical industries. Aiming toward this goal, we develop and apply a systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules. These methods range from a simple sum over atoms, to addition of bond energies, to pairwise interatomic force fields, reaching to the more sophisticated machine learning approaches that are capable of describing collective interactions between many atoms or bonds. In the case of equilibrium molecular geometries, even simple pairwise force fields demonstrate prediction accuracy comparable to benchmark energies calculated using density functional theory with hybrid exchange-correlation functionals; however, accounting for the collective many-body interactions proves to be essential for approaching the “holy grail” of chemical accuracy of 1 kcal/mol for both equilibrium and out-of-equilibrium geometries. This remarkable accuracy is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space. The same representation allows us to predict accurate electronic properties of molecules, such as their polarizability and molecular frontier orbital energies.

Research Organization:
Argonne National Laboratory (ANL), Argonne, IL (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
Grant/Contract Number:
AC02-06CH11357; NSF PP00P2_138932
OSTI ID:
1221601
Journal Information:
Journal of Physical Chemistry Letters, Vol. 6, Issue 12; ISSN 1948-7185
Publisher:
American Chemical SocietyCopyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 495 works
Citation information provided by
Web of Science

References (20)

Chemical space journal December 2004
Using neural networks to represent potential surfaces as sums of products journal November 2006
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Atom-centered symmetry functions for constructing high-dimensional neural network potentials journal February 2011
Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons journal April 2010
Prediction of Intramolecular Polarization of Aromatic Amino Acids Using Kriging Machine Learning journal July 2014
Virtual Exploration of the Small-Molecule Chemical Universe below 160 Daltons journal February 2005
Virtual Exploration of the Chemical Universe up to 11 Atoms of C, N, O, F:  Assembly of 26.4 Million Structures (110.9 Million Stereoisomers) and Analysis for New Ring Systems, Stereochemistry, Physicochemical Properties, Compound Classes, and Drug Discovery journal January 2007
The Blue Obelisk—Interoperability in Chemical Informatics journal May 2006
Generalized Gradient Approximation Made Simple journal October 1996
Rationale for mixing exact exchange with density functional approximations journal December 1996
Robust and Affordable Multicoefficient Methods for Thermochemistry and Thermochemical Kinetics:  The MCCM/3 Suite and SAC/3 journal May 2003
Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization Energies journal July 2013
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning journal January 2012
III - Bond energies journal September 1965
Text categorization with Support Vector Machines: Learning with many relevant features book January 1998
Homometric Structures journal June 1939
Machine learning of molecular electronic properties in chemical compound space journal September 2013
Quantum chemistry structures and properties of 134 kilo molecules journal August 2014
Quantum chemistry structures and properties of 134 kilo molecules text January 2014

Cited By (167)

Prediction of the Atomization Energy of Molecules Using Coulomb Matrix and Atomic Composition in a Bayesian Regularized Neural Networks
  • Tchagang, Alain B.; Valdés, Julio J.; Tetko, Igor V.
  • Artificial Neural Networks and Machine Learning – ICANN 2019: Workshop and Special Sessions: 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, p. 793-803 https://doi.org/10.1007/978-3-030-30493-5_75
book September 2019
A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility journal February 2020
Simulation and design of energy materials accelerated by machine learning journal June 2019
A new kind of atlas of zeolite building blocks journal October 2019
wACSF—Weighted atom-centered symmetry functions as descriptors in machine learning potentials journal June 2018
Size‐Extensive Molecular Machine Learning with Global Representations journal February 2020
Reactive molecular dynamics: From small molecules to proteins journal August 2018
Prediction of atomization energy using graph kernel and active learning journal January 2019
Hierarchical modeling of molecular energies using a deep neural network journal June 2018
Learning from the density to correct total energy and forces in first principle simulations journal October 2019
Molecular force fields with gradient-domain machine learning: Construction and application to dynamics of small molecules with coupled cluster forces journal March 2019
Understanding machine-learned density functionals: Understanding Machine-Learned Density Functionals journal November 2015
Mapping and classifying molecules from a high-throughput structural database journal February 2017
Band Gap Prediction for Large Organic Crystal Structures with Machine Learning journal July 2019
Improving the accuracy of Møller-Plesset perturbation theory with neural networks journal October 2017
Electronic transport of organic-inorganic hybrid perovskites from first-principles and machine learning journal February 2019
Machine learning of molecular properties: Locality and active learning journal June 2018
Chemical diversity in molecular orbital energy predictions with kernel ridge regression journal May 2019
Quantifying Chemical Structure and Machine‐Learned Atomic Energies in Amorphous and Liquid Silicon journal April 2019
Towards exact molecular dynamics simulations with machine-learned force fields journal September 2018
Machine learning-based screening of complex molecules for polymer solar cells journal June 2018
From DFT to machine learning: recent approaches to materials science–a review journal May 2019
Perspective: Machine learning potentials for atomistic simulations journal November 2016
A reactive, scalable, and transferable model for molecular energies from a neural network approach based on local information journal June 2018
Structure-based sampling and self-correcting machine learning for accurate calculations of potential energy surfaces and vibrational levels journal June 2017
Machine-learned electron correlation model based on correlation energy density at complete basis set limit journal July 2019
ANI-1, A data set of 20 million calculated off-equilibrium conformations for organic molecules journal December 2017
Supervised machine learning of ultracold atoms with speckle disorder journal April 2019
Neural-network-enhanced evolutionary algorithm applied to supported metal nanoparticles journal May 2018
Human versus Robots in the Discovery and Crystallization of Gigantic Polyoxometalates journal August 2017
Making machine learning a useful tool in the accelerated discovery of transition metal complexes journal July 2019
Electronic spectra from TDDFT and machine learning in chemical space journal August 2015
A new control method based on fuzzy controller, time delay estimation, deep learning, and non-dominated sorting genetic algorithm-III for powertrain mount system journal December 2019
A Critical Review of Machine Learning of Energy Materials journal January 2020
Processing Optimization and Property Predictions of Hot‐Extruded Bi–Te–Se Thermoelectric Materials via Machine Learning journal November 2019
Machine learning model for non-equilibrium structures and energies of simple molecules journal January 2019
Machine learning of accurate energy-conserving molecular force fields journal May 2017
Compressing physics with an autoencoder: Creating an atomic species representation to improve machine learning models in the chemical sciences journal August 2019
Metadynamics for training neural network model chemistries: A competitive assessment journal June 2018
Machine Learning a General-Purpose Interatomic Potential for Silicon journal December 2018
Quantum-chemical insights from deep tensor neural networks journal January 2017
FCHL revisited: Faster and more accurate quantum machine learning journal January 2020
Quantifying Chemical Structure and Machine-Learned Atomic Energies in Amorphous and Liquid Silicon journal April 2019
Genarris: Random generation of molecular crystal structures and fast screening with a Harris approximation journal June 2018
Machine learning modeling of Wigner intracule functionals for two electrons in one-dimension journal April 2019
Crowd-sourcing materials-science challenges with the NOMAD 2018 Kaggle competition journal November 2019
Alchemical and structural distribution based representation for universal quantum machine learning journal June 2018
New hybrid between SPEA/R with deep neural network: Application to predicting the multi-objective optimization of the stiffness parameter for powertrain mount systems journal August 2019
Communication: Understanding molecular representations in machine learning: The role of uniqueness and target similarity journal October 2016
Data‐Driven Materials Science: Status, Challenges, and Perspectives journal September 2019
Gaussian process regression for geometry optimization journal March 2018
Bypassing the Kohn-Sham equations with machine learning journal October 2017
What makes Data Science different? A discussion involving Statistics2.0 and Computational Sciences journal February 2018
Inverse molecular design using machine learning: Generative models for matter engineering journal July 2018
Machine learning in materials science journal August 2019
Machine Learning in Materials Science book January 2016
A neural network protocol for electronic excitations of N -methylacetamide journal May 2019
BAND NN: A Deep Learning Framework for Energy Prediction and Geometry Optimization of Organic Small Molecules journal December 2019
Machine-learned approximations to Density Functional Theory Hamiltonians journal February 2017
Deep Learning Spectroscopy: Neural Networks for Molecular Excitation Spectra journal January 2019
The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics journal January 2018
Universal nanohydrophobicity predictions using virtual nanoparticle library journal January 2019
Applying machine learning techniques to predict the properties of energetic materials journal June 2018
First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems journal August 2017
Constructing convex energy landscapes for atomistic structure optimization journal December 2019
Machine learning and artificial neural network accelerated computational discoveries in materials science journal November 2019
SchNet – A deep learning architecture for molecules and materials journal June 2018
Dataset’s chemical diversity limits the generalizability of machine learning predictions journal November 2019
Enumeration of de novo inorganic complexes for chemical discovery and machine learning journal January 2020
Representing molecular and materials data for unsupervised machine learning journal April 2018
ChemML : A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data journal January 2020
Physics-informed machine learning for inorganic scintillator discovery journal June 2018
TGMin: An efficient global minimum searching program for free and surface‐supported clusters journal December 2018
Chemical machine learning with kernels: The impact of loss functions journal February 2019
Machine-Learning-Assisted De Novo Design of Organic Molecules and Polymers: Opportunities and Challenges journal January 2020
Classification of spatially resolved molecular fingerprints for machine learning applications and development of a codebase for their implementation journal January 2018
Solid harmonic wavelet scattering for predictions of molecule properties journal June 2018
Constant size descriptors for accurate machine learning models of molecular properties journal June 2018
Accelerating atomic structure search with cluster regularization journal June 2018
Data‐Driven Materials Science: Status, Challenges, and Perspectives journal November 2019
Human versus Robots in the Discovery and Crystallization of Gigantic Polyoxometalates journal August 2017
Python for Scientific Computing journal January 2007
Machine learning of accurate energy-conserving molecular force fields text January 2017
ANI-1, A data set of 20 million calculated off-equilibrium conformations for organic molecules text January 2017
Development of a machine learning potential for graphene text January 2018
Machine Learning a General-Purpose Interatomic Potential for Silicon text January 2018
Applying Machine Learning Techniques to Predict the Properties of Energetic Materials posted_content February 2018
A reactive, scalable, and transferable model for molecular energies from a neural network approach based on local information text January 2018
Electronic spectra from TDDFT and machine learning in chemical space text January 2015
Communication: Understanding molecular representations in machine learning: The role of uniqueness and target similarity text January 2016
Machine learning meets volcano plots: Computational discovery of cross-coupling catalysts text January 2018
Quantum Machine Learning in Chemistry and Materials book January 2020
Quantifying Chemical Structure and Machine-Learned Atomic Energies in Amorphous and Liquid Silicon text January 2019
A reactive, scalable, and transferable model for molecular energies from a neural network approach based on local information text January 2018
Deep Learning Spectroscopy: Neural Networks for Molecular Excitation Spectra text January 2019
Reactive molecular dynamics: From small molecules to proteins text January 2018
Applying Machine Learning Techniques to Predict the Properties of Energetic Materials posted_content February 2018
Electronic Spectra from TDDFT and Machine Learning in Chemical Space text January 2015
Comparing molecules and solids across structural and alchemical space text January 2016
By-passing the Kohn-Sham equations with machine learning text January 2016
Machine-learned approximations to Density Functional Theory Hamiltonians preprint January 2016
Quantum-Chemical Insights from Deep Tensor Neural Networks text January 2016
Machine Learning of Accurate Energy-Conserving Molecular Force Fields text January 2016
Mapping and Classifying Molecules from a High-Throughput Structural Database preprint January 2016
Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks text January 2017
Accurate Force Field for Molybdenum by Machine Learning Large Materials Data text January 2017
The TensorMol-0.1 Model Chemistry: a Neural Network Augmented with Long-Range Physics preprint January 2017
WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials text January 2017
SchNet - a deep learning architecture for molecules and materials text January 2017
Metadynamics for Training Neural Network Model Chemistries: a Competitive Assessment text January 2017
Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface text January 2018
Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation text January 2018
Learning from the Density to Correct Total Energy and Forces in First Principle Simulations text January 2018
Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces text January 2019
Design and Analysis of Machine Learning Exchange-Correlation Functionals via Rotationally Invariant Convolutional Descriptors text January 2019
Data-driven materials science: status, challenges and perspectives text January 2019
FCHL revisited: faster and more accurate quantum machine learning text January 2019
Gaussian Process Regression for Geometry Optimization text January 2020
Bayesian molecular design with a chemical language model journal March 2017
Improving accuracy of interatomic potentials: more physics or more data? A case study of silica journal March 2019
Machine Learning Force Fields journal March 2021
PubChemQC PM6: Data Sets of 221 Million Molecules with Optimized Molecular Geometries and Electronic Properties journal October 2020
Transferable Dynamic Molecular Charge Assignment Using Deep Neural Networks journal July 2018
Machine Learning the Voltage of Electrode Materials in Metal-Ion Batteries journal April 2019
Learning a Local-Variable Model of Aromatic and Conjugated Systems journal December 2017
Aromatic molecules on low-index coinage metal surfaces: Many-body dispersion effects journal December 2016
Deep learning for UV absorption spectra with SchNarc: First steps toward transferability in chemical compound space journal October 2020
Mean-field density matrix decompositions journal December 2020
Machine learning with bond information for local structure optimizations in surface science journal December 2020
A mixed quantum chemistry/machine learning approach for the fast and accurate prediction of biochemical redox potentials and its large-scale application to 315,000 redox reactions journal April 2019
Molecular Property Prediction: A Multilevel Quantum Interactions Modeling Perspective journal July 2019
Deep Learning for Deep Chemistry: Optimizing the Prediction of Chemical Patterns journal November 2019
Localized Coulomb Descriptors for the Gaussian Approximation Potential preprint January 2016
Conditional molecular design with deep generative models text January 2018
Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes text January 2018
Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials preprint January 2018
Supervised machine learning of ultracold atoms with speckle disorder text January 2018
Compressing physical properties of atomic species for improving predictive chemistry text January 2018
Independent Vector Analysis for Data Fusion Prior to Molecular Property Prediction with Machine Learning preprint January 2018
Steerable Wavelet Scattering for 3D Atomic Systems with Application to Li-Si Energy Prediction preprint January 2018
Chemical diversity in molecular orbital energy predictions with kernel ridge regression text January 2018
An Atomistic Machine Learning Package for Surface Science and Catalysis preprint January 2019
DScribe: Library of Descriptors for Machine Learning in Materials Science text January 2019
Designing compact training sets for data-driven molecular property prediction preprint January 2019
PANNA: Properties from Artificial Neural Network Architectures text January 2019
Learning and Interpreting Potentials for Classical Hamiltonian Systems preprint January 2019
Gated Graph Recursive Neural Networks for Molecular Property Prediction preprint January 2019
Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights text January 2019
High-Dimensional Potential Energy Surfaces for Molecular Simulations text January 2019
Assisting human experts in the interpretation of their visual process: A case study on assessing copper surface adhesive potency preprint January 2019
Deep Learning for Optoelectronic Properties of Organic Semiconductors text January 2019
libmolgrid: GPU Accelerated Molecular Gridding for Deep Learning Applications preprint January 2019
Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches text January 2019
Autoencoding Undirected Molecular Graphs With Neural Networks preprint January 2020
Machine Learning in Materials Modeling -- Fundamentals and the Opportunities in 2D Materials preprint January 2020
Molecule Property Prediction and Classification with Graph Hypernetworks preprint January 2020
Incorporating electronic information into Machine Learning potential energy surfaces via approaching the ground-state electronic energy as a function of atom-based electronic populations preprint January 2020
QM7-X: A comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules text January 2020
The MLIP package: Moment Tensor Potentials with MPI and Active Learning preprint January 2020
On the role of gradients for machine learning of molecular energies and forces preprint January 2020
Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids preprint January 2020
Molecular machine learning with conformer ensembles preprint January 2020
Physics-inspired structural representations for molecules and materials text January 2021
Machine learning based energy-free structure predictions of molecules (closed and open-shell), transition states, and solids text January 2021
An automated approach for developing neural network interatomic potentials with FLAME preprint January 2021
Artificial Intelligence based Autonomous Molecular Design for Medical Therapeutic: A Perspective preprint January 2021
Improving Molecular Force Fields Across Configurational Space by Combining Supervised and Unsupervised Machine Learning text January 2021

Similar Records

Related Subjects

74 ATOMIC AND MOLECULAR PHYSICS
chemical compound space
machine learning
atomization energies
molecular properties
many-body potentials