Search Menu
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information
Search Scholarly Publications

Title: Teaching a neural network to attach and detach electrons from molecules

Abstract

Interatomic potentials derived with Machine Learning algorithms such as Deep-Neural Networks (DNNs), achieve the accuracy of high-fidelity quantum mechanical (QM) methods in areas traditionally dominated by empirical force fields and allow performing massive simulations. Most DNN potentials were parametrized for neutral molecules or closed-shell ions due to architectural limitations. In this work, we propose an improved machine learning framework for simulating open-shell anions and cations. We introduce the AIMNet-NSE (Neural Spin Equilibration) architecture, which can predict molecular energies for an arbitrary combination of molecular charge and spin multiplicity with errors of about 2–3 kcal/mol and spin-charges with error errors ~0.01e for small and medium-sized organic molecules, compared to the reference QM simulations. The AIMNet-NSE model allows to fully bypass QM calculations and derive the ionization potential, electron affinity, and conceptual Density Functional Theory quantities like electronegativity, hardness, and condensed Fukui functions. We show that these descriptors, along with learned atomic representations, could be used to model chemical reactivity through an example of regioselectivity in electrophilic aromatic substitution reactions.

Authors:
; ORCiD logo; ORCiD logo; ORCiD logo; ORCiD logo
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
USDOE Office of Science (SC); USDOE Laboratory Directed Research and Development (LDRD) Program; National Science Foundation (NSF)
OSTI Identifier:
1812946
Alternate Identifier(s):
OSTI ID: 1828729; OSTI ID: 1844146
Report Number(s):
LA-UR-21-30591; LA-UR-20-24346
Journal ID: ISSN 2041-1723; 4870; PII: 24904
Grant/Contract Number:  
89233218CNA000001; CHE-1802789; CHE-2041108; CHE-200122; ACI-1053575; OAC-1818253
Resource Type:
Published Article
Journal Name:
Nature Communications
Additional Journal Information:
Journal Name: Nature Communications Journal Volume: 12 Journal Issue: 1; Journal ID: ISSN 2041-1723
Publisher:
Nature Publishing Group
Country of Publication:
United Kingdom
Language:
English
Subject:
36 MATERIALS SCIENCE; Material Science

Citation Formats

Zubatyuk, Roman, Smith, Justin S., Nebgen, Benjamin T., Tretiak, Sergei, and Isayev, Olexandr. Teaching a neural network to attach and detach electrons from molecules. United Kingdom: N. p., 2021. Web. doi:10.1038/s41467-021-24904-0.
Copy to clipboard
Zubatyuk, Roman, Smith, Justin S., Nebgen, Benjamin T., Tretiak, Sergei, & Isayev, Olexandr. Teaching a neural network to attach and detach electrons from molecules. United Kingdom. https://doi.org/10.1038/s41467-021-24904-0
Copy to clipboard
Zubatyuk, Roman, Smith, Justin S., Nebgen, Benjamin T., Tretiak, Sergei, and Isayev, Olexandr. Wed . "Teaching a neural network to attach and detach electrons from molecules". United Kingdom. https://doi.org/10.1038/s41467-021-24904-0.
Copy to clipboard
@article{osti_1812946,
title = {Teaching a neural network to attach and detach electrons from molecules},
author = {Zubatyuk, Roman and Smith, Justin S. and Nebgen, Benjamin T. and Tretiak, Sergei and Isayev, Olexandr},
abstractNote = {Interatomic potentials derived with Machine Learning algorithms such as Deep-Neural Networks (DNNs), achieve the accuracy of high-fidelity quantum mechanical (QM) methods in areas traditionally dominated by empirical force fields and allow performing massive simulations. Most DNN potentials were parametrized for neutral molecules or closed-shell ions due to architectural limitations. In this work, we propose an improved machine learning framework for simulating open-shell anions and cations. We introduce the AIMNet-NSE (Neural Spin Equilibration) architecture, which can predict molecular energies for an arbitrary combination of molecular charge and spin multiplicity with errors of about 2–3 kcal/mol and spin-charges with error errors ~0.01e for small and medium-sized organic molecules, compared to the reference QM simulations. The AIMNet-NSE model allows to fully bypass QM calculations and derive the ionization potential, electron affinity, and conceptual Density Functional Theory quantities like electronegativity, hardness, and condensed Fukui functions. We show that these descriptors, along with learned atomic representations, could be used to model chemical reactivity through an example of regioselectivity in electrophilic aromatic substitution reactions.},
doi = {10.1038/s41467-021-24904-0},
journal = {Nature Communications},
number = 1,
volume = 12,
place = {United Kingdom},
year = {Wed Aug 11 00:00:00 EDT 2021},
month = {Wed Aug 11 00:00:00 EDT 2021}
}
Copy to clipboard
Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
https://doi.org/10.1038/s41467-021-24904-0
Copyright Statement
Other availability
Search WorldCat Search WorldCat to find libraries that may hold this journal
Save / Share:
Export Metadata
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.

Works referenced in this record:

QTPIE: Charge transfer with polarization current equalization. A fluctuating charge model with correct asymptotics
journal, April 2007

A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer
journal, January 2021

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

Incorporating Electronic Information into Machine Learning Potential Energy Surfaces via Approaching the Ground-State Electronic Energy as a Function of Atom-Based Electronic Populations
journal, June 2020

  • Xie, Xiaowei; Persson, Kristin A.; Small, David W.
  • Journal of Chemical Theory and Computation, Vol. 16, Issue 7
  • DOI: 10.1021/acs.jctc.0c00217

UniChem: a unified chemical structure cross-referencing and identifier tracking system
journal, January 2013

  • Chambers, Jon; Davies, Mark; Gaulton, Anna
  • Journal of Cheminformatics, Vol. 5, Issue 1
  • DOI: 10.1186/1758-2946-5-3

ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost
journal, January 2017

  • Smith, J. S.; Isayev, O.; Roitberg, A. E.
  • Chemical Science, Vol. 8, Issue 4
  • DOI: 10.1039/C6SC05720A

The ORCA program system: The ORCA program system
journal, June 2011

  • Neese, Frank
  • Wiley Interdisciplinary Reviews: Computational Molecular Science, Vol. 2, Issue 1
  • DOI: 10.1002/wcms.81

Charge equilibration for molecular dynamics simulations
journal, April 1991

  • Rappe, Anthony K.; Goddard, William A.
  • The Journal of Physical Chemistry, Vol. 95, Issue 8
  • DOI: 10.1021/j100161a070

Deep Learning for Nonadiabatic Excited-State Dynamics
journal, November 2018

  • Chen, Wen-Kai; Liu, Xiang-Yang; Fang, Wei-Hai
  • The Journal of Physical Chemistry Letters, Vol. 9, Issue 23
  • DOI: 10.1021/acs.jpclett.8b03026

Electronegativity equalization: application and parametrization
journal, February 1985

  • Mortier, Wilfried J.; Van Genechten, Karin; Gasteiger, Johann
  • Journal of the American Chemical Society, Vol. 107, Issue 4
  • DOI: 10.1021/ja00290a017

A General Quantum Mechanically Derived Force Field (QMDFF) for Molecules and Condensed Phase Simulations
journal, September 2014

  • Grimme, Stefan
  • Journal of Chemical Theory and Computation, Vol. 10, Issue 10
  • DOI: 10.1021/ct500573f

Less is more: Sampling chemical space with active learning
journal, June 2018

  • Smith, Justin S.; Nebgen, Ben; Lubbers, Nicholas
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5023802

Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network
journal, August 2019

  • Zubatyuk, Roman; Smith, Justin S.; Leszczynski, Jerzy
  • Science Advances, Vol. 5, Issue 8
  • DOI: 10.1126/sciadv.aav6490

Machine learning of molecular electronic properties in chemical compound space
journal, September 2013

Intensive and extensive properties
journal, February 1970

SchNet – A deep learning architecture for molecules and materials
journal, June 2018

  • Schütt, K. T.; Sauceda, H. E.; Kindermans, P. -J.
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5019779

Nonadiabatic Excited-State Dynamics with Machine Learning
journal, September 2018

  • Dral, Pavlo O.; Barbatti, Mario; Thiel, Walter
  • The Journal of Physical Chemistry Letters, Vol. 9, Issue 19
  • DOI: 10.1021/acs.jpclett.8b02469

Philicity:  A Unified Treatment of Chemical Reactivity and Selectivity
journal, June 2003

  • Chattaraj, Pratim Kumar; Maiti, Buddhadev; Sarkar, Utpal
  • The Journal of Physical Chemistry A, Vol. 107, Issue 25
  • DOI: 10.1021/jp034707u

Fast and accurate prediction of the regioselectivity of electrophilic aromatic substitution reactions
journal, January 2018

  • Kromann, Jimmy C.; Jensen, Jan H.; Kruszyk, Monika
  • Chemical Science, Vol. 9, Issue 3
  • DOI: 10.1039/C7SC04156J

Quantum Chemistry in the Age of Machine Learning
journal, March 2020

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges
journal, April 2019

  • Unke, Oliver T.; Meuwly, Markus
  • Journal of Chemical Theory and Computation, Vol. 15, Issue 6
  • DOI: 10.1021/acs.jctc.9b00181

Machine Learning of Partial Charges Derived from High-Quality Quantum-Mechanical Calculations
journal, February 2018

  • Bleiziffer, Patrick; Schaller, Kay; Riniker, Sereina
  • Journal of Chemical Information and Modeling, Vol. 58, Issue 3
  • DOI: 10.1021/acs.jcim.7b00663

A Predictive Tool for Electrophilic Aromatic Substitutions Using Machine Learning
journal, October 2018

  • Tomberg, Anna; Johansson, Magnus J.; Norrby, Per-Ola
  • The Journal of Organic Chemistry, Vol. 84, Issue 8
  • DOI: 10.1021/acs.joc.8b02270

Capturing intensive and extensive DFT/TDDFT molecular properties with machine learning
journal, August 2018

  • Pronobis, Wiktor; Schütt, Kristof T.; Tkatchenko, Alexandre
  • The European Physical Journal B, Vol. 91, Issue 8
  • DOI: 10.1140/epjb/e2018-90148-y

Hierarchical modeling of molecular energies using a deep neural network
journal, June 2018

  • Lubbers, Nicholas; Smith, Justin S.; Barros, Kipton
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5011181

High-Throughput Screening Approach for the Optoelectronic Properties of Conjugated Polymers
journal, June 2018

  • Wilbraham, Liam; Berardo, Enrico; Turcani, Lukas
  • Journal of Chemical Information and Modeling, Vol. 58, Issue 12
  • DOI: 10.1021/acs.jcim.8b00256

Discovering a Transferable Charge Assignment Model Using Machine Learning
journal, July 2018

  • Sifain, Andrew E.; Lubbers, Nicholas; Nebgen, Benjamin T.
  • The Journal of Physical Chemistry Letters, Vol. 9, Issue 16
  • DOI: 10.1021/acs.jpclett.8b01939

Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
journal, January 2012

GFN2-xTB—An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions
journal, January 2019

  • Bannwarth, Christoph; Ehlert, Sebastian; Grimme, Stefan
  • Journal of Chemical Theory and Computation, Vol. 15, Issue 3
  • DOI: 10.1021/acs.jctc.8b01176

Prediction of organic homolytic bond dissociation enthalpies at near chemical accuracy with sub-second computational cost
journal, May 2020

The Pilot Way to Grid Resources Using glideinWMS
conference, March 2009

  • Sfiligoi, Igor; Bradley, Daniel C.; Holzman, Burt
  • 2009 WRI World Congress on Computer Science and Information Engineering
  • DOI: 10.1109/CSIE.2009.950

The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics
journal, January 2018

  • Yao, Kun; Herr, John E.; Toth, David W.
  • Chemical Science, Vol. 9, Issue 8
  • DOI: 10.1039/C7SC04934J

Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning
journal, July 2019

Transferable Dynamic Molecular Charge Assignment Using Deep Neural Networks
journal, July 2018

  • Nebgen, Benjamin; Lubbers, Nicholas; Smith, Justin S.
  • Journal of Chemical Theory and Computation, Vol. 14, Issue 9
  • DOI: 10.1021/acs.jctc.8b00524

Conceptual Density Functional Theory
journal, May 2003

  • Geerlings, P.; De Proft, F.; Langenaeker, W.
  • Chemical Reviews, Vol. 103, Issue 5
  • DOI: 10.1021/cr990029p

B97-3c: A revised low-cost variant of the B97-D density functional method
journal, February 2018

  • Brandenburg, Jan Gerit; Bannwarth, Christoph; Hansen, Andreas
  • The Journal of Chemical Physics, Vol. 148, Issue 6
  • DOI: 10.1063/1.5012601

Approximation capabilities of multilayer feedforward networks
journal, January 1991

Multitask prediction of site selectivity in aromatic C–H functionalization reactions
journal, January 2020

  • Struble, Thomas J.; Coley, Connor W.; Jensen, Klavs F.
  • Reaction Chemistry & Engineering, Vol. 5, Issue 5
  • DOI: 10.1039/D0RE00071J

Combining SchNet and SHARC: The SchNarc Machine Learning Approach for Excited-State Dynamics
journal, April 2020

  • Westermayr, Julia; Gastegger, Michael; Marquetand, Philipp
  • The Journal of Physical Chemistry Letters, Vol. 11, Issue 10
  • DOI: 10.1021/acs.jpclett.0c00527

Perspective on "Density functional approach to the frontier-electron theory of chemical reactivity"
journal, February 2000

  • Ayers, Paul W.; Levy, Mel
  • Theoretical Chemistry Accounts: Theory, Computation, and Modeling (Theoretica Chimica Acta), Vol. 103, Issue 3-4
  • DOI: 10.1007/s002149900093

Machine learning unifies the modeling of materials and molecules
journal, December 2017

  • Bartók, Albert P.; De, Sandip; Poelking, Carl
  • Science Advances, Vol. 3, Issue 12
  • DOI: 10.1126/sciadv.1701816

On the Foundations of Chemical Reactivity Theory
journal, March 2007

  • Cohen, Morrel H.; Wasserman, Adam
  • The Journal of Physical Chemistry A, Vol. 111, Issue 11
  • DOI: 10.1021/jp066449h

The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules
journal, May 2020

Computational Explorations of Mechanisms and Ligand-Directed Selectivities of Copper-Catalyzed Ullmann-Type Reactions
journal, May 2010

  • Jones, Gavin O.; Liu, Peng; Houk, K. N.
  • Journal of the American Chemical Society, Vol. 132, Issue 17
  • DOI: 10.1021/ja100739h

A Structure-Based Platform for Predicting Chemical Reactivity
journal, June 2020

Theory and modeling of stereoselective organic reactions
journal, March 1986

ChEMBL web services: streamlining access to drug discovery data and utilities
journal, April 2015

  • Davies, Mark; Nowotka, Michał; Papadatos, George
  • Nucleic Acids Research, Vol. 43, Issue W1
  • DOI: 10.1093/nar/gkv352

Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons
journal, April 2010

ChEMBL: towards direct deposition of bioassay data
journal, November 2018

  • Mendez, David; Gaulton, Anna; Bento, A. Patrícia
  • Nucleic Acids Research, Vol. 47, Issue D1
  • DOI: 10.1093/nar/gky1075

Chemical reactivity indexes in density functional theory
journal, January 1999

Quantum-chemical insights from deep tensor neural networks
journal, January 2017

  • Schütt, Kristof T.; Arbabzadah, Farhad; Chmiela, Stefan
  • Nature Communications, Vol. 8, Issue 1
  • DOI: 10.1038/ncomms13890

The Mass of the Electric Carrier in Copper, Silver and Aluminium
journal, February 1917

Frontier molecular orbital theory of cycloaddition reactions
journal, November 1975

Holistic prediction of enantioselectivity in asymmetric catalysis
journal, July 2019

Machine learning of accurate energy-conserving molecular force fields
journal, May 2017

  • Chmiela, Stefan; Tkatchenko, Alexandre; Sauceda, Huziel E.
  • Science Advances, Vol. 3, Issue 5
  • DOI: 10.1126/sciadv.1603015

Quantum chemical calculation of electron ionization mass spectra for general organic and inorganic molecules
journal, January 2017

  • Ásgeirsson, Vilhjálmur; Bauer, Christoph A.; Grimme, Stefan
  • Chemical Science, Vol. 8, Issue 7
  • DOI: 10.1039/C7SC00601B

Deep learning in neural networks: An overview
journal, January 2015

Machine learning for molecular and materials science
journal, July 2018

Extending the Applicability of the ANI Deep Learning Molecular Potential to Sulfur and Halogens
journal, June 2020

  • Devereux, Christian; Smith, Justin S.; Davis, Kate K.
  • Journal of Chemical Theory and Computation, Vol. 16, Issue 7
  • DOI: 10.1021/acs.jctc.0c00121
    Sort by:
    [ × clear filter / sort ]

    Similar Records in DOE PAGES and OSTI.GOV collections: