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

Title: Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network

Journal Article · · Science Advances
ORCiD logo [1]; ORCiD logo [2]; ORCiD logo [3]; ORCiD logo [4]
  1. Univ. of North Carolina, Chapel Hill, NC (United States); Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Jackson State Univ., Jackson, MS (United States)
  2. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
  3. Jackson State Univ., Jackson, MS (United States)
  4. Univ. of North Carolina, Chapel Hill, NC (United States)
+ Show Author Affiliations

Atomic and molecular properties could be evaluated from the fundamental Schrodinger’s equation and therefore represent different modalities of the same quantum phenomena. Here, we present AIMNet, a modular and chemically inspired deep neural network potential. We used AIMNet with multitarget training to learn multiple modalities of the state of the atom in a molecular system. The resulting model shows on several benchmark datasets state-of-the-art accuracy, comparable to the results of orders of magnitude more expensive DFT methods. It can simultaneously predict several atomic and molecular properties without an increase in the computational cost. With AIMNet, we show a new dimension of transferability: the ability to learn new targets using multimodal information from previous training. The model can learn implicit solvation energy (SMD method) using only a fraction of the original training data and an archive median absolute deviation error of 1.1 kcal/mol compared to experimental solvation free energies in the MNSol database.

Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
Grant/Contract Number:
89233218CNA000001
OSTI ID:
1570635
Report Number(s):
LA-UR-18-29299
Journal Information:
Science Advances, Vol. 5, Issue 8; ISSN 2375-2548
Publisher:
AAASCopyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 118 works
Citation information provided by
Web of Science

References (45)

Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning journal January 2012
The Pilot Way to Grid Resources Using glideinWMS conference March 2009
Inverse molecular design using machine learning: Generative models for matter engineering journal July 2018
Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning journal July 2019
Improving solvation energy predictions using the SMD solvation method and semiempirical electronic structure methods journal September 2018
The ORCA program system: The ORCA program system journal June 2011
Less is more: Sampling chemical space with active learning journal June 2018
Predicting electronic structure properties of transition metal complexes with neural networks journal January 2017
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Toward Accurate Conformational Energies of Smaller Peptides and Medium-Sized Macrocycles: MPCONF196 Benchmark Energy Data Set journal February 2018
Brain mechanisms linking language and action journal June 2005
A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians journal October 2018
A Comparison of Quantum and Molecular Mechanical Methods to Estimate Strain Energy in Druglike Fragments journal May 2017
Transferability in Machine Learning for Electronic Structure via the Molecular Orbital Basis journal July 2018
SchNet – A deep learning architecture for molecules and materials journal June 2018
Exploring the use of adaptive gradient methods in effective deep learning systems conference April 2018
The open science grid journal July 2007
Accelerating Chemical Discovery with Machine Learning: Simulated Evolution of Spin Crossover Complexes with an Artificial Neural Network journal February 2018
Minimal Basis Iterative Stockholder: Atoms in Molecules for Force-Field Development journal July 2016
Machine learning of accurate energy-conserving molecular force fields journal May 2017
First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems journal August 2017
Machine learning for molecular and materials science journal July 2018
Planning chemical syntheses with deep neural networks and symbolic AI journal March 2018
Multitask Learning journal January 1997
An overlap fitted chain of spheres exchange method journal October 2011
Active learning of linearly parametrized interatomic potentials journal December 2017
Discovering a Transferable Charge Assignment Model Using Machine Learning journal July 2018
RPLS-Net: pulmonary lobe segmentation based on 3D fully convolutional networks and multi-task learning journal April 2021
Machine learning of accurate energy-conserving molecular force fields text January 2017
ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost text January 2017
Imitation of Life: Structure, Agency and Discourse in Theatrical Performance journal July 2012
Fast and accurate modeling of molecular atomization energies with machine learning text January 2012
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning text January 2011
Minimal Basis Iterative Stockholder: Atoms in Molecules for Force-Field Development text January 2016
Machine Learning of Accurate Energy-Conserving Molecular Force Fields text January 2016
Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks text January 2017
The TensorMol-0.1 Model Chemistry: a Neural Network Augmented with Long-Range Physics preprint January 2017
SchNet - a deep learning architecture for molecules and materials text January 2017
Transferability in Machine Learning for Electronic Structure via the Molecular Orbital Basis text January 2018
Correction to “A Comparison of Quantum and Molecular Mechanical Methods to Estimate Strain Energy in Druglike Fragments” journal March 2019
AI designs organic syntheses journal March 2018
The S66x8 benchmark for noncovalent interactions revisited: explicitly correlated ab initio methods and density functional theory journal January 2016
The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics journal January 2018
Active learning of linearly parametrized interatomic potentials text January 2016
A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians preprint January 2018

Cited By (4)

Neural Network Potentials: A Concise Overview of Methods journal April 2022
A Scalable Molecular Force Field Parameterization Method Based on Density Functional Theory and Quantum-Level Machine Learning preprint January 2019
Deep Learning in Protein Structural Modeling and Design preprint January 2020
Embedding quantum statistical excitations in a classical force field text January 2021

Figures / Tables (6)

Fig. 1(p. 3)figure Fig. 1
Fig. 2(p. 5)figure Fig. 2
Fig. 3(p. 5)figure Fig. 3
Fig. 4(p. 6)figure Fig. 4
Fig. 5(p. 6)figure Fig. 5
Fig. 6(p. 7)figure Fig. 6

Similar Records

Teaching a neural network to attach and detach electrons from molecules
Journal Article · Wed Aug 11 00:00:00 EDT 2021 · Nature Communications · OSTI ID:1570635
+2 more
Graphical Gaussian process regression model for aqueous solvation free energy prediction of organic molecules in redox flow batteries
Journal Article · Thu Oct 28 00:00:00 EDT 2021 · Physical Chemistry Chemical Physics. PCCP · OSTI ID:1570635
+5 more
Graph-Based Approaches for Predicting Solvation Energy in Multiple Solvents: Open Datasets and Machine Learning Models
Journal Article · Wed Jun 30 00:00:00 EDT 2021 · Journal of Physical Chemistry. A, Molecules, Spectroscopy, Kinetics, Environment, and General Theory · OSTI ID:1570635
+5 more

Related Subjects

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY