Machine learning of molecular electronic properties in chemical compound space

The combination of modern scientific computing with electronic structure theory can lead to an unprecedented amount of data amenable to intelligent data analysis for the identification of meaningful, novel and predictive structure–property relationships. Such relationships enable high-throughput screening for relevant properties in an exponentially growing pool of virtual compounds that are synthetically accessible. Here, we present a machine learning model, trained on a database of ab initio calculation results for thousands of organic molecules, that simultaneously predicts multiple electronic ground- and excited-state properties. The properties include atomization energy, polarizability, frontier orbital eigenvalues, ionization potential, electron affinity and excitation energies. The machine learning model is based on a deep multi-task artificial neural network, exploiting the underlying correlations between various molecular properties. The input is identical to ab initio methods, i.e. nuclear charges and Cartesian coordinates of all atoms. For small organic molecules, the accuracy of such a ‘quantum machine’ is similar, and sometimes superior, to modern quantum-chemical methods—at negligible computational cost.

• Publication year

  2013

• Venue

  New Journal of Physics

• Publication date

  2013-05-30

• Fields of study

  Physics, Chemistry, Computer Science

• Identifiers
  DOI 10.1088/1367-2630/15/9/095003arXiv 1305.7074
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Learning Invariant Representations of Molecules for Atomization Energy Prediction

  2012cited by this paper
• First principles view on chemical compound space: Gaining rigorous atomistic control of molecular properties

  2012cited by this paper
• [Lecture Notes in Computer Science] Neural Networks: Tricks of the Trade Volume 7700 || Deep Learning via Semi-supervised Embedding

  2012cited by this paper
• Wavelet analysis in neurodynamics

  2012cited by this paper
• Resolution-of-identity approach to Hartree–Fock, hybrid density functionals, RPA, MP2 and GW with numeric atom-centered orbital basis functions

  2012cited by this paper
• Neural Networks: Tricks of the Trade

  2012cited by this paper
• Accurate and efficient method for many-body van der Waals interactions.

  2012influential reference
• Toward Quantitative Structure-Property Relationships for Charge Transfer Rates of Polycyclic Aromatic Hydrocarbons.

  2011cited by this paper
• Kernel Analysis of Deep Networks

  2011cited by this paper
• Intramolecular polarisable multipolar electrostatics from the machine learning method Kriging

  2011cited by this paper
• Support vector machine regression (LS-SVM)--an alternative to artificial neural networks (ANNs) for the analysis of quantum chemistry data?

  2011cited by this paper
• Fast and accurate modeling of molecular atomization energies with machine learning.

  2011cited by this paper
• Atom-centered symmetry functions for constructing high-dimensional neural network potentials.

  2011cited by this paper
• The Harvard Clean Energy Project: Large-Scale Computational Screening and Design of Organic Photovoltaics on the World Community Grid

  2011cited by this paper
• Virtual screening: an endless staircase?

  2010cited by this paper
• Finding Nature’s Missing Ternary Oxide Compounds Using Machine Learning and Density Functional Theory

  2010cited by this paper
• Potential energy surfaces fitted by artificial neural networks.

  2010cited by this paper
• Deep, Big, Simple Neural Nets for Handwritten Digit Recognition

  2010cited by this paper
• Finding Nature′s Missing Ternary Oxide Compounds Using Machine Learning and Density Functional Theory.

  2010cited by this paper
• Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons.

  2009cited by this paper
• Neural network approach to quantum-chemistry data: accurate prediction of density functional theory energies.

  2009cited by this paper
• 970 million druglike small molecules for virtual screening in the chemical universe database GDB-13.

  2009cited by this paper
• Ab initio molecular simulations with numeric atom-centered orbitals

  2009cited by this paper
• Dynamically Polarizable Water Potential Based on Multipole Moments Trained by Machine Learning.

  2009cited by this paper
• On Relevant Dimensions in Kernel Feature Spaces

  2008cited by this paper
• Generalized neural-network representation of high-dimensional potential-energy surfaces.

  2007cited by this paper
• 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

  2007cited by this paper
• Scaling learning algorithms towards AI

  2007cited by this paper
• Semiempirical Molecular Orbital Methods

  2007cited by this paper
• The Blue Obelisk—Interoperability in Chemical Informatics

  2006cited by this paper
• A nested molecule-independent neural network approach for high-quality potential fits.

  2006cited by this paper
• Computation of vertical excitation energies of retinal and analogs: Scope and limitations

  2006cited by this paper
• Virtual exploration of the small-molecule chemical universe below 160 Daltons.

  2005cited by this paper
• Hopping transport in conductive heterocyclic oligomers: reorganization energies and substituent effects.

  2005cited by this paper
• SIMILARITY OF MOLECULAR DESCRIPTORS : THE EQUIVALENCE OF ZAGREB INDICES AND WALK COUNTS

  2004cited by this paper
• Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks

  2004cited by this paper
• Tests of second-generation and third-generation density functionals for thermochemical kineticsElectronic supplementary information (ESI) available: Mean errors for pure and hybrid DFT methods. See http://www.rsc.org/suppdata/cp/b3/b316260e/

  2004cited by this paper
• The Signature Molecular Descriptor. 1. Using Extended Valence Sequences in QSAR and QSPR Studies

  2003cited by this paper
• Comparative assessment of a new nonempirical density functional: Molecules and hydrogen-bonded complexes

  2003cited by this paper
• Small Representative Benchmarks for Thermochemical Calculations

  2003cited by this paper
• A generalized exchange-correlation functional: the Neural-Networks approach ☆

  2003cited by this paper
• Combined first-principles calculation and neural-network correction approach for heat of formation

  2003cited by this paper
• Handbook of Molecular Descriptors

  2002cited by this paper
• Neural Networks: Tricks of the Trade

  2002cited by this paper
• GradientBased Learning Applied to Document Recognition

  2001cited by this paper
• A Model of Inductive Bias Learning

  2000cited by this paper
• QSAR Comparative Study of Wiener Descriptors for Weighted Molecular Graphs

  2000cited by this paper
• A Chemist's Guide to Density Functional Theory

  2000cited by this paper
• Assessment of Gaussian-3 and density functional theories for a larger experimental test set

  2000cited by this paper
• Assessment of the Perdew–Burke–Ernzerhof exchange-correlation functional

  1999influential reference
• Gradient-based learning applied to document recognition

  1998cited by this paper
• Multitask Learning

  1997cited by this paper
• Generalized Gradient Approximation Made Simple [Phys. Rev. Lett. 77, 3865 (1996)]

  1997cited by this paper
• Generalized Gradient Approximation Made Simple.

  1996cited by this paper
• Rationale for mixing exact exchange with density functional approximations

  1996cited by this paper
• UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations

  1992cited by this paper
• Acceleration of stochastic approximation by averaging

  1992cited by this paper
• Potential energy surfaces for macromolecules. A neural network technique

  1992cited by this paper
• Stochastic Gradient Learning in Neural Networks

  1991cited by this paper
• Approximation by superpositions of a sigmoidal function

  1989cited by this paper
• SMILES. 2. Algorithm for generation of unique SMILES notation

  1989cited by this paper
• Phoneme recognition using time-delay neural networks

  1989cited by this paper
• SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules

  1988cited by this paper
• Learning representations by back-propagating errors

  1986cited by this paper
• Learning representations by back-propagation errors, nature

  1986cited by this paper
• An intermediate neglect of differential overlap theory for transition metal complexes: Fe, Co and Cu chlorides

  1979cited by this paper
• Inhomogeneous Electron Gas

  1973cited by this paper
• An intermediate neglect of differential overlap technique for spectroscopy: Pyrrole and the azines

  1973cited by this paper
• Self-Consistent Equations Including Exchange and Correlation Effects

  1965cited by this paper
• NEW METHOD FOR CALCULATING THE ONE-PARTICLE GREEN'S FUNCTION WITH APPLICATION TO THE ELECTRON-GAS PROBLEM

  1965cited by this paper
• HARD AND SOFT ACIDS AND BASES

  1963cited by this paper
• Stochastic Gradient Learning in Neural Networks

  year unknowncited by this paper

• Transforming MOF Modeling with Machine-Learned Potentials: Progress and Perspectives.

  2026cites this paper
• Machine learning electronic structure and atomistic properties from the external potential

  2026cites this paper
• Cost-Effective Multi-Channel MolOrbImage for Machine-Learned Excited-State Properties of Practical Photofunctional Materials

  2026cites this paper
• On the use of chemical bonding descriptors in machine learning

  2026cites this paper
• Machine-Learning Framework for Excitation Energies of Chromophores in Polarizable Environments.

  2026cites this paper
• Optimal invariant sets for atomistic machine learning

  2026cites this paper
• A Comparison of Electronic Structure Methods for Predicting the Hydrogenation Energies of Candidate Molecules for Hydrogen Storage

  2025cites this paper
• Enhancing the Scalability and Applicability of Kohn-Sham Hamiltonians for Molecular Systems

  2025cites this paper
• Black-Box Uncertainty Estimation for Deep Learning Models in Atomistic Simulations

  2025cites this paper
• Multimodal Quantum Vision Transformer for Enzyme Commission Classification from Biochemical Representations

  2025cites this paper
• Electron-Informed Coarse-Graining Molecular Representation Learning for Real-World Molecular Physics

  2025cites this paper
• Artificial Intelligence-Driven Modeling for Hydrogel Three-Dimensional Printing: Computational and Experimental Cases of Study

  2025cites this paper
• Deciphering conductivity in PEDOT guided by machine learning: From solvent baths to charge paths

  2025cites this paper
• Acquisition of absorption and fluorescence spectral data using chatbots

  2025cites this paper
• Generalized convolutional many-body distribution functional representations

  2025cites this paper
• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie

  2025cites this paper
• Predicting Molecular Energies of Small Organic Molecules With Multi‐Fidelity Methods

  2025cites this paper
• Accelerating battery innovation: AI-powered molecular discovery.

  2025cites this paper
• Efficient solving of Schrödinger equation using deep convolutional neural network model with an attention mechanism and transfer learning

  2025cites this paper
• Machine learning accelerates Raman computations from molecular dynamics for materials science.

  2025cites this paper
• Information-theoretic multi-scale geometric pre-training for enhanced molecular property prediction

  2025cites this paper
• Q‐DFTNet: A Chemistry‐Informed Neural Network Framework for Predicting Molecular Dipole Moments via DFT‐Driven QM9 Data

  2025cites this paper
• A Dataset of Raman and Infrared Spectra as an Extension to the ChEMBL

  2025cites this paper
• Optimal Invariant Bases for Atomistic Machine Learning

  2025cites this paper
• Machine learning approach to predict L -edge x-ray absorption spectra of light transition metal ion compounds

  2025cites this paper
• Group Contribution Methods Designed using Machine Learning to Predict Enthalpies in Carboxylic Acids and Anhydrides

  2025cites this paper
• Functions on Symmetric Matrices and Point Clouds via Lightweight Invariant Features from Galois Theory

  2025cites this paper
• SchNet++: A Novel Framework Based on Dynamic Cutoff Radius & Multi-head Prediction for Molecule Energy Prediction

  2025cites this paper
• High throughput tight binding calculation of electronic HOMO–LUMO gaps and its prediction for natural compounds

  2025cites this paper
• The QCML dataset, Quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations

  2025cites this paper
• Combining molecular fragmentation and machine learning for accurate prediction of adiabatic ionization potentials

  2025cites this paper
• Benchmarking data efficiency in Δ-ML and multifidelity models for quantum chemistry.

  2025cites this paper
• Conceptual DFT Meets Machine Learning: A New Route to Enhanced Diels–Alder Reactivity

  2025cites this paper
• Quantum-inspired AI models for high-precision spectral property prediction

  2025cites this paper
• Adapting hybrid density functionals with machine learning

  2025cites this paper
• Lean CNNs for Mapping Electron Charge Density Fields to Material Properties

  2025cites this paper
• MORE-Q, a dataset for molecular olfactorial receptor engineering by quantum mechanics

  2025cites this paper
• Rapid high-fidelity quantum simulations using multi-step nonlinear autoregression and graph embeddings

  2025influential citation
• Machine Learning Prediction of Optical Properties of Coumarin Derivatives Using Gaussian-Weighted Graph Convolution and Subgraph Modular Input

  2025cites this paper
• Enhancing the prediction of TADF emitter properties using Δ-machine learning: A hybrid semi-empirical and deep tensor neural network approach.

  2025cites this paper
• IR Spectra for the EMIM-TFSI Ion Pair Using Deep Potentials.

  2025cites this paper
• Δ-Machine Learning of Polarizability Tensors Using a Dipole Interaction Model.

  2025cites this paper
• Gaussian Process Regression - Neural Network Hybrid with Optimized Redundant Coordinates

  2025cites this paper
• Chemical Space Exploration with Artificial "Mindless" Molecules

  2025cites this paper
• Structure generation and rolling strain prediction of layered composite materials based on Transformer model

  2025cites this paper
• LFaB: Low fidelity as Bias for Active Learning in the chemical configuration space

  2025influential citation
• Multi-View Graph Learning with Graph-Tuple

  2025cites this paper
• A Deep Learning Potential for Accurate Shock Response Simulations in Tin

  2025cites this paper
• Exploring the Design Space of Machine Learning Models for Quantum Chemistry with a Fully Differentiable Framework

  2025cites this paper
• Machine learning assisted approximation of descriptors (CO and OH) binding energy on Cu-based bimetallic alloys.

  2025cites this paper
• Survey on recent progress of AI for chemistry: methods, applications, and opportunities

  2025cites this paper
• Kernel regression methods for prediction of materials properties: Recent developments

  2025cites this paper
• Hybrid Unsupervised/Supervised Machine Learning for Identifying Molecular Structural Fingerprints From Ensemble Property

  2025cites this paper
• Effect of Molecular Structure on the B3LYP-Computed HOMO–LUMO Gap: A Structure −Property Relationship Using Atomic Signatures

  2025cites this paper
• A critical review on electronic materials properties and multifunctional applications

  2025cites this paper
• High-Fidelity Ring Fragments for Molecular Design and Spectroscopy: the PCS-LCB25-Nano-LEGO Framework.

  2025cites this paper
• Representative Random Sampling of Chemical Space

  2025cites this paper
• Quantum chemical properties of chlorinated polycyclic aromatic hydrocarbons for delta machine learning

  2025cites this paper
• Drug repurposing identifies novel Wee1 kinase inhibitors for triple negative breast cancer therapeutics.

  2025cites this paper
• tmQMg* Data Set: Excited State Properties of 74k Transition Metal Complexes

  2025cites this paper
• Leveraging natural language processing to curate the tmCAT, tmPHOTO, tmBIO, and tmSCO datasets of functional transition metal complexes.

  2024cites this paper
• Dataset for quantum-mechanical exploration of conformers and solvent effects in large drug-like molecules

  2024cites this paper
• Molecule Graph Networks with Many-Body Equivariant Interactions

  2024cites this paper
• A Comprehensive Graph Pooling Benchmark: Effectiveness, Robustness and Generalizability

  2024cites this paper
• QeMFi: A Multifidelity Dataset of Quantum Chemical Properties of Diverse Molecules

  2024cites this paper
• Predicting electronic screening for fast Koopmans spectral functional calculations

  2024cites this paper
• Advanced Design of Soft Robots with Artificial Intelligence

  2024cites this paper
• Quantum Mechanics Characterization of Non-Covalent Interaction in Nucleotide Fragments

  2024cites this paper
• Neural Polarization: Toward Electron Density for Molecules by Extending Equivariant Networks

  2024cites this paper
• Generalized Semiclassical Ehrenfest Method: A Route to Wave Function-Free Photochemistry and Nonadiabatic Dynamics with Only Potential Energies and Gradients.

  2024cites this paper
• QCDGE database, Quantum Chemistry Database with Ground- and Excited-state Properties of 450 Kilo Molecules

  2024cites this paper
• Quantum mechanical dataset of 836k neutral closed-shell molecules with up to 5 heavy atoms from C, N, O, F, Si, P, S, Cl, Br

  2024cites this paper
• Transfer Learning for Molecular Property Predictions from Small Data Sets

  2024cites this paper
• Response Matching for Generating Materials and Molecules

  2024cites this paper
• Deepmol: an automated machine and deep learning framework for computational chemistry

  2024cites this paper
• Hund’s matrix: A structure-based descriptor to predict the magnetic properties

  2024cites this paper
• X-ray-induced atomic transitions via machine learning: A computational investigation

  2024cites this paper
• Biomolecular dynamics with machine-learned quantum-mechanical force fields trained on diverse chemical fragments

  2024cites this paper
• Exact solution for minimization of root mean square deviation with G-RMSD to determine molecular similarity

  2024cites this paper
• Machine Learning in the Era of Smart Automation for Renewable Energy Materials

  2024cites this paper
• Novel machine learning insights into the QM7b and QM9 quantum mechanics datasets

  2024cites this paper
• Transfer learning with graph neural networks for improved molecular property prediction in the multi-fidelity setting

  2024cites this paper
• A review on the applications of graph neural networks in materials science at the atomic scale

  2024cites this paper
• Betulin: a novel triterpenoid anti-cancerous agent targeting cervical cancer through epigenetic proteins

  2024cites this paper
• Tandem SERS and MS/MS Profiling of Plasma Extracellular Vesicles for Early Ovarian Cancer Biomarker Discovery.

  2024cites this paper
• Energy Decomposition Scheme for Rectangular Graphene Flakes

  2024cites this paper
• Transfer Learning in Materials Informatics: structure-property relationships through minimal but highly informative multimodal input

  2024cites this paper
• Hybrid DFT Geometries and Properties for 17k Lanthanoid Complexes─The LnQM Data Set

  2024cites this paper
• A Smart Strategy for Photoresponsive Molecules: Utilizing Generative Pre-trained Transformer and TDDFT Calculations in Drug Delivery

  2024cites this paper
• Combining Transfer Learning and Representation Learning to Improve Predictive Analytics on Small Materials Data

  2024cites this paper
• QM9star, two Million DFT-computed Equilibrium Structures for Ions and Radicals with Atomic Information

  2024cites this paper
• A foundation machine learning potential with polarizable long-range interactions for materials modelling

  2024cites this paper
• The emergence of machine learning force fields in drug design

  2024cites this paper
• The Case for a Defect Genome Initiative

  2024cites this paper
• Many-body expansion based machine learning models for octahedral transition metal complexes

  2024cites this paper
• Extendibility of fermionic states and rigorous ground state approximations of interacting fermionic systems

  2024cites this paper
• Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks

  2024cites this paper
• Global reaction neural networks with embedded stoichiometry and thermodynamics for learning kinetics from reactor data

  2024cites this paper
• Automatic Prediction of Peak Optical Absorption Wavelengths in Molecules Using Convolutional Neural Networks

  2024cites this paper
• Benchmarking Data Efficiency in $\Delta$-ML and Multifidelity Models for Quantum Chemistry

  2024cites this paper