Graph neural networks for materials science and chemistry

Published in

Communications Materials(2022)

External Links:

Generate Graph DownloadPDF

TL;DR

An overview of the basic principles of GNNs, widely used datasets, and state-of-the-art architectures are provided, followed by a discussion of a wide range of recent applications, and a possible road-map for their further development is indicated.

Abstract

Machine learning plays an increasingly important role in many areas of chemistry and materials science, being used to predict materials properties, accelerate simulations, design new structures, and predict synthesis routes of new materials. Graph neural networks (GNNs) are one of the fastest growing classes of machine learning models. They are of particular relevance for chemistry and materials science, as they directly work on a graph or structural representation of molecules and materials and therefore have full access to all relevant information required to characterize materials. In this Review, we provide an overview of the basic principles of GNNs, widely used datasets, and state-of-the-art architectures, followed by a discussion of a wide range of recent applications of GNNs in chemistry and materials science, and concluding with a road-map for the further development and application of GNNs. Graph neural networks are machine learning models that directly access the structural representation of molecules and materials. This Review discusses state-of-the-art architectures and applications of graph neural networks in materials science and chemistry, indicating a possible road-map for their further development.

Authors

Pascal Friederich

5 papers

Patrick Reiser

3 papers

Marlen Neubert

1 papers

References325 items

L2XGNN: learning to explain graph neural networks

TUCAN: A molecular identifier and descriptor applicable to the whole periodic table from hydrogen to oganesson

Efficient and interpretable graph network representation for angle-dependent properties applied to optical spectroscopy

On scientific understanding with artificial intelligence

SELFIES and the future of molecular string representations

Graph neural networks for materials science and chemistry

Published in

Communications Materials(2022)

External Links:

Generate Graph DownloadPDF

TL;DR

Abstract

Authors

Pascal Friederich

5 papers

Patrick Reiser

3 papers

Marlen Neubert

1 papers

References325 items

L2XGNN: learning to explain graph neural networks

TUCAN: A molecular identifier and descriptor applicable to the whole periodic table from hydrogen to oganesson

Efficient and interpretable graph network representation for angle-dependent properties applied to optical spectroscopy

On scientific understanding with artificial intelligence

SELFIES and the future of molecular string representations

Andr'e Eberhard

1 papers

Luca Torresi

2 papers

Houssam Metni

1 papers

Clint van Hoesel

1 papers

Henrik Schopmans

2 papers

Machine learning in scanning transmission electron microscopy

MolNet: A Chemically Intuitive Graph Neural Network for Prediction of Molecular Properties

Deep Generative Model for Periodic Graphs

FastFlows: Flow-Based Models for Molecular Graph Generation

Prediction and Interpretation of Polymer Properties Using the Graph Convolutional Network

Convergence Acceleration in Machine Learning Potentials for Atomistic Simulations

ML4CO: Is GCNN All You Need? Graph Convolutional Neural Networks Produce Strong Baselines For Combinatorial Optimization Problems, If Tuned and Trained Properly, on Appropriate Data

Δ-Quantum machine-learning for medicinal chemistry

Generative Adversarial Networks

Simulation Intelligence: Towards a New Generation of Scientific Methods

Crystal graph attention networks for the prediction of stable materials

Combining crystal graphs and domain knowledge in machine learning to predict metal-organic frameworks performance in methane adsorption

GNES: Learning to Explain Graph Neural Networks

Explainable graph neural networks for organic cages

DeepTMC: A Deep Learning Platform to Targeted Design Doped Transition Metal Compounds

Graph network based deep learning of bandgaps.

Crystal Diffusion Variational Autoencoder for Periodic Material Generation

Geometric and Physical Quantities improve E(3) Equivariant Message Passing

Atomistic Graph Neural Networks for metals: Application to bcc iron

Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties.

Predicting Drug ADMET Properties Based on Graph Attention Network

Phase–Property Diagrams for Multicomponent Oxide Systems toward Materials Libraries

A geometric-information-enhanced crystal graph network for predicting properties of materials

Correction to "XGraphBoost: Extracting Graph Neural Network-Based Features for a Better Prediction of Molecular Properties"

CNN-G: Convolutional Neural Network Combined With Graph for Image Segmentation With Theoretical Analysis

Predicting energy and stability of known and hypothetical crystals using graph neural network

SARS-Net: COVID-19 detection from chest x-rays by combining graph convolutional network and convolutional neural network

Excited state non-adiabatic dynamics of large photoswitchable molecules using a chemically transferable machine learning potential

NewtonNet: a Newtonian message passing network for deep learning of interatomic potentials and forces

Deep Retrosynthetic Reaction Prediction using Local Reactivity and Global Attention

MOF Synthesis Prediction Enabled by Automatic Data Mining and Machine Learning

Graph representational learning for bandgap prediction in varied perovskite crystals

De Novo Drug Design Using Reinforcement Learning with Graph-Based Deep Generative Models

Geometric deep learning on molecular representations

Edge Representation Learning with Hypergraphs

3DMol-Net: Learn 3D Molecular Representation Using Adaptive Graph Convolutional Network Based on Rotation Invariance

GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles

Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation

SE(3)-equivariant prediction of molecular wavefunctions and electronic densities

Atomistic Line Graph Neural Network for improved materials property predictions

GemNet: Universal Directional Graph Neural Networks for Molecules

How Attentive are Graph Attention Networks?

Machine-learned potentials for next-generation matter simulations

Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture

XGraphBoost: Extracting Graph Neural Network-Based Features for a Better Prediction of Molecular Properties

HamNet: Conformation-Guided Molecular Representation with Hamiltonian Neural Networks

SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects

Inverse design of glass structure with deep graph neural networks

Implementing graph neural networks with TensorFlow-Keras

Weisfeiler and Lehman Go Topological: Message Passing Simplicial Networks

Using graph convolutional neural networks to learn a representation for glycans

ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations

Persistent Message Passing

E(n) Equivariant Graph Neural Networks

Graph neural networks for automated de novo drug design.

A systematic comparison study on hyperparameter optimisation of graph neural networks for molecular property prediction

Equivariant message passing for the prediction of tensorial properties and molecular spectra

A deep learning framework for high-throughput mechanism-driven phenotype compound screening and its application to COVID-19 drug repurposing

Bayesian reaction optimization as a tool for chemical synthesis

MutagenPred-GCNNs: A Graph Convolutional Neural Network-Based Classification Model for Mutagenicity Prediction with Data-Driven Molecular Fingerprints

E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials

Symmetry-adapted graph neural networks for constructing molecular dynamics force fields

Anomalous collapses of Nares Strait ice arches leads to enhanced export of Arctic sea ice

Learning properties of ordered and disordered materials from multi-fidelity data

ResGNet-C: A graph convolutional neural network for detection of COVID-19

Graph Neural Networks for COVID-19 Drug Discovery

Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction.

A compact review of molecular property prediction with graph neural networks.

Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules

Comparison of Atom Representations in Graph Neural Networks for Molecular Property Prediction

Transfer learning with graph neural networks for optoelectronic properties of conjugated oligomers.

Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures

Accelerating Discovery of Metal-Organic Frameworks for Methane Adsorption with Hierarchical Screening and Deep Learning.

A general-purpose machine-learning force field for bulk and nanostructured phosphorus

Machine learning of solvent effects on molecular spectra and reactions

Scientific intuition inspired by machine learning-generated hypotheses

Structure-Based Synthesizability Prediction of Crystals Using Partially Supervised Learning.

The Open Catalyst 2020 (OC20) Dataset and Community Challenges

Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials

On the Universality of Rotation Equivariant Point Cloud Networks

Retrospective on a decade of machine learning for chemical discovery

Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties

Learning Graph Normalization for Graph Neural Networks

Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models

BonDNet: a graph neural network for the prediction of bond dissociation energies for charged molecules

Quantum machine learning using atom-in-molecule-based fragments selected on the fly

Machine Learning-Enabled Design of Point Defects in 2D Materials for Quantum and Neuromorphic Information Processing.

Regio-selectivity prediction with a machine-learned reaction representation and on-the-fly quantum mechanical descriptors

Pushing the boundaries of molecular representation for drug discovery with graph attention mechanism.

Graph networks for molecular design

Orbital Graph Convolutional Neural Network for Material Property Prediction

Materials Graph Transformer predicts the outcomes of inorganic reactions with reliable uncertainties

Coarse graining molecular dynamics with graph neural networks.

CCDCGAN: Inverse design of crystal structures

100

TUDataset: A collection of benchmark datasets for learning with graphs

101

OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Features

102

Deep learning for UV absorption spectra with SchNarc: First steps toward transferability in chemical compound space.

103

DRACON: disconnected graph neural network for atom mapping in chemical reactions.

104

Graph attention convolutional neural network model for chemical poisoning of honey bees' prediction.

105

Molecule Edit Graph Attention Network: Modeling Chemical Reactions as Sequences of Graph Edits

106

rdScaffoldNetwork: The Scaffold Network Implementation in RDKit

107

Graph neural network based coarse-grained mapping prediction

108

SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks

109

MoFlow: An Invertible Flow Model for Generating Molecular Graphs

110

Coupling a Crystal Graph Multilayer Descriptor to Active Learning for Rapid Discovery of 2D Ferromagnetic Semiconductors/Half‐Metals/Metals

111

Learning Graph Models for Retrosynthesis Prediction

112

RetroXpert: Decompose Retrosynthesis Prediction like a Chemist

113

GCIceNet: a graph convolutional network for accurate classification of water phases.

114

On the Bottleneck of Graph Neural Networks and its Practical Implications

115

Prediction of Drug–Target Interactions From Multi-Molecular Network Based on Deep Walk Embedding Model

116

Generating transition states of isomerization reactions with deep learning.

117

Message Passing Neural Networks for Partial Charge Assignment to Metal–Organic Frameworks

118

Graph Neural Networks for Prediction of Fuel Ignition Quality

119

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

120

A deep neural network for molecular wave functions in quasi-atomic minimal basis representation.

121

Inverse design of nanoporous crystalline reticular materials with deep generative models

122

Machine-enabled inverse design of inorganic solid materials: promises and challenges

123

Improvement in ADMET Prediction with Multitask Deep Featurization.

124

A Graph to Graphs Framework for Retrosynthesis Prediction

125

Uncertainty-Quantified Hybrid ML/DFT High Throughput Screening Method for Crystals.

126

Practical Deep-Learning Representation for Fast Heterogeneous Catalyst Screening.

127

Directional Message Passing for Molecular Graphs

128

Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy

129

Unveiling the predictive power of static structure in glassy systems

130

AIRSS data for carbon at 10GPa and the C+N+H+O system at 1GPa

131

Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Data

132

Neural Message Passing on High Order Paths

133

A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility

134

Molecule Attention Transformer

135

A Comprehensive Survey on Geometric Deep Learning

136

Combining SchNet and SHARC: The SchNarc Machine Learning Approach for Excited-State Dynamics

137

Deep Learning of Activation Energies

138

Introducing Machine Learning: Science and Technology

139

A critical examination of compound stability predictions from machine-learned formation energies

140

Building attention and edge message passing neural networks for bioactivity and physical–chemical property prediction

141

U1 snRNP regulates cancer cell migration and invasion in vitro

142

Retrosynthesis Prediction with Conditional Graph Logic Network

143

Predicting Retrosynthetic Reactions Using Self-Corrected Transformer Neural Networks

144

Normalizing Flows for Probabilistic Modeling and Inference

145

Exploring chemical compound space with quantum-based machine learning

146

Hierarchical Graph Pooling with Structure Learning

147

Graph convolutional networks: a comprehensive review

148

Advances, Updates, and Analytics for the Computation-Ready, Experimental Metal–Organic Framework Database: CoRE MOF 2019

149

Deep Learning for Optoelectronic Properties of Organic Semiconductors

150

Hyperbolic Graph Neural Networks

151

Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules

152

Predicting materials properties without crystal structure: deep representation learning from stoichiometry

153

Text-mined dataset of inorganic materials synthesis recipes

154

B-Spline CNNs on Lie Groups

155

Deep Learning for Automated Classification and Characterization of Amorphous Materials

156

FCHL revisited: Faster and more accurate quantum machine learning.

157

Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks.

158

Multitask prediction of site selectivity in aromatic C–H functionalization reactions

159

Molecule Property Prediction Based on Spatial Graph Embedding

160

Modeling Physico-Chemical ADMET Endpoints with Multitask Graph Convolutional Networks

161

Prediction and Interpretable Visualization of Retrosynthetic Reactions Using Graph Convolutional Networks

162

iPool - Information-based Pooling in Hierarchical Graph Neural Networks

163

Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions

164

Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery

165

Cormorant: Covariant Molecular Neural Networks

166

Graph convolutional networks for computational drug development and discovery

167

Next-Generation Experimentation with Self-Driving Laboratories

168

GraphNVP: An Invertible Flow Model for Generating Molecular Graphs

169

Provably Powerful Graph Networks

170

Edge Contraction Pooling for Graph Neural Networks

171

Automatic Retrosynthetic Pathway Planning Using Template-free Models

172

Materials property prediction using symmetry-labeled graphs as atomic position independent descriptors

174

Graph Convolutional Networks with EigenPooling

175

Exploiting machine learning for end-to-end drug discovery and development

176

Self-Attention Graph Pooling

177

PubChemQC PM6: Data Sets of 221 Million Molecules with Optimized Molecular Geometries and Electronic Properties

178

Applications of machine learning in drug discovery and development

179

Analyzing Learned Molecular Representations for Property Prediction

180

GNNExplainer: Generating Explanations for Graph Neural Networks

181

Fast Graph Representation Learning with PyTorch Geometric

182

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges.

183

LanczosNet: Multi-Scale Deep Graph Convolutional Networks

184

Graph Transformation Policy Network for Chemical Reaction Prediction

185

Graph Neural Networks: A Review of Methods and Applications

186

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

187

Machine Learning of Coarse-Grained Molecular Dynamics Force Fields

188

Graph Convolutional Neural Networks for Polymers Property Prediction

189

Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction

190

A graph-convolutional neural network model for the prediction of chemical reactivity

191

Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks

192

How Powerful are Graph Neural Networks?

193

Relational Graph Attention Networks

194

Capsule Graph Neural Network

195

Hypergraph Neural Networks

196

SphereNet: Learning Spherical Representations for Detection and Classification in Omnidirectional Images

197