Graph Neural Networks for Molecules
arXiv:2209.05582 · doi:10.1007/978-3-031-37196-7_2
Abstract
Graph neural networks (GNNs), which are capable of learning representations from graphical data, are naturally suitable for modeling molecular systems. This review introduces GNNs and their various applications for small organic molecules. GNNs rely on message-passing operations, a generic yet powerful framework, to update node features iteratively. Many researches design GNN architectures to effectively learn topological information of 2D molecule graphs as well as geometric information of 3D molecular systems. GNNs have been implemented in a wide variety of molecular applications, including molecular property prediction, molecular scoring and docking, molecular optimization and de novo generation, molecular dynamics simulation, etc. Besides, the review also summarizes the recent development of self-supervised learning for molecules with GNNs.
arXiv commentA chapter for the book "Machine Learning in Molecular Sciences". 31 pages, 4 figures
References in corpus (42)
- The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
- Geometric deep learning: going beyond Euclidean data
- Automatic chemical design using a data-driven continuous representation of molecules
- Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
- ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
- Molecular Graph Convolutions: Moving Beyond Fingerprints
- Quantum-Chemical Insights from Deep Tensor Neural Networks
- Modeling polypharmacy side effects with graph convolutional networks
- Machine Learning of Accurate Energy-Conserving Molecular Force Fields
- PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
- Machine learning for molecular simulation
- Molecular Contrastive Learning of Representations via Graph Neural Networks
- GuacaMol: Benchmarking Models for De Novo Molecular Design
- Machine Learning Unifies the Modelling of Materials and Molecules
- Deep learning for molecular design - a review of the state of the art
- Self-Referencing Embedded Strings (SELFIES): A 100% robust molecular string representation
- Machine Learning of Molecular Electronic Properties in Chemical Compound Space
- ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction
- DeepAffinity: Interpretable Deep Learning of Compound-Protein Affinity through Unified Recurrent and Convolutional Neural Networks
- Deep neural network solution of the electronic Schrödinger equation
- ANI-1: A data set of 20M off-equilibrium DFT calculations for organic molecules
- Hierarchical modeling of molecular energies using a deep neural network
- Mol-CycleGAN - a generative model for molecular optimization
- OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Features
- Machine learning force fields and coarse-grained variables in molecular dynamics: application to materials and biological systems
- MoFlow: An Invertible Flow Model for Generating Molecular Graphs
- Conditional molecular design with deep generative models
- Inverse design of 3d molecular structures with conditional generative neural networks
- SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization
- Molecular graph generation with Graph Neural Networks
- Orbital Graph Convolutional Neural Network for Material Property Prediction
- Scaffold-based molecular design using graph generative model
- Molecular geometry prediction using a deep generative graph neural network
- A Deep Generative Model for Molecule Optimization via One Fragment Modification
- Combining docking pose rank and structure with deep learning improves protein-ligand binding mode prediction
- Deep Reinforcement Learning Optimizes Graphene Nanopores for Efficient Desalination
- Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrast
- Three-Dimensionally Embedded Graph Convolutional Network (3DGCN) for Molecule Interpretation
- Graph Neural Networks Accelerated Molecular Dynamics
- KekuleScope: prediction of cancer cell line sensitivity and compound potency using convolutional neural networks trained on compound images
- Fast Neural Network Approach for Direct Covariant Forces Prediction in Complex Multi-Element Extended Systems
- Denoise Pretraining on Nonequilibrium Molecules for Accurate and Transferable Neural Potentials