Learning Graph-Level Representation for Drug Discovery

Abstract

Predicating macroscopic influences of drugs on human body, like efficacy and toxicity, is a central problem of small-molecule based drug discovery. Molecules can be represented as an undirected graph, and we can utilize graph convolution networks to predication molecular properties. However, graph convolutional networks and other graph neural networks all focus on learning node-level representation rather than graph-level representation. Previous works simply sum all feature vectors for all nodes in the graph to obtain the graph feature vector for drug predication. In this paper, we introduce a dummy super node that is connected with all nodes in the graph by a directed edge as the representation of the graph and modify the graph operation to help the dummy super node learn graph-level feature. Thus, we can handle graph-level classification and regression in the same way as node-level classification and regression. In addition, we apply focal loss to address class imbalance in drug datasets. The experiments on MoleculeNet show that our method can effectively improve the performance of molecular properties predication.

arXiv admin note: text overlap with arXiv:1703.00564, arXiv:1611.03199 by other authors

References in corpus (8)

• Semi-Supervised Classification with Graph Convolutional Networks
• Spectral Networks and Locally Connected Networks on Graphs
• Focal Loss for Dense Object Detection
• Learning Convolutional Neural Networks for Graphs
• Gated Graph Sequence Neural Networks
• Massively Multitask Networks for Drug Discovery
• Atomic Convolutional Networks for Predicting Protein-Ligand Binding Affinity
• MoleculeNet: A Benchmark for Molecular Machine Learning

Cited by in corpus (15)

• Open Graph Benchmark: Datasets for Machine Learning on Graphs
• A Comprehensive Survey on Deep Graph Representation Learning
• Strategies for Pre-training Graph Neural Networks
• Do Transformers Really Perform Bad for Graph Representation?
• MR-GNN: Multi-Resolution and Dual Graph Neural Network for Predicting Structured Entity Interactions
• Molecule Attention Transformer
• Subgraph Neural Networks
• All SMILES Variational Autoencoder
• Backdoor Attacks to Graph Neural Networks
• Deep Generative Graph Distribution Learning for Synthetic Power Grids
• Graph convolutions that can finally model local structure
• An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design
• Sparse hierarchical representation learning on molecular graphs
• GABO: Graph Augmentations with Bi-level Optimization
• Edge but not Least: Cross-View Graph Pooling