Hierarchical Generation of Molecular Graphs using Structural Motifs

Abstract

Graph generation techniques are increasingly being adopted for drug discovery. Previous graph generation approaches have utilized relatively small molecular building blocks such as atoms or simple cycles, limiting their effectiveness to smaller molecules. Indeed, as we demonstrate, their performance degrades significantly for larger molecules. In this paper, we propose a new hierarchical graph encoder-decoder that employs significantly larger and more flexible graph motifs as basic building blocks. Our encoder produces a multi-resolution representation for each molecule in a fine-to-coarse fashion, from atoms to connected motifs. Each level integrates the encoding of constituents below with the graph at that level. Our autoregressive coarse-to-fine decoder adds one motif at a time, interleaving the decision of selecting a new motif with the process of resolving its attachments to the emerging molecule. We evaluate our model on multiple molecule generation tasks, including polymers, and show that our model significantly outperforms previous state-of-the-art baselines.

References in corpus (22)

• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Spectral Networks and Locally Connected Networks on Graphs
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Effective Approaches to Attention-based Neural Machine Translation
• Discriminative Embeddings of Latent Variable Models for Structured Data
• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• MolGAN: An implicit generative model for small molecular graphs
• Learning Deep Generative Models of Graphs
• Gated Graph Sequence Neural Networks
• How Powerful are Graph Neural Networks?
• Constrained Graph Variational Autoencoders for Molecule Design
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
• Syntax-Directed Variational Autoencoder for Structured Data
• Conditional molecular design with deep generative models
• Efficient Graph Generation with Graph Recurrent Attention Networks
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization
• Grammar Variational Autoencoder
• Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders
• Generating Focussed Molecule Libraries for Drug Discovery with Recurrent Neural Networks

Cited by in corpus (11)

• Molecular Contrastive Learning of Representations via Graph Neural Networks
• Molecular graph generation with Graph Neural Networks
• A Deep Generative Model for Molecule Optimization via One Fragment Modification
• Learning to Extend Molecular Scaffolds with Structural Motifs
• Simulated annealing for optimization of graphs and sequences
• Learning Interpretable Representations of Entanglement in Quantum Optics Experiments using Deep Generative Models
• BatmanNet: Bi-branch Masked Graph Transformer Autoencoder for Molecular Representation
• Multiresolution Graph Transformers and Wavelet Positional Encoding for Learning Hierarchical Structures
• GraphTune: A Learning-based Graph Generative Model with Tunable Structural Features
• SCGG: A Deep Structure-Conditioned Graph Generative Model
• Equivariant Graph Attention Networks with Structural Motifs for Predicting Cell Line-Specific Synergistic Drug Combinations