GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation

Abstract

Molecular graph generation is a fundamental problem for drug discovery and has been attracting growing attention. The problem is challenging since it requires not only generating chemically valid molecular structures but also optimizing their chemical properties in the meantime. Inspired by the recent progress in deep generative models, in this paper we propose a flow-based autoregressive model for graph generation called GraphAF. GraphAF combines the advantages of both autoregressive and flow-based approaches and enjoys: (1) high model flexibility for data density estimation; (2) efficient parallel computation for training; (3) an iterative sampling process, which allows leveraging chemical domain knowledge for valency checking. Experimental results show that GraphAF is able to generate 68% chemically valid molecules even without chemical knowledge rules and 100% valid molecules with chemical rules. The training process of GraphAF is two times faster than the existing state-of-the-art approach GCPN. After fine-tuning the model for goal-directed property optimization with reinforcement learning, GraphAF achieves state-of-the-art performance on both chemical property optimization and constrained property optimization.

References in corpus (15)

• Auto-Encoding Variational Bayes
• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Pixel Recurrent Neural Networks
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Variational Inference with Normalizing Flows
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Density estimation using Real NVP
• MolGAN: An implicit generative model for small molecular graphs
• MADE: Masked Autoencoder for Distribution Estimation
• Multiplicative Normalizing Flows for Variational Bayesian Neural Networks
• Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
• Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders
• MolecularRNN: Generating realistic molecular graphs with optimized properties

Cited by in corpus (28)

• Graph Neural Networks: A Review of Methods and Applications
• A Comprehensive Survey on Deep Graph Representation Learning
• Equivariant flow-based sampling for lattice gauge theory
• Molecular graph generation with Graph Neural Networks
• Regression Transformer: Concurrent sequence regression and generation for molecular language modeling
• A Deep Generative Model for Molecule Optimization via One Fragment Modification
• Generative AI and Process Systems Engineering: The Next Frontier
• DIG: A Turnkey Library for Diving into Graph Deep Learning Research
• Fast Graph Generation via Spectral Diffusion
• Simulated annealing for optimization of graphs and sequences
• On the design space between molecular mechanics and machine learning force fields
• Guiding Deep Molecular Optimization with Genetic Exploration
• Multi-Objective Latent Space Optimization of Generative Molecular Design Models
• GraphDF: A Discrete Flow Model for Molecular Graph Generation
• MARS: Markov Molecular Sampling for Multi-objective Drug Discovery
• Differentiable Scaffolding Tree for Molecular Optimization
• AdvFlow: Inconspicuous Black-box Adversarial Attacks using Normalizing Flows
• JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Design
• Categorical Normalizing Flows via Continuous Transformations
• Advancing Biomedicine with Graph Representation Learning: Recent Progress, Challenges, and Future Directions
• Optimizing ZX-Diagrams with Deep Reinforcement Learning
• SCGG: A Deep Structure-Conditioned Graph Generative Model
• HelixMO: Sample-Efficient Molecular Optimization in Scene-Sensitive Latent Space
• Weakly Supervised Concept Map Generation through Task-Guided Graph Translation
• Auto-decoding Graphs
• Partition and Code: learning how to compress graphs
• Graph Context Encoder: Graph Feature Inpainting for Graph Generation and Self-supervised Pretraining
• Learning Graphon Autoencoders for Generative Graph Modeling