Molecular Attributes Transfer from Non-Parallel Data

Abstract

Optimizing chemical molecules for desired properties lies at the core of drug development. Despite initial successes made by deep generative models and reinforcement learning methods, these methods were mostly limited by the requirement of predefined attribute functions or parallel data with manually pre-compiled pairs of original and optimized molecules. In this paper, for the first time, we formulate molecular optimization as a style transfer problem and present a novel generative model that could automatically learn internal differences between two groups of non-parallel data through adversarial training strategies. Our model further enables both preservation of molecular contents and optimization of molecular properties through combining auxiliary guided-variational autoencoders and generative flow techniques. Experiments on two molecular optimization tasks, toxicity modification and synthesizability improvement, demonstrate that our model significantly outperforms several state-of-the-art methods.

References in corpus (16)

• Adam: A Method for Stochastic Optimization
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Improved Training of Wasserstein GANs
• Variational Inference with Normalizing Flows
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Style Transfer from Non-Parallel Text by Cross-Alignment
• MADE: Masked Autoencoder for Distribution Estimation
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
• Syntax-Directed Variational Autoencoder for Structured Data
• Improving Variational Inference with Inverse Autoregressive Flow
• Learning Multimodal Graph-to-Graph Translation for Molecular Optimization
• Multi-Objective Molecule Generation using Interpretable Substructures
• Hierarchical Graph-to-Graph Translation for Molecules
• Toward Diverse Text Generation with Inverse Reinforcement Learning
• Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation