Document Type
Article
Publication Title
arXiv
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. © 2021, CC BY-NC-SA.
DOI
10.48550/arXiv.2111.15146
Publication Date
11-30-2021
Keywords
Attribute functions; Chemical molecules; Drug development; Generative model; Learn+; Molecular optimization; Parallel data; Property; Reinforcement learning method; Transfer problems
Recommended Citation
S. Zheng, Y. Song, P. Zhang, L. Song, C. Li, and Y. Yang, "Molecular attributes transfer from non-parallel data," 2021, arXiv:2111.15146
Comments
Preprints: arXiv