Efficient learning of non-autoregressive graph variational autoencoders for molecular graph generation

With the advancements in deep learning, deep generative models combined with graph neural networks have been successfully employed for data-driven molecular graph generation. Early methods based on the non-autoregressive approach have been effective in generating molecular graphs quickly and efficie...

Full description

Saved in:
Bibliographic Details
Published in:Journal of cheminformatics Vol. 11; no. 1; pp. 70 - 10
Main Authors: Kwon, Youngchun, Yoo, Jiho, Choi, Youn-Suk, Son, Won-Joon, Lee, Dongseon, Kang, Seokho
Format: Journal Article
Language:English
Published: Cham Springer International Publishing 21.11.2019
BioMed Central Ltd
Springer Nature B.V
BMC
Subjects:
Artificial neural networks
Chemistry
Chemistry and Materials Science
Computational Biology/Bioinformatics
Computer Applications in Chemistry
Deep learning
Documentation and Information in Chemistry
Graph matching
Graph neural network
Graph neural networks
Graphs
Machine learning
Molecular graph
Neural networks
Organic chemistry
Research Article
Theoretical and Computational Chemistry
Variational autoencoder
Deep learning
Variational autoencoder
Molecular graph
Graph neural network
ISSN:1758-2946, 1758-2946
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the advancements in deep learning, deep generative models combined with graph neural networks have been successfully employed for data-driven molecular graph generation. Early methods based on the non-autoregressive approach have been effective in generating molecular graphs quickly and efficiently but have suffered from low performance. In this paper, we present an improved learning method involving a graph variational autoencoder for efficient molecular graph generation in a non-autoregressive manner. We introduce three additional learning objectives and incorporate them into the training of the model: approximate graph matching, reinforcement learning, and auxiliary property prediction. We demonstrate the effectiveness of the proposed method by evaluating it for molecular graph generation tasks using QM9 and ZINC datasets. The model generates molecular graphs with high chemical validity and diversity compared with existing non-autoregressive methods. It can also conditionally generate molecular graphs satisfying various target conditions.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1758-2946
1758-2946
DOI:10.1186/s13321-019-0396-x