Learning Cell-Type-Specific Gene Regulation Mechanisms by Multi-Attention Based Deep Learning With Regulatory Latent Space

Minji Kang; Sangseon Lee; Dohoon Lee; Sun Kim

doi:10.3389/fgene.2020.00869

Learning Cell-Type-Specific Gene Regulation Mechanisms by Multi-Attention Based Deep Learning With Regulatory Latent Space

Front Genet. 2020 Sep 30:11:869. doi: 10.3389/fgene.2020.00869. eCollection 2020.

Authors

Minji Kang¹, Sangseon Lee¹, Dohoon Lee², Sun Kim^{1

2

3}

Affiliations

¹ Bioinformatics Institute, Seoul National University, Seoul, South Korea.
² Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, South Korea.
³ Department of Computer Science and Engineering, Institute of Engineering Research, Seoul National University, Seoul, South Korea.

Abstract

Epigenetic gene regulation is a major control mechanism of gene expression. Most existing methods for modeling control mechanisms of gene expression use only a single epigenetic marker and very few methods are successful in modeling complex mechanisms of gene regulations using multiple epigenetic markers on transcriptional regulation. In this paper, we propose a multi-attention based deep learning model that integrates multiple markers to characterize complex gene regulation mechanisms. In experiments with 18 cell line multi-omics data, our proposed model predicted the gene expression level more accurately than the state-of-the-art model. Moreover, the model successfully revealed cell-type-specific gene expression control mechanisms. Finally, the model was used to identify genes enriched for specific cell types in terms of their functions and epigenetic regulation.

Keywords: cell-type-specific; deep learning; gene regulation mechanism; gene regulatory network; multi-omics.