EPISCORE: cell type deconvolution of bulk tissue DNA methylomes from single-cell RNA-Seq data

Andrew E Teschendorff,Charles E Breeze,Tianyu Zhu,Stephan Beck

doi:10.1186/s13059-020-02126-9

Abstract

Cell type heterogeneity presents a challenge to the interpretation of epigenome data, compounded by the difficulty in generating reliable single-cell DNA methylomes for large numbers of cells and samples. We present EPISCORE, a computational algorithm that performs virtual microdissection of bulk tissue DNA methylation data at single cell-type resolution for any solid tissue. EPISCORE applies a probabilistic epigenetic model of gene regulation to a single-cell RNA-seq tissue atlas to generate a tissue-specific DNA methylation reference matrix, allowing quantification of cell-type proportions and cell-type-specific differential methylation signals in bulk tissue data. We validate EPISCORE in multiple epigenome studies and tissue types.

Highlights

DNA methylation is a key cell type-specific epigenetic mark associated with gene expression that plays a key role in development and differentiation [1]
Construction and validation of a lung-specific mRNA expression reference Since EPISCORE is primarily aimed at dissecting the cellular heterogeneity of complex solid tissues, we first focused on lung, a tissue for which ample scRNA-Seq and DNA methylation (DNAm) data are available, allowing for rigorous validation
Seq and DNAm data of purified samples from Epigenomics Roadmap and Stem-Cell-Matrix Compendium-2 (SCM2), we identify genes for which differential DNAm at their regulatory elements across the samples is predicted by corresponding gene expression

Summary

Introduction

DNA methylation is a key cell type-specific epigenetic mark associated with gene expression that plays a key role in development and differentiation [1]. The challenge posed by cell type heterogeneity is best addressed with single-cell technologies [8, 9], which consortia such as the Human and Mouse Cell Atlas (HCA/MCA) projects [10,11,12,13] are using to generate tissue-specific single-cell RNA-Seq (scRNA-Seq) atlases at high cellular resolution Such tissue-specific scRNA-Seq atlases provide a nearly unbiased catalog of all major cell types present in a tissue and constitute a resource that is already being exploited to enable cell type deconvolution of bulk mRNA expression profiles [14].

Objectives

Methods

Results

Conclusion