Deciphering Cellular Heterogeneity by Integrative Single-Cell and Bulk RNA-Seq Data Analysis

• Activity Number: 561 - Fine-Scale Inference from Aggregate-Level Genomic Data
• Type: Invited
• Date/Time: Thursday, August 6, 2020 : 3:00 PM to 4:50 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Mingyao Li* and Yafei Lyu and Jiaxin Fan and Rui Xiao
• Companies: University of Pennsylvania and University of Pennsylvania and University of Pennsylvania and University of Pennsylvania
• Keywords: single cell RNA-seq; deconvolution; allele specific expression

Abstract

Knowledge of cell type composition in disease relevant tissues is an important step towards the identification of cellular targets of disease. In this talk, I will present a method that utilizes cell-type specific gene expression from single-cell RNA-seq data to characterize cell type compositions from bulk RNA-seq data in complex tissues from diverse samples. By iteratively identifying cell type invariant genes between disease conditions and appropriately weighting of genes showing cross-subject and cross-cell consistency, our method can transfer cell type-specific gene expression information from one data set to another, and infer cell type compositions in diverse samples. We further show that the estimated cell type proportions allow us to characterize allele-specific expression (ASE) with cell type resolution in bulk RNA-seq data. To do so, we regress the bulk level allele-specific read counts over estimated cell-type proportions through a linear mixed-effect model, and test for the presence of ASE in each cell type. Extensive evaluations show that this method is powerful in detecting cell type-specific ASE effect even for rare cell types.