iDESC: Identifying differential expression in single-cell RNA sequencing data with multiple subjects

Abstract

Single-cell RNA sequencing (scRNA-seq) enables assessment of transcriptome-wide changes at single-cell resolution. However, dominant subject effect in scRNA-seq datasets with multiple subjects severely confounds cell-type-specific differential expression (DE) analysis. We developed iDESC to separate subject effect from disease effect with consideration of dropouts to identify DE genes. iDESC was shown to have well-controlled type I error and high power compared to existing methods and obtained the best consistency between datasets and disease relevance in two scRNA-seq datasets from same disease, suggesting the importance of considering subject effect and dropouts in the DE analysis of scRNA-seq data with multiple subjects.