DEsingle for detecting three types of differential expression in single-cell RNA-seq data

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Single-cell RNA sequencing (scRNA-seq) data is characterized by a high proportion of zero counts, stemming from both biological variability ('real' zeros) and technical limitations ('dropout' zeros).
• Current differential expression (DE) analysis methods struggle to differentiate between these two zero types, potentially leading to inaccurate biological interpretations.

Purpose of the Study:

• To develop a novel computational approach for distinguishing real and dropout zeros in scRNA-seq data.
• To introduce an R package, DEsingle, that accurately identifies three distinct types of differentially expressed genes by accounting for zero-inflation.

Main Methods:

• Utilized a Zero-Inflated Negative Binomial (ZINB) model to estimate the proportions of real and dropout zeros.
• Developed the DEsingle R package implementing the ZINB model for DE gene detection in scRNA-seq datasets.

Main Results:

• DEsingle effectively estimates the proportion of real versus dropout zeros.
• The package accurately defines and detects three types of DE genes, improving upon existing methods.

Conclusions:

• DEsingle provides a more accurate method for differential gene expression analysis in scRNA-seq data by addressing the challenge of zero-inflation.
• The R package offers a valuable tool for researchers studying gene expression at the single-cell level.