scDAPA: detection and visualization of dynamic alternative polyadenylation from single cell RNA-seq data

Motivation: Alternative polyadenylation (APA) plays a key post-transcriptional regulatory role in mRNA stability and functions in eukaryotes. Single cell RNA-seq (scRNA-seq) is a powerful tool to discover cellular heterogeneity at gene expression level. Given 3 enriched strategy in library construction, the most commonly used scRNA-seq protocol-10x Genomics enables us to improve the study resolution of APA to the single cell level. However, currently there is no computational tool available for investigating APA profiles from scRNA-seq data. Results: Here, we present a package scDAPA for detecting and visualizing dynamic APA from scRNA-seq data. Taking bam/sam files and cell cluster labels as inputs, scDAPA detects APA dynamics using a histogram-based method and the Wilcoxon rank-sum test, and visualizes candidate genes with dynamic APA. Benchmarking results demonstrated that scDAPA can effectively identify genes with dynamic APA among different cell groups from scRNA-seq data.