Identification and quantification of alternative polyadenylation sites in single cell RNA-seq data using scPAISO

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Alternative polyadenylation (APA) is a key mechanism for generating transcript diversity through distinct mRNA isoforms.
• Current single-cell RNA sequencing (scRNA-seq) methods for APA analysis are limited by their reliance on Read2 data, hindering precise cleavage site (CS) and polyadenylation site (PAS) identification.

Purpose of the Study:

• To introduce scPAISO, a novel computational pipeline for accurate de novo PAS identification and PAS isoform quantification in single cells.
• To leverage Read1 data from 3' tag-based scRNA-seq for enhanced APA analysis.

Main Methods:

• Development of the scPAISO computational pipeline.
• Utilizing Read1 data from 3' tag-based scRNA-seq for PAS identification and quantification.
• Comparative analysis with existing scRNA-seq methods for APA.

Main Results:

• scPAISO enables direct capture of mRNA 3' end cleavage sites, showing improved AAUAAA motif enrichment and PAS peak resolution.
• The pipeline accurately identified PASs and quantified APA dynamics across diverse datasets and biological systems.
• Demonstrated stage-specific 3' UTR lengthening in hematopoietic differentiation, 3' UTR remodeling in systemic sclerosis, and tissue-specific polyadenylation preferences in mice.

Conclusions:

• scPAISO offers an accurate and scalable framework for single-cell APA analysis.
• This method enhances the high-resolution characterization of post-transcriptional regulation and transcriptome diversity.
• Facilitates deeper understanding of APA in development, physiology, and disease contexts.