APALORD: An R-based tool for differential alternative polyadenylation analysis of long-read RNA-seq data

Abstract Alternative polyadenylation (APA) is a critical co/post-transcriptional process that enhances RNA isoform diversity, regulating mRNA stability, localization and translation in a spatiotemporal manner. Over the past decade, long-read (LR) RNA sequencing techniques have advanced rapidly, producing datasets that could offer insights into APA mechanisms. Here we introduce APALORD (Alternative Polyadenylation Analysis of LOng-ReaDs), an R-based analysis tool for APA analysis of LR RNA-seq data. Leveraging precise 3’ end information from 3’-primed LR RNA-seq data, APALORD identifies polyadenylation sites (PASs) and quantifies PAS usage (PAU) at individual sites for each sample. It conducts APA analysis at the gene level (using Kolmogorov-Smirnov test) and at the level of individual PAS (using DEXSeq) across sample conditions. APALORD was applied to direct RNA-seq data from human embryonic stem cells (hESCs) and hESC-derived neurons. PASs were identified with high accuracy and a transcriptome-wide 3’UTR lengthening trend was found, consistent with previous studies. APALORD analysis of PacBio cDNA data from human tissues confirmed a 3’UTR lengthening trend in cortex compared to liver. R2C2 libraries generated on the Nanopore platform were analyzed with APALORD and it revealed APA change associated with polysome fractions in human neural progenitor cells. In summary, APALORD offers a comprehensive framework for differential APA analysis using LR RNA-Seq data, empowering researchers to investigate 3’ end dynamics across diverse biological contexts.