PNNL

Abstract

The application of RNA-sequencing has led to numerous breakthroughs related to querying gene expression levels in numerous complex biological systems. Among these are knowledge of how organisms, such as the vertebrate model organism zebrafish (Danio rerio), respond to toxicant exposure. While RNA-seq is a powerful tool, its cost, and the requirement of tens of thousands of sequenced reads of RNA, means that it cannot always be applied to all sample types and biological systems. Recently, the development of 3’ RNA-seq (focused only on sequencing the 3’ end of RNA transcripts) has allowed for the determination of gene expression levels with a fraction of the required reads compared to standard RNA-seq (focused on sequencing and aligning all parts of the transcript). While 3’ RNA-seq has many advantages, there is still an open question regarding how well it can perform against standard RNA-seq with respect to alignment and identification of differentially expressed genes (DEGs) or enriched functions. Here, we compared 3’ RNA-seq data to standard sequencing in samples collected from zebrafish exposed to control conditions or an elevated level of perfluorobutane sulfonamide (FBSA). We found that 3’ RNA-seq and standard showed specific alignment advantages when focusing on annotated or unannotated regions of the genome, with 3’ RNA-seq showing better alignment to annotated regions. We also found that 3’ RNA-seq was far better at detecting DEGs under conditions of sparse data (fewer reads) but that DEGs identified by standard RNA-seq led to a greater number of statistically enriched biological functions. These results identify specific experimental conditions where 3’ RNA-seq may be advantageous (sparse data focused on well annotated genomes) compared to conditions where standard RNA-seq may be better. Our conclusions will help guide future experiments using RNA sequencing to query gene expression levels in a range of biological systems to gain the most accurate and widespread knowledge