A Comparison of Low Read Depth QuantSeq 3' Sequencing to Total RNA-Seq in FUS Mutant Mice.

Seth Jarvis,Maria Secrier,Vincent Plagnol,Nicol Birsa,Pietro Fratta

doi:10.3389/fgene.2020.562445

Abstract

Transcriptomics is a developing field with new methods of analysis being produced which may hold advantages in price, accuracy, or information output. QuantSeq is a form of 3′ sequencing produced by Lexogen which aims to obtain similar gene-expression information to RNA-seq with significantly fewer reads, and therefore at a lower cost. QuantSeq is also able to provide information on differential polyadenylation. We applied both QuantSeq at low read depth and total RNA-seq to the same two sets of mouse spinal cord RNAs, each comprised by four controls and four mutants related to the neurodegenerative disease amyotrophic lateral sclerosis. We found substantial differences in which genes were found to be significantly differentially expressed by the two methods. Some of this difference likely due to the difference in number of reads between our QuantSeq and RNA-seq data. Other sources of difference can be explained by the differences in the way the two methods handle genes with different primary transcript lengths and how likely each method is to find a gene to be differentially expressed at different levels of overall gene expression. This work highlights how different methods aiming to assess expression difference can lead to different results.

Highlights

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease with multiple causes
While some of the differences will have come from the fact that QuantSeq uses only mRNA and total RNA-seq uses all RNApresent in the cell, we have identified some other possible causes of the differences in which genes are recognized as differentially expressed, including differences in how the two methods handle genes based on the number of reads they have and the length of their primary transcript
We performed our methods comparison on two sets of samples derived from embryonic spinal cords of: (a) four Fus KO mice and WT littermate controls, and (b) four Fus d14 mice with their own littermate WT controls

Summary

Introduction

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease with multiple causes. Mutations in genes which encode RNA binding proteins have been associated with the development of ALS. There is increasing interest in how genetic mutations can affect changes in RNA expression and lead to ALS (Boylan, 2015). One of the genes that has been shown to cause ALS is Fused in sarcoma (FUS); mutations in its coding sequence cause FUS, which is usually prevalently nuclear, to mislocalize to the cytoplasm. This mislocalization leads to cytoplasmic inclusions that are the hallmark of disease and further to nuclear depletion of the protein (Kwiatkowski et al, 2009; Vance et al, 2009). Whether the excess of FUS in the cytoplasm or its loss from the nucleus are the drivers of disease remains an open question, and there is evidence supporting FUS mutations causing gain of function, the loss of nuclear function may contribute to disease (Scekic-Zahirovic et al, 2016; Birsa et al, 2020)

Objectives

Methods

Results

Conclusion