Identifying stably expressed genes from multiple RNA-Seq data sets.

Bin Zhuo,Yanming Di,Jeff H Chang,Sarah Emerson

doi:10.7717/peerj.2791

Abstract

We examined RNA-Seq data on 211 biological samples from 24 different Arabidopsis experiments carried out by different labs. We grouped the samples according to tissue types, and in each of the groups, we identified genes that are stably expressed across biological samples, treatment conditions, and experiments. We fit a Poisson log-linear mixed-effect model to the read counts for each gene and decomposed the total variance into between-sample, between-treatment and between-experiment variance components. Identifying stably expressed genes is useful for count normalization and differential expression analysis. The variance component analysis that we explore here is a first step towards understanding the sources and nature of the RNA-Seq count variation. When using a numerical measure to identify stably expressed genes, the outcome depends on multiple factors: the background sample set and the reference gene set used for count normalization, the technology used for measuring gene expression, and the specific numerical stability measure used. Since differential expression (DE) is measured by relative frequencies, we argue that DE is a relative concept. We advocate using an explicit reference gene set for count normalization to improve interpretability of DE results, and recommend using a common reference gene set when analyzing multiple RNA-Seq experiments to avoid potential inconsistent conclusions.

Highlights

RNA sequencing (RNA-Seq) has become the technology of choice for transcriptome profiling over the last few years
We identify stably expressed genes from RNA-Seq data sets based on a numerical measure—the sum of three variance components estimated from a mixed-effect model
In ‘Factors affecting stability ranking’, we further demonstrate that when using a numerical a b c measure to quantify gene expression stability, the outcome will depend on the specific numeric measure used

Summary

Introduction

RNA sequencing (RNA-Seq) has become the technology of choice for transcriptome profiling over the last few years. The exponential growth in RNA-Seq studies have produced a large amount of Arabidopsis thaliana (Arabidopsis) data under a variety of experimental/environmental conditions. It is only natural to begin exploring how the large amount of existing data sets can help the analysis of future data. We discuss identifying stably expressed genes from multiple existing RNA-Seq data sets based on a numerical measure of stability. We envision that such identified stably expressed genes could be used as a reference set or prior information for count normalization and differential expression (DE) analysis of future RNA-Seq data sets obtained from similar or comparable experiments. We fit a random-effect model to the read counts for each gene and decompose the total variance into between-sample, between-treatment and

Methods

Results

Conclusion