Researchers at Oregon State University examined RNA-Seq data on 211 biological samples from 24 different Arabidopsis experiments carried out by different labs. They grouped the samples according to tissue types, and in each of the groups, they identified genes that are stably expressed across biological samples, treatment conditions, and experiments. The researchers 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 they 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, the researchers argue that DE is a relative concept. They 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.
Expression profiles of 15 genes—as measured by RNA-Seq CPM
across 91 samples in the multi-tissue group
The 15 genes include (from top to bottom) (A) five stably expressed genes (randomly selected out of the top 100) identified from the multi-tissue group RNA-Seq data using the total variance measure σˆ2, (B) five stably expressed identified by Czechowski et al. (2005) according to the CV measure from a developmental series of microarray experiments, and (C) five traditional house-keeping genes (HKG) discussed in Czechowski et al. (2005).