RNA-Seq is replacing more and more the microarray analysis for quantitative comparison of expression levels in different samples. Several studies have compared independently both techniques with respect to accuracy and reproducibility. All of these have reported the superiority of RNA-Seq (Richard et al., 2010 and Sultan et al., 2008).
Microarray systems rely on hybridization. Background hybridization levels or signal saturation constrict accurate quantification of the expression pattern of low and high abundance transcripts, respectively (e.g. Gautier et al., 2004 and Mortazavi et al., 2008). Provided that sufficient sequencing depth is applied, RNA-Seq does not show these kinds of limitations. This was demonstrated by e.g. Wang et al., 2008 who compared RNA-Seq read counts to published qRT–PCR measurements for hundreds of genes. They have observed almost linear relationship across five orders of magnitude.
The added value of RNA-Seq is that it enables the identification of new splicing variants and new transcription initiation sites. In addition, the quantification of particular exon and splicing isoform expression at a genome-wide scale is possible. Using RNA-Seq significant differences in gene isoform expression level and transcriptions start sites were found recently between normal and Alzheimer’s disease brain tissue (Twine et al., 2011). This result provided a big step forward in Alzheimer’s disease research.
Given the better accuracy and the add-on values in my opinion RNA-Seq will become an essential technique for researchers analyzing genome-wide gene regulation in any species. This essentially being true where reference sequences are available. Further decrease in costs of next generations sequencing will even facilitate this development.
Have you made first experiences with RNA-Seq in terms of expression profiling that you want to share with other blog visitors?