The Center for Medical Genomics considers a variety of factors to decide which RNA protocol to use: Total RNA-seq vs. mRNA-seq.
Normally, total RNA-Seq (also known as whole-transcriptome sequencing) refers to the sequencing of RNA that has been depleted of ribosomal RNA (rRNA), which represents the majority of RNA molecules, both coding and noncoding. Total RNA, when originally isolated, is composed of multiple RNA species, including rRNA, precursor messenger RNA (pre-mRNA), messenger RNA (mRNA), and several types of noncoding RNA (ncRNA), such as transfer RNA (tRNA), microRNA (miRNA), and long ncRNA (lncRNA; transcripts longer than 200 nucleotides not translated into protein). The removal of rRNA in the total RNA-Seq procedure results in improved sequencing data that enables the characterization of these diverse non-rRNA species.
Using a selection method to enrich for polyadenylated (poly(A)) RNA, mRNA-seq targets on the messenger RNA molecules (and some long noncoding RNA that are polyadenylated), which represents only a small percentage of the total RNA molecules. If the research goal is to focus primarily on the coding region, mRNA-Seq is the most efficient and cost-effective procedure, and therefore represents the best choice.
When the RNA quality is high (RNA Integrate Number, RIN>7), although total RNA-seq offers broader profiling of the RNA molecules, the data quality of this procedure is normally much noisier than mRNA-seq. Much higher sequencing depth is required, and a much higher percentage of intergenic and intronic sequencing reads are observed. On the contrary, mRNA-seq offers much “cleaner” data. In addition, mRNA-seq can also measure most of the noncoding RNA species that are polyadenylated.
When the RNA quality is low (RIN<7), regardless of the experimental objective, the center recommends total RNA-seq. This is because mRNA-seq primarily targets on the RNA fragments with polyA, this assay will lead to excessive 3’-biase (most signals will be concentrating on the 3’-end of the gene), and the gene expression quantification will be interfered by the RNA quality.