Impact of RNA degradation on mRNA seq data

In eukaryotes, and some other organisms, mRNA transcripts are characterized by a poly-A tail at the 3’ end. Illumina mRNA-specific library preparation workflows utilize oligo-dT beads to select for polyadenylated RNAs, which are then prepared into sequencing libraries.

Figure 1. Purifying and Fragmenting mRNA

Illumina recommends using high quality RNA input (RINs of at least 8) for the TruSeq Stranded mRNA workflow, and the TruSeq RNA v2 workflow. The following addresses the possible issues that can arise when degraded RNA (RINs < 8) is used with mRNA-specific library preparation methods.

3’ bias

This bias is due to the 3’ oligo-dT selection of the poly-A tail of the transcript. Depending on the degradation status, the 5’ end of the transcript can be lost. This issue can lead to mis-identification or loss of information regarding splice variants and loss of coverage.

Reduced alignment efficiency

Depending on the extent of the degradation, the alignment efficiency can be negatively impacted. Sequencing RNA samples of lower RNA quality can result in lower percentages of mappable reads. An increase in intergenic reads and decrease of intronic reads may also be observed.

Effect on RNA-seq output

Several issues can arise in data sets generated with degraded RNA:

· Uneven degradation rates among the transcripts can cause protein-coding genes to degrade faster than pseudogenes. Higher %GC content and increased length of both the 3′ UTR and CDS are also associated with faster degradation.

· Unexpected gene expression variation.

· Bias in gene expression studies. It has been shown that the read per kilobase transcript per million (RPKM) values are positively correlated with the RIN, with RNA samples of low quality displaying lower RPKM values.

For any feedback or questions regarding this article (Illumina Knowledge Article #2731), contact Illumina Technical Support [e-mail].