Circular RNAs as a population are typically present at much lower levels, at about 5~10% of linear RNAs. The cross circular junction sequences are even lower. At a typical RNA-seq depth, less than 5% of circular RNAs (red circle) may be reliably quantified (Fig. 6). Even at great sequencing depth of > 300 mil at high costs, the accuracy gains are only modest.
In practice, generic RNA-seq, mostly intended for mRNAs, are inadequate or simply unavailable as a provided service for circular RNA profiling. Circular RNA sequencing requires very deep sequencing depth and paired-end chemistry. Read mapping and data analysis require specialized database, de novo transcript assembly, special algorithms and complex computational pipeline. circRNA annotations such as microRNA binding as sponges are typically not included. Novel circRNA discovery, a consideration of using RNA-seq, is actually not available for the above reasons.
On the other hand, circRNA microarrays use circular junction probes, combined with enzymatic linear RNA removal, to interrogate circular RNAs highly specifically. Array hybridization is relatively independent of other high abundance RNAs. High sensitivity at one transcript per cell can be achieved. Overall, microarray is more efficient and robust in sample labeling than RNA sequencing library prep. At present, it is a mature technology that outperforms RNA-seq in circRNA profiling.
Figure 1. RNA-seq quantification reliability vs read depth. Typical RNA-seq has a depth of < 30 mil reads for mRNAs (blue circle), which is < 0.5 mil for cross circular junction reads (red circle). Less than 5% circular junctions can be reliably quantified. Adopted from Labaj et al, (2011) Bioinformatics [PMID 21685096].
Circular RNA Array Service
Circular RNA qPCR Service