Novel applications of high-throughput RNA sequencing: mapping RNA structure and discovering circular RNAs
High-throughput RNA sequencing (RNA-Seq), although still novel, has primarily been applied as a method for assessing differential RNA abundance or mapping of primary structure of linear transcripts, e.g. inference of splice junctions. I report on two novel applications of RNA-Seq for which I developed computational pipelines. The first (FragSeq) is a coupling of classic enzymatic RNA structure probing with RNA-Seq in order to obtain high-throughput, single-base-resolution endonuclease accessibility maps of entire transcriptomes, thus yielding RNA structure information. A proof-of-principle application of this method on two mouse nuclear RNA samples showed that ssRNA regions of known nuclear ncRNA structures are accurately mapped. Also, mapping of novel structures was validated by follow-up probing. The second application is my pipeline for discovery of RNA circularization from RNA-Seq reads that I applied to a broad unpublished dataset spanning 21 archaeal species and a bacterium, uncovering evidence that C/D RNA guide transcripts are circularized in hyperthermophiles. My findings agree with published findings of circular C/D RNA in three species (P. furiosus, S. acidocaldarius, and S. solfataricus) and provide high-confidence evidence for broad C/D RNA circularization in at least two new species (I. hospitalis and T. kodakaraensis), arguing that this circularization is phylogenetically widespread. Interestingly, the crenarchaeal hyperthermophile P. aerophilum has circularization of transcripts anti-sense to C/D RNAs. This is currently the broadest study of circularization in any domain of life.