Gene Yang ‘19

Although the RNA molecules involved in protein synthesis have been extensively studied, other novel types of RNA are still relatively unknown and continue being researched. One such novel type of RNA is circular RNA (circRNA), an endogenous and diverse class of molecules that is present in most eukaryotic cells. Circular RNAs are created from the splicing of pre-mRNA molecules  as they form closed loop structures after splicing. Although traditionally thought to be noncoding sequences, recent findings suggest that some circRNA do indeed produce proteins in vivo. Researchers from Zhejiang University, China, have developed an integrated bioinformatics tool, CircPro, that can detect coding circRNAs from sequencing data.

The workflow of CircPro can be categorized into three steps. The first involves the detection of circRNA sequences from a given dataset. This is accomplished using the program CIRI2, or the leading de novo circRNA detection tool that identifies these spliced sequences using the statistical maximum likelihood method. Once these circRNAs are identified within a genome, the second step involves determining whether each circRNA is protein coding or non-protein coding based on past known research (i.e. existing transcript annotations). The final step consists of verifying the results generated in the second step via a technique known as ribosome profiling (Ribo-Seq), which utilizes specialized mRNA sequencing in order to determine which parts of the genome are being actively translated.

This new CircPro tool was then tested on real biological sequencing data of human MCF7 cell lines, or sequences with known circRNAs locations. A total of 2036 circRNAs were predicted using CircPro estimated at 87.97% accuracy. In particular, 6 out of 10 of the known polysome-associated circRNAs were detected using this method. Although more research into circRNAs is needed to categorize its specific function, CircPro accelerates this process by introducing new and efficient bioinformatics techniques.

References

1. X. Meng, et al., CircPro: an integrated tool for the identification of circRNAs with protein-coding potential. Bioinformatics 33, 3314-3316 (2017). doi: 10.1093/bioinformatics/btx446
2. Image retrieved from: https://c1.staticflickr.com/6/5671/23523994309_05cdcdf38e_b.jpg