Theoretical model of circular RNA prediction and classification

Abstract

Circular RNAs are reported as new member in RNA world, whose ends are covalently joined and not code for any protein reported till now. Importance of circular RNA as biomarkers for various types of cancers is well reported. Currently, circular RNA candidate prediction is considered as one of the most important steps to further explore role of Circular RNAs in cellular system. Researchers are trying to find circular candidates that are based on the model of events which are responsible for the biogenesis of the circular RNAs. The exact mechanism of circular RNAs biogenesis is not known till now, but there are many proposed methodologies suggested for the genesis of the circular RNAs like exons skipping, back-splicing, exons scrambling, alternate splicing etc. The reported methods till now for finding circular candidate are based on the back splicing event which requires prediction of back splicing sites. This process requires mapping the reads under consideration to their corresponding reference genome. We proposed a de novo-assembly of polyA minus reads based approach that does not requires mapping of reads with reference genome and not depend on prediction of back splice sites to find the circular candidates. We have predicted potential circular RNA candidates by mimicking the output of arrangement of nucleotide sequences after back splicing event. We then perform classifications of circular RNA candidates into coding and non-coding on the basis of origination of candidates RNAs on genome.