Circular RNAs (circRNAs) belong to a recently re-discovered species of RNA that emerge during RNA maturation through a process called back-splicing. A downstream 5’ splice site is linked to an upstream 3’ splice site to form a circular transcript instead of a canonical linear transcript. Recent advances in next-generation sequencing (NGS) have brought circRNAs back into the focus of many scientists. Since then, several studies reported that circRNAs are differentially expressed across tissue types and developmental stages, implying that they are actively regulated and not merely a by-product of splicing. Though functional studies have shown that some circRNAs could act as miRNA-sponges, the function of most circRNAs remains unknown.

To expand our understanding of possible roles of circular RNAs, researchers from the Klaus Tschira Institute for Integrative Computational Cardiology propose a new pipeline that could fully characterizes candidate circRNA structure from RNAseq data – FUCHS: FUllCHaracterization of circular RNA using RNASequencing. Currently, most computational prediction pipelines use back-spliced reads to identify circular RNAs. FUCHS extends this concept by considering all RNA-seq information from long reads (typically > 150 bp) to learn more about the exon coverage, the number of double break point fragments, the different circular isoforms arising from one host-gene, and the alternatively spliced exons within the same circRNA boundaries.

(a) FUCHS work-flow: First FASTQ files have to be mapped, second, circRNAs have to be detected, these steps are executed by the user before running FUCHS. Running FUCHS will automatically start all steps that are encircled by the dashed line. Most of these steps do not depend on each other and may be skipped with the -sS flag. The output of FUCHS can be used to identify false positive circRNAs. The resulting BED files can be used as input for differential motif enrichment or for miRNA seed analysis to gain knowledge about functionally interesting target circRNAs. (b) All reads: This is a simplified view of a BAM file before circular read extraction. This file may contain linear and chimeric reads, thus being rather cluttered. (c) Reads by circle: Separating chimeric reads by circleIDs has the advantage that the structure of these circRNAs is easy to observe, especially when a host-gene harbours many different circular isoforms.

This new knowledge will enable the user to carry out differential-motif enrichment and miRNA-seed analysis to determine potential regulators during circRNA biogenesis. FUCHS is an easy-to-use Python based pipeline that contributes a new aspect to the circRNA research.