Assessment of different enrichment methods revealed the optimal approach to identify bovine circRNAs

To better study the bovine circRNA, we assessed the outcomes of circRNA identification using six enrichment approaches. RNA-sequencing analysis revealed that different approaches led to varied the ratio of uniquely mapped reads, false positive rate of identifying circRNAs, and the number of circRNAs per million clean reads. Besides, 507 of 4,051 identified bovine high confident circRNAs had shared splicing sites with human circRNAs. Overall design: After extracting the total RNA of the liver and rumen from three (Kinsella Composite) beef steers, six enrichment approaches with the combination of ribosomal RNAs removal (Ribo); linear RNAs degradation (R); linear RNAs and RNAs with structured 3' ends degradation (RTP); ribosomal RNAs coupled with linear RNAs (Ribo-R) elimination; ribosomal RNA, linear RNAs and RNAs with poly (A) tailing elimination (Ribo-RP); and ribosomal RNA, linear RNAs and RNAs with structured 3' ends elimination (Ribo-RTP) were performed before RNA-seq libraries construction, respectively. Besides, five raw RNA-seq data with the same libraries construction strategies as ribosomal RNAs removal were downloaded from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) with the accession number GSE116775. These five reanalysis data were aggregated into the Ribo group. Subsequently, we compared circRNAs identified by different enrichment methods based on all RNA sequencing results.