Fig. 3

RCMs are required for circRNA production. a Length distributions of introns flanking circRNA-producing exon(s), compared with the lengths of all introns. The median values were shown. Numbers in the brackets are numbers of introns used for analysis. b Number of RCMs in one pair of flanking introns of circRNA compared with those in control exons (exon 2 and exon 8). Values are shown as mean ± SEM. Numbers in the brackets are numbers of intron pairs used for analysis. p values in a & b are from Kruskal-Wallis test with Dunn’s post-hoc test for multiple comparisons. ****, p < 0.0001. c Schematic plot showing that RCMs promote circRNA production and the strategy to disturb one of the RCMs by CRISPR-Cas9. d Quantification of linear mRNA and circRNA in wild-type N2 strain and RCM deletion mutant strains of six circRNA genes. Error bars are the standard deviations of three biological replicates. n.d.: not detected (Ct values not determined or bigger than those in no-template controls). Two-tail student’s t-test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. ns: not significant