Extended Data Fig. 6: YTHDC2 recognition of seRNA m6A.
a, Read densities of YTHDC1 and YTHDC2 iCLIP-seq at the center and 1 kilobases flanking regions of m6A-seRNAs in HEK293T cells. Data from GSE51633, GSE156400, GSE92879, GSE114543 and GSE78030. P value was examined by two-sided Wilcoxon rank-sum test. b, Overlap between YTHDC2-binding peaks in seRNAs and hyper-m6A seRNAs in PDAC samples. c, Venn plots of overlap between YTHDC1 CLIP peaks in seRNAs and CFL1-related m6A and H3K4me3 regions in PANC-1 cells (left) or hyper-m6A seRNAs in PDAC samples (right). d, REMSA showing no interaction between m6A-seRNA oligo or unmethylated-seRNA oligo and recombinant YTHDC1 protein. e, Western blot analysis of RNA pull-down products of cell lysate with oligo-A or oligo-m6A shows binding of YTHDC2 but not YTHDC1 to oligo-m6A. f, The genomic distribution of YTHDC2 PAR-CLIP-seq peaks in PANC-1 nuclear RNA. g, YTHDC2 enrichment with five representative seRNAs or seRNA without m6A site in CFL1 or METTL3 KO PDAC cells. h, The absence of YTHDC1 enrichment in five representative seRNAs or seRNA without m6A site in CFL1 or METTL3 KO PDAC cells. i−k, Enrichment of m6A (i), YTHDC2 (j) and H3K4me3 (k) with five representative seRNAs or seRNA without m6A site in CFL1 KO cells transfected with dPspCas13b-METTL3 gRNA or its control dPspCas13b-METTL3 NT. l, The absence of YTHDC1 enrichment in five representative seRNAs or seRNA without m6A site in CFL1 KO along with dPspCas13b-METTL3 system transfection. Results in (g-l) are mean ± SD from 3 independent experiments. P by one-way ANOVA with Dunnett’s T3 multiple-comparison test in g−l. n.s., not significant.
