Fig. 1: RBP binding close to a miRNA target site is associated with enhanced miRNA targeting. | Nature Communications

Fig. 1: RBP binding close to a miRNA target site is associated with enhanced miRNA targeting.

From: The regulatory impact of RNA-binding proteins on microRNA targeting

Fig. 1

a Overview of the analysis to investigate the effect of RBP binding on the miRNA targeting (MT) efficacy. Genes were grouped together based on the distance between a miRNA target site (MTS) and the nearest RBP-binding site (RBS) on the 3′UTR, denoted as dMTS-RBS. The genes with an MTS that overlaps with an RBS were categorized as ‘overlapped’. The association between MT efficacy and dMTS-RBS was analyzed by measuring the mRNA fold change. b Association analysis between dMTS-RBS and MT efficacy. The mean mRNA fold changes of 3′UTRs with a single 7, 8mer MTS obtained from mRNA-seq (HepG2) or microarray datasets (HeLa and other human cancer cell lines) that monitored the whole-transcriptome response to overexpressed miRNAs/siRNAs are shown. The 3′UTRs were grouped with respect to dMTS-RBS, as depicted in (a). The mRNA fold changes were compared between the 3′UTR groups (two-sided Wilcoxon’s rank-sum test). The number of 3′UTRs used for measuring dMTS-RBS is shown on top. mRNA fold changes are displayed in the log2 scale and the error bars represent 95% confidence intervals. c Residuals of the mRNA fold change for each group from (b) after correcting for four known confounding features of MT (local AU content, target abundance (TA), seed-pairing stability (SPS), and 3′UTR length). The regression residuals represent the remaining information after reducing the contribution from the confounding features. Otherwise as in (b). d Association analysis between dMTS-RBS and MT efficacy after a more rigorous correction for the confounding features. A subset of 3′UTRs analyzed in (b) were selected and split into four subgroups with respect to the dMTS-RBS. Each subgroup was carefully chosen to have statistically indistinguishable confounding features between each other (see Supplementary Fig. 1f for full versions). The mean values of confounding features, dMTS-RBS, and log2(mRNA fold change) are displayed (***P < 0.001). Otherwise as in (b). e Association analysis between dMTS-RBS and MT efficacy after deleting DROSHA or DICER. Five miRNAs whose targets show the strongest derepression in response to miRNA removal were chosen and the association between dMTS-RBS and MT efficacy was evaluated. See Supplementary Fig. 1h for full versions. Otherwise as in (d).

Back to article page