Figure 1. CLEAR-CLIP unambiguously identifies endogenous in vivo miRNA–target interactions.

(a) In CLEAR-CLIP, AGO–target contacts are cross-linked in vivo by ultraviolet irradiation. Endogenous AGO is immunopurified from tissue lysates and washed under stringent conditions that disrupt the interaction of AGO–miRNA with non-cross-linked target RNAs. Target regions cannot be cloned from no-ultraviolet controls, indicating that cross-linking of AGO to target mRNA (shown as ‘X') is required. Cross-linking of the miRNA may not be necessary, because the AGO–miRNA interaction is uniquely strong and survives stringent washing. After washing, RNA ends are modified to facilitate miRNA–target ligation and joined with T4 RNA Ligase I treatment, yielding miRNA–target chimeric RNAs in two orientations at the indicated frequencies. All depicted post-IP manipulations up to SDS–PAGE occur on beads. Correlation plots of miRNA abundance of all miR-first (b) and miR-last (c) chimeras versus small RNA sequencing data in the brain67. Pearson's correlation coefficients (r) are shown. CDF plots of cognate miRNA seed matches in target regions relative to ligation site for all miR-first chimeras in plus-ligase (d) and no-ligase (e) samples, and for all miR-last chimeras in plus-ligase (f) and no-ligase (g) samples. (h) Distribution of standard AGO CLIP and miRNA–target chimeras in transcript regions. (i) CLEAR-CLIP confirmed known miRNA regulation, here exemplified by miR-124 regulation of the Ptbp1 3′-UTR. Other examples are shown in Supplementary Fig. 3c.