Isolation and sequencing of AGO-bound RNAs reveals characteristics of mammalian stem-loop processing in vivo

Ian M. Silverman; Sager J. Gosai; Nicholas Vrettos; Shawn W. Foley; Nathan D. Berkowitz; Zissimos Mourelatos; Brian D. Gregory

We reasoned that the ratio of mature miRNAs to pre-miRNAs represents a reasonable estimate for in vivo miRNA processing efficiency. As some miRNAs had no detectable pre-miRNA reads or vice versa, we used a generalized log odds ratio of miRNA-seq to pre-miRNA-seq reads (see METHODS) to compute a mature to precursor abundance ratio (MPAR) for miRNAs expressed in human HEK293T and mouse MEF cells (Tables S4 and S5 Among the maximum scoring high confidence miRNAs in humans was hsa-miR-338, which had no detectable pre-miRNA-seq reads and 1,358 reads per million (RPM) in miRNA-seq libraries (MPAR = 11.4) ( Table S5 ). In contrast, we found a number of miRNAs that had many more pre-miRNA-seq reads than mature miRNA-seq reads. To validate that our MPAR predictions are representative, we performed northern blotting on one human (hsa-miR-1296) and two mouse (mmu-miR-671 and mmu-miR-503) miRNAs with low MPAR scores (Figs. 4E-G) .