Additional file 4: of Precise mapping of the transcription start sites of human microRNAs using DROSHA knockout cells

List of pri-miRNAs analyzed in this study. From the list of authentic miRNAs, which were established previously [20], the list of pri-miRNAs was generated based on the genomic locations of mature miRNAs. The pri-miRNAs were divided into two groups (A and B) depending on their overlap with protein-coding genes. By inspecting the RNA-seq data, the signal intensity was annotated as either ‘High’, ‘Low’, or ‘No’ (see Methods). (A) For intergenic pri-miRNAs with no overlap with protein-coding genes, the enrichment of RNA signals in the DROSHA knockout cells, compared to those in wild-type cells, was determined. Thirty-four pri-miRNAs that met our criteria are shown with bold letters. For the clustered miRNAs among the pri-miRNAs analyzed by RACE, the evidence of co-transcription was indicated with citations for 11 out of 20 clustered miRNAs. For six pri-miRNAs, we identified the expressed sequence tags (ESTs) with sequences spanning all miRNA members of each cluster and included the GenBank IDs of representative ESTs. For the remaining 3 miRNA clusters, we performed PCR experiments to confirm that the transcripts spanning all miRNA members are increased in the cDNA made from DROSHA knockout cells compared to that from wild-type cells. We included the results into Additional file 5. (B) For intronic miRNAs, which reside inside the introns of protein-coding genes, the accumulation of RNA signals at the miRNA-containing intron in the DROSHA knockout cells, compared to that in wild-type cells, was determined. (XLSX 17 kb)