MicroRNAs are RNAs of about 22 nucleotides that guide Argonaute proteins to partially complementary sites in mRNAs, which represses their translation and speeds up their decay. They are cut from hairpins in longer primary transcripts: Drosha releases the precursor hairpin, Dicer removes its loop, and one strand of the remaining duplex, the mature miRNA, is loaded into Argonaute while the other strand, the star or passenger, is usually degraded.
This track shows the human microRNA genes of MirGeneDB 3.0, a database of animal microRNAs that are curated by hand from small RNA sequencing data with uniform criteria for biogenesis and a nomenclature that follows the evolutionary history of each family. The track has 573 loci of 514 genes. Each item is a primary miRNA region as defined by MirGeneDB: the precursor hairpin plus 30 nucleotides on each side. The two arms of the hairpin are drawn on top of each item, colored by whether MirGeneDB annotates them as the dominant (mature) arm, the star arm, or as co-mature. For the larger, uncurated set of microRNAs registered in miRBase, see the miRBase track.
Clicking an item opens a page with the miRBase ID and accessions, family, seed, the positions of the arms and the secondary structure of the primary miRNA region predicted with RNAfold. The structure is drawn with forna and can be moved and zoomed; the nucleotides of the arms are highlighted in the colors of the track.
Each item is labeled with its MirGeneDB ID, e.g. Hsa-Mir-8-P1a. Some genes have several loci with the suffixes -v1 to -v4; these are variants of the same hairpin that MirGeneDB annotates with slightly different ends of the precursor and the arms. The arms are drawn as colored blocks over the item:
| Dominant arm (5p or 3p), the mature miRNA | |
| Star arm (5p* or 3p*), the passenger strand | |
| Co-mature arm (3p_co), both arms are about equally abundant |
The same colors mark the arm nucleotides in the structure on the details page. Items can be searched in the position box by MirGeneDB ID, gene name without the -v suffix, miRBase ID, family, and miRBase precursor and mature accessions.
MirGeneDB annotates microRNAs from public small RNA sequencing data, found with miSRA, and genome assemblies. Reads are processed with miSRA or miRTrace, genomes are scanned with MirMachine for conserved families, and candidate loci predicted with MirMiner are curated by hand using synteny and comparisons of the sequences within each family. A locus is accepted only if it meets the MirGeneDB biogenesis criteria, which rely on the pattern of reads on both arms of the hairpin. The arm with more reads is the mature arm, the other the star arm; when the read counts of the two arms are within a factor of two of each other, MirGeneDB calls them co-mature. Version 3.0 has 21,822 microRNA genes from 114 species. See Clarke et al., 2025, for details.
The human annotation was downloaded from https://mirgenedb.org/static/data/hsa/hsa-all.bed, which contains all features of all loci on hg38, together with the GFF file https://mirgenedb.org/gff/hsa?all=1&sort=pos for the miRBase precursor accessions and the human browse table for the miRBase IDs, families and mature accessions. Variant loci (-v2 to -v4) take these from the gene's entry in the browse table, and the precursor accession from their -v1 locus. Of the features in the BED file, the track uses the primary regions (573) as items and the 5p, 3p, 5p*, 3p* and 3p_co arms (1,145) as the colored blocks; the precursor, loop, flank and motif features are not shown. The seed of each locus, nucleotides 2 to 8 of its dominant or co-mature arms, was read from hg38 at the seed features of the BED file and agrees with the browse table for all 514 genes. The seeds of the 59 -v2 to -v4 loci differ from those of their -v1 locus, because their arms are shifted. The sequence of every primary region was taken from hg38 and folded with RNAfold (ViennaRNA 2.7.2, option -d2, default energy parameters) to predict the minimum free energy (MFE) structure. At 32 loci, the precursor part of the MFE structure is not one hairpin, mostly because the flanks pair into the arms, or an arm reaches into the hairpin loop. These loci were refolded with the flanks unpaired (-C), and where that still did not give one hairpin with both arms in the stem, the lowest-energy such structure from RNAsubopt (ViennaRNA 2.7.2) is shown. Hsa-Mir-451 keeps its MFE structure, in which the 5p arm reaches into the loop: its short precursor is processed by Argonaute 2 instead of Dicer, and the arm cannot fit into the stem. The Structure row of the details page says which structure is shown. The steps are documented in the makeDoc file, and the scripts are in the kent source tree.
The data can be explored interactively in table format with the Table Browser or the Data Integrator and exported from there to spreadsheet or tab-separated tables. From scripts, the data can be accessed through our API, track=mirGeneDb.
For automated download and analysis, the annotation is stored in two bigBed files that can be downloaded from our download server: mirGeneDb.bb has the primary regions with all fields shown on the details page, including the sequence and dot-bracket structure, and mirGeneDbArms.bb has the arms. Individual regions or the whole annotation can be obtained using our tool bigBedToBed, which can be compiled from the source code or downloaded as a precompiled binary for your system. Instructions for downloading source code and binaries are on our download page. The tool can also be used to obtain features within a given range, e.g. bigBedToBed http://hgdownload.soe.ucsc.edu/gbdb/hg38/mirGeneDb/mirGeneDb.bb -chrom=chr21 -start=0 -end=46709983 stdout
The original annotation can be downloaded from the MirGeneDB download page.
This track was created by Andrea Tanzer, University of Vienna, from the MirGeneDB 3.0 human annotation.
MirGeneDB is curated by Bastian Fromm at The Arctic University Museum of Norway, UiT The Arctic University of Norway, and Kevin Peterson at Dartmouth College, together with the other members of the MirGeneDB consortium.
Secondary structures were predicted with RNAfold from the ViennaRNA Package, developed by Ronny Lorenz, Ivo Hofacker and colleagues at the University of Vienna. The structure display on the details page uses fornac by Peter Kerpedjiev, Stefan Hammer and Ronny Lorenz.
Clarke AW, Høye E, Hembrom AA, Paynter VM, Vinther J, Wyrożemski Ł, Biryukova I, Formaggioni A, Ovchinnikov V, Herlyn H et al. MirGeneDB 3.0: improved taxonomic sampling, uniform nomenclature of novel conserved microRNA families and updated covariance models. Nucleic Acids Res. 2025 Jan 6;53(D1):D116-D128. PMID: 39673268; PMC: PMC11701709
Kerpedjiev P, Hammer S, Hofacker IL. Forna (force-directed RNA): Simple and effective online RNA secondary structure diagrams. Bioinformatics. 2015 Oct 15;31(20):3377-9. PMID: 26099263; PMC: PMC4595900
Lorenz R, Bernhart SH, Höner Zu Siederdissen C, Tafer H, Flamm C, Stadler PF, Hofacker IL. ViennaRNA Package 2.0. Algorithms Mol Biol. 2011 Nov 24;6:26. PMID: 22115189; PMC: PMC3319429