Skip to content


miRBase 16 and deep sequencing data

It’s taken a while.  miRBase is, after all, more than 8 years old. But we’ve dragged ourselves kicking and screaming into the blogosphere, on the off-chance a miRBase blog might be of some minor interest.  So here you will find, in time, news of miRBase releases, discussion of new features, ideas for future development and more.  All, of course, with questions, discussion, feedback and comment from you!

miRBase 16 has been available for a month or so.  You may or may not have noticed the new interface to view short RNA deep sequencing data mapped to miRNA sequences (see, for example, the dme-mir-282 entry). The patterns of mapped reads allow us to see the diversity of mature miRNA sequences (sometimes called isomiRs) expressed from a given locus.  Read counts, summed over many experiments, act as a proxy for miRNA expression.  We can therefore show the relative abundance of miR, miR* and isomiR sequences.  Reads can be filtered by count, experiment, tissue and stage.  We can also use the read data as detailed evidence for miRNA annotations, and so revisit previous dubious annotations.  A new search interface allows us to search for miRNAs by tissue- or stage-specific expression.  We have so far added reads from just a few datasets imported from GEO, but we’ll be adding many more in the coming weeks and months.  A manuscript describing our approaches for dealing with deep sequencing data has been accepted for publication in the NAR 2011 database issue, so look out for that!

We’re keen to hear what you think, and suggestions for improvements are always welcome, here or by email.

Posted in new features, releases.


29 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Nico says

    great job guys !

  2. Francis Nunes says

    We are planning to use microarray plataform for microRNAs analysis. Most companies have their platforms based on miRBase updates. Recently, 267 novel miRNAs were described for honey bees (Apis mellifera) by Chen et al (2010), published in Insect Molecular Biology journal. I presume the authors sent the novel sequences to miRBase, but only after the release 16 became avaiable. Please, could you tell me:
    1- if you already know these sequences
    2 – when will be the next release
    3- if these sequences will be avaiable in the release #17

    Thanks in advance.

    Best,
    Francis

    • sam says

      Yep, those sequences will be in the next release. I’m afraid we don’t yet have an anticipated release date for miRBase 17. We are aiming to work towards a schedule of pre-announced and regular releases, but we’re a little way from that right now.

    • Christoph says

      Hi Francis,
      If you have or can get the sequence information from the authors you can use custom arrays – there companies that offer this service at reasonable cost. Not sure if it’s ok to mention names here but you can just google “custom miRNA microarrays”.
      Good luck – Christoph

  3. Kevin says

    How to find the pri-miRNA sequence?

  4. Moreno Colaiacovo says

    miRBase is an invaluable resource for research in the miRNA field. However, I have three suggestions that will in my opinion would improve the usefulness of the database.

    1. What about building a section with all the miRNA-related papers which are published daily? In this way, one could be sure not to miss any important publication on this subject.

    2. Wouldn’t be a good idea to force people who experimentally validate miRNAs to add them to the database? As far as I know it is not compulsory.

    3. What do you think about creating a better classification system for miRNAs? I think it’s correct to classify them on the basis of the precursor, but often they act accordingly to their mature sequence, so it would be nice to have a sort of classification level that takes into account also the mature sequence similarity. It could happen that miRNAs which have the same mature sequence (and consequently target the same mRNA) are in fact classified in different families because their precursors are different.

    Please keep on doing the precious work you’ve done until now

    Best regards
    Moreno

    • sam says

      Hi Moreno

      1. What about building a section with all the miRNA-related papers which are
      published daily? In this way, one could be sure not to miss any important
      publication on this subject.

      Nice idea, but it’s tough for us to resource this right now. The miRNA blog is the closest thing I know to this. Automatic updates from PubMed or PubCrawler are a good way to do this on a personal level.

      2. Wouldn’t be a good idea to force people who experimentally validate miRNAs to add them to the database? As far as I know it is not compulsory.

      It kinda is compulsory! This is enforced by journal editors, and some are more likely to know about it than others. See the original guidelines here. Basically, authors should submit sequences to miRBase after their manuscript is accepted. We then assign names quickly enough to get them used in the final text of the paper.

      3. What do you think about creating a better classification system for miRNAs? I think it’s correct to classify them on the basis of the precursor, but often they act accordingly to their mature sequence, so it would be nice to have a sort of classification level that takes into account also the mature sequence similarity.

      You’re talking about the family classification, right? The miRBase families are currently not that great, but they are supposed to represent gene families at the moment. You’re correct of course that we should also build families based on the mature sequences, and this is already on the list.

  5. Veronica says

    I would like to know how miRNAs sequence human mature and immature are described so far in miRBase
    Thanks

  6. Wang Qinhu says

    Does your new NAR paper supply a tool for mapping and visualization deep reads mapped to miRNA sequences? When will the paper be available?
    Thanks.

    • sam says

      Not exactly. We’re not (at least currently) aiming to produce a standalone tool to map and visualise your own data, rather an interface to see datasets that are already published in GEO. Other people may be able to post suggestions here of the most useful tools to do what you want.

      I’m not sure exactly when the paper will be available, but I’m guessing advanced access is imminent. I’ll, of course, post here when it’s out.

      • sam says

        I’m not sure exactly when the paper will be available, but I’m guessing advanced access is imminent. I’ll, of course, post here when it’s out.

        It is now! See this post.

    • Petrus Tang says

      If you are interested in the visulaization of reads mapped to miRNA sequences, you can try the deep sequecning smallRNA analysis pipeline (DSAP) http://dsap.cgu.edu.tw.
      Wish that it helps to solve part of your problem.

  7. yjhua2110 says

    Last year, we have published a novel database, deepBase ( http://deepbase.sysu.edu.cn/ ), which we have developed to facilitate the comprehensive annotation and discovery of small RNAs from deep-sequencing data.

    We have mapped reads from deep-sequencing experiments to microRNAs downloaded from miRBase (release v13) and developed a variety of interfaces and graphical visualization to view these mapped reads (http://deepbase.sysu.edu.cn/browseNasRNA.php and http://deepbase.sysu.edu.cn/browseExpress.php).

    Do you known what our deepBase database have provided?

    I would like to find you to communicate and improve our deepBase database ( http://deepbase.sysu.edu.cn/ ).

    • sam says

      Thanks for this. Several people have pointed me at your deepbase recently. Naturally, we will be focussing on miRNA data, whereas you are doing much more! Right now, we are primarily aiming to facilitate queries by expression, and assessing confidence of individual miRNA annotations. Of course, we’re very happy to talk about common goals and different approaches any time.

      • yjhua2110 says

        Thanks sam!

  8. Jack says

    Thank you for providing the very useful update. I have been using miRBase as a resource for some time now and have moved into the realm of NGS these days.

    I have one question about future release annotation for NGS related entries. Do you forsee adding isomir (isoform) specific identifiers for mature sequences? This could prove quite useful when evaluating the most abundant forms in differential expression experiments.

    I apologize if this has been adressed previously.

    Best,
    Jack

    • sam says

      I don’t think we can name individual isomirs (at least right now I don’t see a way we can do that sensibly) — there will be huge numbers of them, and a human-readable nomenclature would probably therefore have little use anyway. However, stable accessions are a possibility — indeed, each distinct read sequence currently gets an accession, which should be stable, and therefore can be useful to refer to the specific isoform it represents.

  9. Francis Nunes says

    Hi all,

    I´d like to know if the release #17 is coming, and also if the novel ant miRNAs (from genomes recently published of Pogonomyrmex barbatus, Linepithema humile, Solenopsis invicta and Atta cephalotes) will be avaiable.

    Best, Francis

  10. sam says

    Release 17 should be along within the next couple of months — I’m afraid there is no definite timeline yet. I saw the ant genome papers, but as far as I can see, noone has published microRNA sets yet have they? Please point me in the right direction if I’ve missed something. In the absence of those data, miRBase 17 won’t contain ant microRNAs. At some point we will likely do the prediction of conserved microRNAs in those genomes, but that won’t be fast.

    • Jennifer Williams says

      Hi Sam,
      Is there a timeline for release 17 yet?
      Thanks for any info!

      • sam says

        We’re working on a release right now. We don’t have a firm date yet, but are aiming for before the end of the month. I’ll post here a predicted release date when we have one.

  11. Francis Nunes says

    Hi Sam,
    from genomic data of six ant species, I found miRNA predictions (sometimes validation by qPCR) for Camponotus floridanus (96 miRNAs) and Harpegnathos saltator (159), Atta cephalotes (57), and Linepithema humile (70). However, precursor and mature sequences were not presented neither in the main paper nor in the supporting information. I hope authors get in touch with you soon.

    • sam says

      OK — thanks Frances!

  12. Dubbe says

    Hello, how to find the pre-miRNA?

  13. sam says

    The hairpin sequences we have (for example, on every entry page, and on the FTP site: ftp://mirbase.org/pub/mirbase/CURRENT/hairpin.fa.gz) include the pre-miRNA (the product of Drosha cleavage), but also usually contains additional residues either side. That additional sequence is of somewhat arbitrary length, and often represents the extent of the predicted hairpin. If you want the true pre-miRNA sequence, I’m afraid you need to do some work. Where there are mature sequences defined from both arms of the hairpin, the ends of the mature sequences define the pre-miRNA. If there is only one mature sequence annotated from the hairpin, the mature sequence defines one end of the precursor, and the other can be predicted based on the 2 nt 3′ overhang. We have a plan to allow users to retrieve the pre-miRNA with or without a flanking sequence, but, as usual, I’m afraid that will take a little time.