[BioC] R: How can I trace back transcription target genes from the miRNAs file downloadable from miRbase ?

[Sean Davis]

On Fri, Oct 2, 2009 at 7:51 AM,  <mauede at alice.it> wrote:
> My first task  was to download a set (as big as posssible) of experimentally
> Validated miRNAs from miRecords with their relative target genes
> and the 3'UTR sequences., limited to Homo sapiens.
> The XLS file  from miRecords related the miRNA identier ("hsa-miR-xxx) with
> its target genes identifier. I never found a clear way to download
> the miRNA sequence and the relative target 3'UTR sequence from miRecords.
> The many different links  bring to pages of sequences that
> are not expressively stated to be what I need.  Therefore I downloaded the
> Validated  miRNAs file from miRbase, matched the miRNA identifier
> with miRecords to get the miRNA sequence. Then I used the gene identifier
> (NM_yyyy) from miRecords to quey BioMart and get the 3'UTR sequences.
> There are many unresolved miRNAs because I cannot find an exact match
> between the miRecords and miRbase. For example in mirBase I
> found two miRNAs whose identifiers differ only by the last digit but their
> sequences are different beyond the seed region so their are (I think)
> two different entities:
>
> hsa-miR-26a-1* MIMAT0004499 Homo sapiens miR-26a-1*
>                            "CCUAUUCUUGGUUACUUGCACG"
>> val_miRNA[830]
> hsa-miR-26a-2* MIMAT0004681 Homo sapiens miR-26a-2*
>                            "CCUAUUCUUGAUUACUUGUUUC"

Please read:

http://www.mirbase.org/help/nomenclature.shtml

The point is that the nomenclature for miRNAs is only a weak surrogate
for sequence information.

> miRecords XLS file only contains "hsa-miR-26a"  that I cannot match to
> either one above mentioned.
>  I can only use the Validated miRNAs from miRecords for which I find a match
> in mirBase.
>
> My question is: if I restrict my search to mirBase, where can I find the
> experimentally Validated (not just predicted)
> target genes associated to the miRNAs in the downloadable files containing
> records like the above shown ones ?

I would suggest that you move your queries about the specifics of the
mirbase and mirecords databases to the respective developers.  Both
miRecords and mirbase supply email links.

> The data MIMATsssss does not seem to bring me anywhere ....
> To complete my task I have to find the Validated target identifiers (for
> instance NM_xxxxxx) and then use this data to
> query BioMart and get the 3'UTR sequences.

I think you are probably best directing your questions of specifics of
the databases and their contents to their respective developers.

Sean

> -----Messaggio originale-----
> Da: Sean Davis [mailto:seandavi at gmail.com]
> Inviato: ven 02/10/2009 4.22
> A: mauede at alice.it
> Cc: Bioconductor List
> Oggetto: Re: [BioC] How can I trace back transcription target genes from the
> miRNAs file downloadable from miRbase ?
>
> On Thu, Oct 1, 2009 at 9:27 PM,  <mauede at alice.it> wrote:
>> I downloaded the Validated miRNAs files (mirbase/CURRENT/mature.fa.gz ,
>> maturestar.fa).
>> How can I trace back to the gene  transcription sequence for the genes
>> that targeted any specifi miRNA '
>> Thank you in advance,
>> Maura
>
> Hi, Maura.  Are you asking to find the targets of miRNAs?  Or are you
> asking for the sequences of transcripts?
>
> Sean
>
>
> Alice Messenger ;-) chatti anche con gli amici di Windows Live Messenger e
> tutti i telefonini TIM!
> Vai su http://maileservizi.alice.it/alice_messenger/index.html?pmk=footer