[ensembl-dev] [SPAM] - Re: [SPAM] - Re: [SPAM] - Re: Transcript variation alleles - Email found in subject - Email found in subject - Email found in subject

[Will McLaren]

### just re-posting this to dev as the list didn't like the message getting
too long ###



A variation is defined by its alleles and a pair of flanking sequence, e.g.:

ATCGTACTGTACGTGTTTATCG [A/G] TGACTTACTATCGTATGACTT

dbSNP (or in some cases Ensembl) use sequence alignment algorithms to map
this sequence to the genomic reference sequence. If it maps to the reverse
strand, we create a variation_feature on the reverse strand. In some cases
the sequence may map more than once, hence a variation object can have
multiple associated variation_feature objects. This is easier to see on the
web views:

http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=rs78197743;vdb=variation

As I mentioned, it makes our lives easier if most things are on the forward
strand, so we "flip" as many of these reverse strand mappings as we can. We
only do this when a variation has a single mapping to the reverse strand.

Will
- Hide quoted text -


On 2 February 2011 16:17, Oliver, Gavin <gavin.oliver at almacgroup.com> wrote:

> Cool, thanks.
>
>
>
> What determines the strand it goes on?
>
>
> ------------------------------
>
> *From:* wmclaren at gmail.com [mailto:wmclaren at gmail.com] *On Behalf Of *Will
> McLaren
> *Sent:* 02 February 2011 16:15
>
> *To:* Oliver, Gavin
> *Cc:* Graham Ritchie; dev at ensembl.org
> *Subject:* Re: [SPAM] - Re: [SPAM] - Re: [SPAM] - Re: [ensembl-dev]
> Transcript variation alleles - Email found in subject - Email found in
> subject - Email found in subject
>
>
>
> variation_feature objects have a strand just like all features in Ensembl -
> most are on the forward strand (as this makes our life easier!) but many are
> also on the reverse strand.
>
>
>
> As I mentioned earlier, you can check this using:
>
>
>
> $vf->seq_region_strand
>
>
>
> which will return a value of 1 (forward) or -1 (reverse).
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20110202/6151884a/attachment.html>