[ensembl-dev] SNP consequence prediction

[Will McLaren]

Hi Andrea,

Answers below

Cheers

Will

On 9 November 2010 14:57, Andrea Edwards <edwardsa at cs.man.ac.uk> wrote:
> Hello
>
> question 1
>
> I was looking at the different types of consequences you can get for a SNP
> with the ensembl api and I would like to cliary the difference between
> splice site and essential splice site
>
> essential splice site:
> first/last 2 bp of an intron
> NOT in exon at all?

Exactly - the variation must fall in the 2bp splice donor or acceptor
site of the intron.

>
> splice site:
> 1-3bp into exon (does this mean at either the start or the end of the exon?)
> 3-8bp into intron (does this mean at either the start or the end of the
> intron?)

Correct; the variation must fall in a region spanning the intron/exon
boundary - this region extends slightly further into the intron than
the exon. In terms of the intronic part, obviously if it's in the
first/last 2 bases then it will be essential splice site as described
above.

>
> ===========================
>
> question 2
>
> Does the api offer any consequence prediction if the SNP is found to be in
> an ncRNA? I didn't see any type of RNA feature on the list of option. Might
> an RNA such as an miRNA be classed as a regulatory region ?

If a variation falls within a non-coding RNA it can be classified as
"WITHIN_MATURE_miRNA" or "WITHIN_NON_CODING_GENE" depending on the
biotype of the transcript. They are not classed as regulatory regions.

>
> Is there a list of the genetic elements that are represented by the
> 'regulatory region' category

The regulatory region calling is at the moment limited - the variation
must fall in a regulatory feature (termed an ExternalFeature by the
Ensembl regulation API) that is linked via an external reference to a
particular transcript. You can retrieve these features using the
ExternalFeatureAdaptor in the Ensembl regulation API.

>
> ====================
>
> Many thamks
>
>
>
> _______________________________________________
> Dev mailing list
> Dev at ensembl.org
> http://lists.ensembl.org/mailman/listinfo/dev
>