[ensembl-dev] Variant Consequence Predictor

[Will McLaren]

Hi Stuart,

I'm not 100% sure what's happening here, but we have the variation on our
website as having alleles C/G, not C/T.

http://www.ensembl.org/Homo_sapiens/Variation/Mappings?r=11:76370918-76371918;v=rs79525962;vdb=variation;vf=1164142

This data comes from dbSNP, however, if you look on the population genetics
page, we have imported data directly from 1000 genomes VCF files, and there
the alleles are given as C/T. So there is some disagreement between the
dbSNP alleles and the 1000 genomes alleles. This may well be resolved in
some future release of dbSNP - in fact if you look at the record on dbSNP,
they also show the C/T alleles being reported under the 1000 genomes
submissions.

http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=rs79525962

but for some reason they haven't added the "T" to the list of alleles.

Basically, the results are OK for the alleles given - C/T gives a peptide
change of A/T, whereas C/G gives a peptide change of A/P. So neither is
necessarily "correct", it just depends what you want to interpret as being
the observed alleles at that position. Of course, there is also the possible
situation where in fact three alleles have genuinely been observed (C/G/T)
at that position, which means they would both be correct!

Hope this helps

Will

On 25 February 2011 09:23, Stuart Meacham <sm766 at cam.ac.uk> wrote:

> Hi list (Will!),
>
> I have been using the Variant Consequence predictor in whole genome mode
> and have (at least one) anomalous results. For example:
>
> Transcript: ENST00000260061
>
> SNP Residue Major Minor AA_Replacement Consequence
>
> rs79525962 407 C T A/P NON_SYNONYMOUS_CODING
> 11_76371418_C/T 407 C T A/T NON_SYNONYMOUS_CODING
>
> top line is data in Ensembl
> second line is data (1000 Genomes) using the predictor. You can see the
> Consequence predictor shows an amino acid change (A/T) at the same position
> as Ensembl shows A/P.
>
> Is one just wrong? And if so how do I check which one?
>
> Cheers
>
> Stuart
>
> _______________________________________________
> Dev mailing list
> Dev at ensembl.org
> http://lists.ensembl.org/mailman/listinfo/dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20110225/bae318b7/attachment.html>