[ensembl-dev] Some RefSeq transcripts seem broken

[Will McLaren]

Yes, my reply still stands.

The protein sequence as stored in $tva->transcript_variation->_peptide is
extracted from the protein sequence as constructed by the Ensembl API (from
the imported RefSeq coords and the genomic sequence), *not* from the
protein sequence as given in the RefSeq record. The reasons for this are
explained in my initial reply.

Unfortunately this is how our whole API is designed, and we are currently
looking for better solutions at dealing with situations where the RefSeq
sequence differs from the genome.

Hopefully that makes more sense?

Will

On 27 September 2016 at 12:57, João Eiras <joao.eiras at gmail.com> wrote:

> Hi.
>
> Thanks for the reply. But I might not have been clear enough.
>
> This is how the plugin looks like
>
> sub run {
> my ($self, $tva) = @_;
> return {reference => $tva->transcript_variation->_peptide};
> }
>
> As you can see, it accesses _peptide which is the wild type sequence
> (I hope, I got that line from the ProteinSeqs plugin).
>
> So, regardless of whether the variant affects the transcript or, I'm
> accessing the wild type sequence... right ?
>
> I used that variant (which is a frameshift) as an example, but could
> just as well do
>
> #CHROM POS ID REF ALT QUAL FILTER INFO FORMAT
> chr5 38300282 . GTC AAA 5000 . . .
>
> the point of the variant it just to access the transcript (since I
> don't know how to query the VEP database directly).
>
> VEP will then report three annotations in transcripts
> ENSMUST00000063136, ENSMUST00000114099 and NM_172709.3. The reference
> sequence for transcripts ENSMUST00000114099 and NM_172709.3 should be
> the same, but they aren't.
>
> So, I'm concerned that the refseq transcripts are not being correctly
> stored.
>
> Thank you.
>
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> Posting guidelines and subscribe/unsubscribe info:
> http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20160927/e0b6fbb2/attachment.html>