[ensembl-dev] VEP - HGVS duplication reported as insertion

[Sarah Hunt]

Hi Dani,

I'm sorry to hear you are having this problem, but we are unable to 
replicate it.

We have observed problems with FASTA index corruption which would cause 
this issue - try deleting 
$HOME/.vep/[species]/[assembly]/[fastafile].index (or .fai) and 
re-running VEP to regenerate it.

Make sure you are using a >=1.6 version of BioPerl - older versions can 
have this problem with large files. The VEP installer installs 1.6.1 for 
you.

Also, try re-installing the API to ensure you are using the precise 
version we tested.

I hope this helps,

Sarah

On 08/01/2016 12:54, Daniel Borras wrote:
> Dear Sir, Madame,
>
> My name is Daniel Borras and I am currently working with your tool 
> Variant Effect Predictor. I usually input HGVS variants obtained from 
> other sources to annotate their effect. However I noticed an 
> unexpected behaviour when it comes to duplications. When feeding VEP 
> with an HGVS duplication this is transformed to an insertion, please 
> the real output below. According to HGVS recommendations, duplications 
> are reported differently than insertions, let me quote HGVS 
> recommendations: g.5dupT (or g.5dup, */not g.5_6insT/*) . This 
> wouldn't be a big issue if it wasn't because for downstream analysis 
> the annotation of the variant changes making previous comparisons and 
> checks to fail since the variant description is different. I believe 
> that this behaviour should be patched, probably is not too complicated 
> to fix, and will make VEP to report correct duplications in HGVS notation.
>
> The results are I obtained are:
> Input: *NM_001009944.2:c.4248dup*
> Output:*NM_001009944.2:c.4248_4249insT*
> Row: chr16    2160919 *NM_001009944.2:c.4248dup*  C    CA    .    . 
> A|frameshift_variant|HIGH|PKD1|5310|Transcript|NM_001009944.2|protein_coding|15/46||*NM_001009944.2:c.4248_4249insT*|NP_001009944.2:p.Gly1417TrpfsTer14|4457-4458|4248-4249|1416-1417|-/X|-/T|||-1|insertion|||YES|||NP_001009944.2||||rseq_mrna_nonmatch&rseq_cds_mismatch&rseq_ens_match_wt|||||||||||||||||||||
>
> The results were obtained by using:
> VEP version: ensembl-tools-release-80
> Assembly: GRCh37
> Cache_version: 80
> Port: 3337
>
>
> Best,
> Dani
>
>
>
>
> _______________________________________________
> 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/20160108/478a5f04/attachment.html>