[ensembl-dev] FW: Mutations in VEP

Sarah Hunt seh at ebi.ac.uk
Tue May 23 13:10:51 BST 2017


Hi Wendy,

Yes - you are interpreting this correctly

KMT2A:c.9982delCinsCA is not compliant HGVS, but we try to support and 
correct such descriptions. The HGVS version we currently support 
requires the alleles be written in their minimal representation - which 
would be an A not CA - and the change described at the 3' most position. 
This allele should therefore be described as c.9983dupA. The prefixing 
of the C appears to be confounding the 3' shift in the first instance - 
we will fix this for our July release.

The latest version of HGVS states the alleles should not be listed, so 
the annotation would be c.9983dup. We plan to update to the new HGVS 
version in July.

Best wishes,

Sarah


From: *Wendy Jones* <wj1 at sanger.ac.uk <mailto:wj1 at sanger.ac.uk>>
Date: 22 May 2017 at 17:08
Subject: FW: Mutations in VEP
To: "dev at ensembl.org <mailto:dev at ensembl.org>" <dev at ensembl.org 
<mailto:dev at ensembl.org>>
Cc: "wm2 at ebi.ac.uk <mailto:wm2 at ebi.ac.uk>" <wm2 at ebi.ac.uk 
<mailto:wm2 at ebi.ac.uk>>


Dear Colleagues,

I wonder if you might be able to answer this quesion in Will's absence?

Many thanks

Best wishes

Wendy
------------------------------------------------------------------------
*Van:* Wendy Jones
*Verzonden:* zondag 21 mei 2017 21:39
*Aan:* wm2 at ebi.ac.uk <mailto:wm2 at ebi.ac.uk>
*Onderwerp:* Mutations in VEP

Dear Will,

I wonder if you might be able to help me.  I am a clinician / staff 
scientist working at the Sanger Institute.  I am trying to interpret 
mutations in /KMT2A/ (human) I have been given by clinicians to put into 
a paper using Assembly: GRCh37.p13. Frustratingly I am often only given 
the coding variant and the predicted protein variant by clinicians (i.e. 
not the genomic co-ordinates or transcript ID).  I am using VEP to find 
out further information about them for example the HGVS coding variant 
location for my transcript of interest ENST00000534358.1, and also to 
work out their genomic co-ordinates.  I find for the frameshift 
mutations that the coding variant position often changes when I ask for 
coding variant in line with HGVS nomenclature (I concluded that this was 
representing re-alignment due to repetitive sequence to the most 3' 
position in line with the way HGVS places variants).  I have been 
running variants through VEP twice to ensure the genomic co-ordinates 
represent the HGVS coding variant (If I have understood the way VEP 
works correctly if you put in a variant the genomic co-ordinates 
represent the position of the variant you entered, however if HGVS 
interprets the variant as being in a slightly different position the 
genomic co-ordinates will be slightly out.

However I have noticed that when putting the following variants through 
VEP a second time the HGVS coding variant nomenclature changes again!  I 
wonder if you might be able to explain why this is?

For example KMT2A:c.9982delCinsCArun through VEP, HGVS nomenclature 
coding variant name changes this to KMT2A:c.9982_9983insA, however when 
run through VEP a second time this becomes: c.9983dupA.

It is also a similar story for the following variants:

KMT2A:c.5166delT
KMT2A:c.5999_6002delTGTT
KMT2A:c.8098_8105delCTGGCATC

I wonder if you might be able to explain why the variants continue to 
change (and what the correct variant is?). And also if I have 
interpreted the way VEP calculates genomic co-ordinates correctly?

Many thanks for your help,

Best wishes

Wendy

-- The Wellcome Trust Sanger Institute is operated by Genome Research 
Limited, a charity registered in England with number 1021457 and a 
company registered in England with number 2742969, whose registered 
office is 215 Euston Road, London, NW1 2BE.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20170523/9dd0f54c/attachment.html>


More information about the Dev mailing list