[ensembl-dev] can VEP output the distance of the variant to the exon junction?

[David Tamborero]

thanks for the tip, I will check this out too!
br
d

El vie., 8 mar. 2019 a las 10:33, Irina Armean (<iarmean at ebi.ac.uk>)
escribió:

> Hi David,
>
> thanks for the feedback! Yes, the question of reporting the distance of a
> variant to the nearest exon junction has come up before.
>
> In your particular case, are you interested in identifying stop gained and
> frameshift variants that are/aren't covered by NMD using the 50 bp rule, or
> having your own customizable rule?
>
> If you are interested in identifying stop gained and frameshifts that
> escape NMD (50bp rule) then this functionality is available and refined in LOFTEE
> (https://github.com/konradjk/loftee); it annotates such variants with LoF
> LC END_TRUNC, the plugin is taking into consideration UTR splicing and GERP
> scores.
>
> If there are cases that you are interested in and are not covered by the
> existing plugins, I would be interested to know. Also, we encourage and
> accept pull requests on the VEP_plugins (
> https://github.com/Ensembl/VEP_plugins) git repository for plugins that
> would be for wider interest.
>
> Kind regards,
>
> Irina
> On 07/03/2019 17:39, David Tamborero wrote:
>
> Hi Irina, I have been playing with the plugin (thanks for that again!).
>
> What I am actually interested is in the distance of the variant to the
> next exon-exon junction, as defined by the CDS and not by the 'absolute'
> bps (including UTR), which is what determines whether the NMD will act (see
> here
> <https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1002296>),
> which is what i want to detect. I was not clear enough on that, apologies.
>
> In any case, what you re retrieving now is actually also interesting for
> other evaluations. Maybe you could include what you have now and also the
> coding sequence values ? Again, note that I address these calculations by
> my own ad hoc scripts, so consider to implement it (together with the
> location of the frameshift-induced stop codon mentioned before) just if you
> believe that it can be of interest for the community (which i think it
> does!)
>
> br
> d
>
> El lun., 4 mar. 2019 a las 18:33, David Tamborero (<
> david.tamborero at gmail.com>) escribió:
>
>> thta s great, i m currently travelling but i will try it as soon as i
>> back to the lab (and feedback you guys as appropriate)
>>
>> thanks again!
>> d
>>
>> El vie., 1 mar. 2019 a las 16:27, Irina Armean (<iarmean at ebi.ac.uk>)
>> escribió:
>>
>>> Hi David,
>>>
>>> The NearestExonJB.pm plugin is now available on github
>>> https://github.com/Ensembl/VEP_plugins. This plugin reports the nearest
>>> exon junction boundary (exon id, bp distance, start or end, exon length)
>>> within the affected transcript.
>>>
>>> I hope you find it useful.
>>>
>>> Best regards,
>>>
>>> Irina
>>> On 25/02/2019 08:40, David Tamborero wrote:
>>>
>>> That s fantastic, thanks!
>>>
>>> Have a great week
>>> D
>>>
>>> El vie., 22 feb. 2019 17:48, Irina Armean <iarmean at ebi.ac.uk> escribió:
>>>
>>>> Hi David,
>>>>
>>>> Thanks for the clarification and the positive feedback for the team. We
>>>> do our best to make the desired functionality available in due time,
>>>> sometimes it's easier/faster than other times.
>>>>
>>>> We plan to add a VEP plugin within the next week, which will report the
>>>> nearest exon junction boundary (exon id, bp distance, start or end, exon
>>>> length) for variants in the protein coding region, this should answer your
>>>> case and also provide some additional use cases.
>>>>
>>>> For the case of frameshifts and premature stop codons, they are handled
>>>> more efficiently in the main VEP code and is more complex to update quickly
>>>> however we plan to cover this use case in the near future.
>>>>
>>>> I will come back with a mail once the plugin is in.
>>>>
>>>> Best regards,
>>>>
>>>> Irina
>>>>
>>>>
>>>> On 21/02/2019 17:17, David Tamborero wrote:
>>>>
>>>> Hi Irina, thanks for your answer, it is always awesome how diligent you
>>>> are guys;
>>>>
>>>> my wishlist:
>>>>
>>>> - in my case, what i m interested in is in the distance (nucleotides)
>>>> from my variant 'within' the exon (ie protein coding-affecting variant) to
>>>> the next exon junction. But note that other users may be interested in
>>>> other distances, as those that you mention.
>>>>
>>>> - FYI, i m interested in that info for the variants that create
>>>> preamture stop codons, so I can estimate whether it will be likely that the
>>>> non-mediated decay is triggered (which depends on how far it falls
>>>> regarding the last junction of the transcript). Note that --for the same
>>>> reason-- I m also interested in, given a frameshift mutation, where the new
>>>> 'cryptic' stop codon will be placed. I currently calculate it by actually
>>>> using the VEP info that can be parsed from the HGVSp column (e.g.
>>>> p.Glu5ValfsTer5, meaning the stop codon in 5 codons after the first
>>>> 'ectopic codon' created by the frameshift, if I remember well).
>>>>
>>>> Again, I currently address both cases by my own ad hoc scripts, but I
>>>> thought that you already 'have' the info to easily output these things,
>>>> which I think it can be of interest for the community (and surely more neat
>>>> that my creepy coding!)
>>>>
>>>> hope it makes sense!
>>>> br
>>>> d
>>>>
>>>> El jue., 21 feb. 2019 a las 17:12, Irina Armean (<iarmean at ebi.ac.uk>)
>>>> escribió:
>>>>
>>>>> Dear David,
>>>>>
>>>>> Thanks for the question. We currently don't have this functionality
>>>>> however we can easily add the distance to the closest exon.
>>>>>
>>>>> To clarify, are you interested in getting the distance to the nearest
>>>>> exon junction for the variants only when it affects a protein coding
>>>>> transcript or also when they are upstream/downstream of genes for example?
>>>>>
>>>>> Best regards,
>>>>>
>>>>> Irina
>>>>>
>>>>>
>>>>> On 21/02/2019 11:09, David Tamborero wrote:
>>>>>
>>>>> Dear all,
>>>>>
>>>>> is any flag (which I cannot find) so the VEP output includes the
>>>>> distance of the (coding) variant to the closest exon junction?
>>>>>
>>>>> If not, somebody has a recommended fancy approach to do so? (My
>>>>> current approach is 'just' to match with my own script the VEP output for a
>>>>> given transcript with the exon coordinates data that can be retrieved in
>>>>> e.g. BioMart)
>>>>>
>>>>> many thanks in advance!
>>>>> d
>>>>>
>>>>> _______________________________________________
>>>>> 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/
>>>>>
>>>>> --
>>>>> Irina Armean PhD, Bioinformatican
>>>>> Ensembl Variation, EMBL-EBI, Hinxton, UK
>>>>> iarmean[at]ebi.ac.uk | A3 - 135
>>>>>
>>>>> --
>>>>
>>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20190308/bb0b79dc/attachment.html>