[ensembl-dev] "Use of uninitialized value in split at .vep/Plugins/LoF.pm" Error Messages

Will McLaren wm2 at ebi.ac.uk
Fri Mar 11 09:37:12 GMT 2016


Hi Wei, Konrad,

>From the code it looks as if this is giving this warning when exon_number
is not defined. This can be the case if the variant does not fall in an
exon, so the plugin should really be checking if it is defined before you
attempt to split it.

On the other hand, my guess is by that point in the code you would hope to
be only operating on exonic variants, so either the variant that is causing
the warning is being called exonic when it isn't, or it is exonic and the
exon_number routine isn't working properly.

The only situation I can think of that might be the case is if the variant
falls in one of the very rare frameshift introns that occur in some Ensembl
transcripts. These are very short (<=12bp) artificial introns that are
introduced into a transcript's structure in order to make the structure,
genome sequence and transcript evidence play nicely together.

If either of you do track down a variant that causes this, and there is
something to fix on the VEP side, please let us know.

Regarding capturing errors, VEP does capture it's own warnings via the
warning_msg() subroutine (they can be directed to STDERR or any output
file), but I don't think it can capture warnings like this. Plugins are run
in an eval block, so errors are caught (and reported via warning_msg()),
but I don't think it can capture warnings like this.

Regards

Will McLaren
Ensembl Variation

On 10 March 2016 at 17:05, Konrad Karczewski <konradk at broadinstitute.org>
wrote:

> Hi Wei,
>
> We've had this issue sporadically as well, but we have not yet been able
> to pinpoint the variants that cause this error. It's on our todo list to
> fix, but it seems rare enough and hasn't been prominent enough to get
> noticed in any manual inspections that we do.
>
> On that note, Ensembl dev folk/Will, is there a way in VEP to provide a
> bit more error logging? I should have asked this a long time ago, but it
> would be great to have a logging level that would print which variant an
> error happens on.
>
> -Konrad
>
> On March 10, 2016 at 11:57:21 AM, Wei Zhao (mikezhaow at gmail.com) wrote:
>
> Dear LOFTEE plugin developers,
>
> I met with this error message when I was using the LOFTEE plugin:
>
> "Use of uninitialized value in split at .vep/Plugins/LoF.pm line 378"
>
> Both VEP and LOFTEE ran smoothly without a problem.  But I just saw this
> error message in the error files returned by our clusters' queuing system.
> Just want to know if this is something I should worry about.  Thank you for
> your kind help!
>
> I am currently using VEP v83 and Perl v5.20.1.
>
> Best,
> Wei
> _______________________________________________
> 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/
>
>
> _______________________________________________
> 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/20160311/9bc458f2/attachment.html>


More information about the Dev mailing list