[ensembl-dev] variant_effect_predictor.pl: warning/error using SNPs on chrs with mixed character/numeric vals

Will McLaren wm2 at ebi.ac.uk
Fri Feb 7 09:05:31 GMT 2014


Hello Janet,

I can confirm this is just a warning and it won't affect your output - I'll
create a fix so the warning doesn't appear in future.

PS don't worry about your BioPerl install being newer - for the purposes of
using the VEP the latest version is fine.

Regards

Will McLaren
Ensembl Variation


On 6 February 2014 18:53, Janet Young <jayoung at fhcrc.org> wrote:

> Hi there,
>
> I think I've found a very minor issue with variant_effect_predictor.pl.
>  As far as I can tell, it's not affecting the output in any major way
> (might cause sorting issues - not sure). I'm still at the early stages of
> using the script, so it's also possible I've messed something up.
>
> Anyway, I'm currently trying the script on some of our drosophila data. If
> my vcf file has SNPs  from a mix of chromosomes with numeric names (e.g. 4)
> and character names (e.g. 3RHet), I get a sorting error.   If I split the
> vcf file so that the chr 4 and chr 3RHet SNPs are in separate files,
> there's no error.
>
> The output looks OK, I think, but it still seemed worth letting you know
> about this (I always get a little paranoid that output will be messed up
> when I see any error - it'd be nice to get it fixed at some point).
> Looking at VEP.pm I think I can see why that error comes up.
>
> I've attached a small example input file. Here's the error:
> Argument "3RHet" isn't numeric in numeric comparison (<=>) at
> /home/jayoung/malik_lab_shared/perl/ensembl-variation/modules/Bio/EnsEMBL/Variation/Utils/VEP.pm
> line 1235, <GEN0> line 6.
>
> and here's the full command and output:
>
> variant_effect_predictor.pl --cache --dir
> /home/jayoung/malik_lab/public_databases/Ensembl/VariantEffectPredictorCache
> --species drosophila_melanogaster --input_file
> testCombined3RHetand4small.vcf -o testCombined3RHetand4small.vcf.annotLocal
>
> 2014-02-06 10:39:31 - Read existing cache info
> 2014-02-06 10:39:35 - Starting...
> 2014-02-06 10:39:35 - Detected format of input file as vcf
> 2014-02-06 10:39:35 - Read 6 variants into buffer
> Argument "3RHet" isn't numeric in numeric comparison (<=>) at
> /home/jayoung/malik_lab_shared/perl/ensembl-variation/modules/Bio/EnsEMBL/Variation/Utils/VEP.pm
> line 1235, <GEN0> line 6.
> 2014-02-06 10:39:35 - Reading transcript data from cache and/or database
> [================================================================================================]
>  [ 100% ]
> 2014-02-06 10:39:35 - Retrieved 253 transcripts (0 mem, 253 cached, 0 DB,
> 0 duplicates)
> 2014-02-06 10:39:35 - Analyzing chromosome 3RHet
> 2014-02-06 10:39:35 - Analyzing variants
> [================================================================================================]
>  [ 100% ]
> 2014-02-06 10:39:35 - Calculating consequences
> [================================================================================================]
>  [ 100% ]
> 2014-02-06 10:39:35 - Analyzing chromosome 4
> 2014-02-06 10:39:35 - Analyzing variants
> [================================================================================================]
>  [ 100% ]
> 2014-02-06 10:39:35 - Calculating consequences
> [================================================================================================]
>  [ 100% ]
> 2014-02-06 10:39:35 - Processed 6 total variants (6 vars/sec, 6 vars/sec
> total)
> 2014-02-06 10:39:35 - Wrote stats summary to
> testCombined3RHetand4small.vcf.annotLocal_summary.html
> 2014-02-06 10:39:35 - Finished!
>
> I have a current version of variant_effect_predictor.pl (74), and should
> have a current version of the ensembl perl API (v74, downloaded Jan 24).
> (as for another question I sent yesterday, here's a disclaimer:  I AM using
> a more recent version of Bioperl than is recommended in the installation
> instructions, I'm afraid (
> http://uswest.ensembl.org/info/docs/api/api_cvs.html) - I needed that for
> other projects, and have not put any time into figuring out how to have old
> and new Bioperls installed simultaneously, and how to point Ensembl towards
> the older one.    I don't think that's causing the issue here, though)
>
> thanks again,
>
> Janet Young
>
> -------------------------------------------------------------------
>
> Dr. Janet Young
>
> Malik lab
> http://research.fhcrc.org/malik/en.html
>
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Avenue N., A2-025,
> P.O. Box 19024, Seattle, WA 98109-1024, USA.
>
> tel: (206) 667 4512
> email: jayoung  ...at...  fhcrc.org
>
> -------------------------------------------------------------------
>
>
> _______________________________________________
> 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/20140207/b34f493e/attachment.html>


More information about the Dev mailing list