[ensembl-dev] VEP and ExAC plugin

Nicolas Thierry-Mieg Nicolas.Thierry-Mieg at imag.fr
Thu Aug 6 10:46:19 BST 2015


Hello,

confirmed fixed, thanks Will!

Regards,
Nicolas


On 08/06/2015 11:33 AM, Will McLaren wrote:
> Hi Nicolas,
>
> Thanks for the report - some of the headers were getting scrambled
> internally.
>
> I've pushed fixes to release/81 of ensembl-tools (the
> variant_effect_predictor.pl <http://variant_effect_predictor.pl> script)
> and the ExAC.pm plugin file - please re-download VEP and re-run the
> installer.
>
> Regards
>
> Will McLaren
> Ensembl Variation
>
> On 5 August 2015 at 17:24, Nicolas Thierry-Mieg
> <Nicolas.Thierry-Mieg at imag.fr <mailto:Nicolas.Thierry-Mieg at imag.fr>> wrote:
>
>     Hi all,
>
>     I'm experimenting with the ExAC plugin for VEP. I've noticed that I
>     only obtain the global ExAC_AF but not the population-specific
>     ExAC_*_AF values when using the --vcf option. Withouth this option
>     it works.
>     Details follow.
>
>
>     For these tests I DL'd the latest VEP v81 available today, used
>     INSTALL.pl, installed homo_sapiens_refseq_vep_81_GRCh37.tar.gz as a
>     cache as well as the ExAC.pm plugin (both via INSTALL.pl). And I
>     DL'd ExAC.r0.3.sites.vep.vcf.gz from the broad ftp.
>
>     $ more input.vcf
>     #CHROM  POS     ID      REF     ALT     QUAL    FILTER  INFO
>     chr1    26131654        .       G       A       67.09   PASS
>
>     $ variant_effect_predictor.pl <http://variant_effect_predictor.pl>
>     -i input.vcf -o ttt.vcf --vcf --offline --refseq  --plugin
>     ExAC,/home/nthierry/Software/ExAC/ExAC.r0.3.sites.vep.vcf.gz  --force
>     2015-08-05 17:34:41 - Read existing cache info
>     2015-08-05 17:34:41 - Loaded plugin: ExAC
>     2015-08-05 17:34:41 - Starting...
>     2015-08-05 17:34:41 - Detected format of input file as vcf
>     2015-08-05 17:34:41 - Read 1 variants into buffer
>     2015-08-05 17:34:41 - Reading transcript data from cache and/or database
>     [===============================================]  [ 100% ]
>     2015-08-05 17:34:41 - Retrieved 92 transcripts (0 mem, 236 cached, 0
>     DB, 0 duplicates)
>     2015-08-05 17:34:41 - Analyzing chromosome 1
>     2015-08-05 17:34:41 - Analyzing variants
>     [===============================================]  [ 100% ]
>     2015-08-05 17:34:41 - Calculating consequences
>     2015-08-05 17:34:41 - Processed 1 total variants (1 vars/sec, 1
>     vars/sec total)
>     2015-08-05 17:34:41 - Wrote stats summary to ttt.vcf_summary.html
>     2015-08-05 17:34:41 - Finished!
>
>
>     So, it seems to run OK, but the produced ttt.vcf ends with
>     |||||0.814||| . I would expect these (empty) fields to contain values.
>
>
>     If I remove --vcf from my command, the output vcf contains
>     population-specific allele freqs, as expected, eg:
>     ExAC_AF_FIN=0.728;IMPACT=MODERATE;ExAC_AF=0.814;ExAC_AF_OTH=0.816;[...]
>
>     As a side note, as you can see without --vcf the fields are called
>     ExAC_AF_* , but in the same file they are documented as ExAC_*_AF, eg:
>     ## ExAC_EAS_AF  : ExAC East Asian Allele frequency
>     This last point is just a minor glitch, but you may want to fix it
>     as well.
>
>
>     Regards,
>     Nicolas




More information about the Dev mailing list