[ensembl-dev] Segmentation fault when running VEP

[Will McLaren]

Hi Emma,

It looks like the fault is being triggered when the FASTA DB is being
prepared.

Are you running the latest (82) VEP? It's possible that our new FASTA
indexer isn't playing nicely with your system. You could try the following:

cd
/mnt/hds/proj/bioinfo/develop/modules/ensembl-tools-release-82/scripts/variant_effect_predictor/
rm -rf FAIDX
rm Faidx.*

Then re-running your command. This will reset your system to using the old
BioPerl FASTA indexer.

Regards

Will McLaren
Ensembl Variation

On 24 November 2015 at 09:32, Emma Sernstad <emma.sernstad at scilifelab.se>
wrote:

> Hi,
>
> we are experiencing problems that are difficult to debug.
> Namely we get this output almost right away:
>
> perl
> /mnt/hds/proj/bioinfo/develop/modules/ensembl-tools-release-82/scripts/variant_effect_predictor/
> variant_effect_predictor.pl --dir_cache
> /mnt/hds/proj/cust003/develop/modules/ensembl-tools-release-82/cache
> --cache --force_overwrite --vcf --fork 4 --buffer_size 20000 --offline
> --chr 21 --hgvs --symbol --numbers --sift p --polyphen p --humdiv --domains
> --protein --ccds --uniprot --biotype --regulatory --tsl -i
> /mnt/hds/proj/bioinfo/develop/cust000/337334/analysis/exomes/337334/bwa/gatk/337334_sorted_smd_rreal_brecal_gvcf_vrecal_vt_BOTH_21.vcf
> -o
> /mnt/hds/proj/bioinfo/develop/cust000/337334/analysis/exomes/337334/bwa/gatk/337334_sorted_smd_rreal_brecal_gvcf_vrecal_vt_vep_BOTH_21.vcf
>
> 2015-11-24 10:10:10 - Read existing cache info
>
> 2015-11-24 10:10:10 - Auto-detected FASTA file in cache directory
>
> Segmentation fault
>
> Do you think it might be a corrupt cache directory? Can it have anything
> to do with the perl installation (v5.18.2)?
>
> We are running on CentOS 6.6.
>
> --
> Emma Sernstad
> Science for Life Laboratory, KISP
> Tomtebodavägen 23A
> SE-171 65 Solna, Sweden
>
> Email: emma.sernstad at scilifelab.se
> Phone: +46 (0)722503877(Mobile)
>
> _______________________________________________
> 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/20151124/5d193531/attachment.html>