[ensembl-dev] Variant Effect Predictor not using HTS to read gzipped fata

[Hollizeck, Sebastian]

Hi,

I installed the standalone script in my module system for usage on our server, but i cant seem to get it working with the cache.
The script works with just using the --database option, but when i use the --cache i get the following error 

MSG: ERROR: Cannot index bgzipped FASTA file with Bio::DB::Fasta

this is the full output with verbose

[hollizeck at kkf0f6ee kleinPipeline]$ variant_effect_predictor.pl  --check_existing --gmaf --maf_1kg --maf_esp --maf_exac  --pubmed --regulatory --species homo_sapiens --port 3337 --buffer_size 40000 --cache --minimal  --sift b --polyphen b --ccds --symbol --biotype --gene_phenotype --variant_class  --fork 6 --force_overwrite --filter_common --dir_cache /opt/modules/i12g/ensembl-tools/85/cachedir -i /archive/sample_data/scn281/scn281pa_fa.recalibrated_variants.vcf.gz --vcf -o /archive/sample_data/scn281/scn281pa_fa.recalibrated_variants_vep.vcf.gz -v#----------------------------------#
# ENSEMBL VARIANT EFFECT PREDICTOR #
#----------------------------------#

version 85

By Will McLaren (wm2 at ebi.ac.uk)

Configuration options:

biotype            1
buffer_size        40000
cache              1
ccds               1
check_existing     1
core_type          core
dir                /opt/modules/i12g/ensembl-tools/85/cachedir
dir_cache          /opt/modules/i12g/ensembl-tools/85/cachedir
dir_plugins        /home/hollizeck/.vep/Plugins
filter_common      1
force_overwrite    1
fork               6
gene_phenotype     1
gmaf               1
host               ensembldb.ensembl.org
input_file         /archive/sample_data/scn281/scn281pa_fa.recalibrated_variants.vcf.gz
maf_1kg            1
maf_esp            1
maf_exac           1
minimal            1
numbers            1
output_file        /archive/sample_data/scn281/scn281pa_fa.recalibrated_variants_vep.vcf.gz
polyphen           b
port               3337
pubmed             1
regulatory         1
sift               b
species            homo_sapiens
stats              HASH(0x58ccbc0)
symbol             1
variant_class      1
vcf                1
verbose            1

--------------------

Will only load v85 databases
Species 'homo_sapiens' loaded from database 'homo_sapiens_core_85_37'
Species 'homo_sapiens' loaded from database 'homo_sapiens_cdna_85_37'
Species 'homo_sapiens' loaded from database 'homo_sapiens_vega_85_37'
Species 'homo_sapiens' loaded from database 'homo_sapiens_otherfeatures_85_37'
Species 'homo_sapiens' loaded from database 'homo_sapiens_rnaseq_85_37'
homo_sapiens_variation_85_37 loaded
homo_sapiens_funcgen_85_37 loaded
No ancestral database found
No ontology database found
No taxonomy database found
No ensembl_metadata database found
No production database or adaptor found
2016-12-07 14:04:31 - Connected to core version 85 database and variation version 85 database
2016-12-07 14:04:31 - Read existing cache info
2016-12-07 14:04:32 - Auto-detected FASTA file in cache directory

-------------------- EXCEPTION --------------------
MSG: ERROR: Cannot index bgzipped FASTA file with Bio::DB::Fasta

STACK Bio::EnsEMBL::Variation::Utils::FastaSequence::setup_fasta /opt/modules/i12g/ensembl-tools/85/modules/Bio/EnsEMBL/Variation/Utils/FastaSequence.pm:194
STACK main::configure /opt/modules/i12g/ensembl-tools/85/bin/variant_effect_predictor.pl:835
STACK toplevel /opt/modules/i12g/ensembl-tools/85/bin/variant_effect_predictor.pl:146
Date (localtime)    = Wed Dec  7 14:04:32 2016
Ensembl API version = 85
---------------------------------------------------

This looks like the program does not select the Bio:DB:HTS module but it is present and also in the perl path

[hollizeck at kkf0f6ee kleinPipeline]$ ll /opt/modules/i12g/ensembl-tools/85/modules/Bio/DB/HTS.pm
-rw-rw-rw- 1 root root 77700 Sep 21 18:44 /opt/modules/i12g/ensembl-tools/85/modules/Bio/DB/HTS.pm
[hollizeck at kkf0f6ee kleinPipeline]$ echo $PERL5LIB
/opt/modules/i12g/perl/5.24.0/perl5lib/lib/perl5:/opt/modules/i12g/ensembl-tools/85/modules

I do not know, if I even look at the right spot.
Can you please help me to debug this further?

Is there any way to specify which plugin to use?


Sebastian Hollizeck
Bioinformatics

Dr. von Hauner Children's Hospital
Kubus Research Center
Lindwurmstraße 2a
D-80337 München
phone +49 89-4400-57488
fax +49 89-4400-57979