[ensembl-dev] problem with VEP v80 - regression

[Nicolas Thierry-Mieg]

Dear developers,

we use VEP with custom annotations (in a targeted sequencing project).
The following test was working fine with VEP v76, but now fails with the 
latest v80.

The test is performed in a clean subdir.
The computer is running Centos6 x86_64 fully up-to-date.

The test files are as follows:

[nthierry at timc-bcm-07 Test_VEP_80]$ cat transcript_reorder.gff
3	protein_coding	exon	113077591	113077946	.	-	.	gene_id toto; 
transcript_id toto_a; exon_number 1;
3	protein_coding	CDS	113077591	113077746	.	-	0	gene_id toto; 
transcript_id toto_a; product_id toto_a; exon_number 1;
3	protein_coding	start_codon	113077744	113077746	.	-	.	gene_id toto; 
transcript_id toto_a; product_id toto_a;
3	protein_coding	exon	113063155	113063561	.	-	.	gene_id toto; 
transcript_id toto_a; exon_number 2;
3	protein_coding	stop_codon	113063352	113063354	.	-	.	gene_id toto; 
transcript_id toto_a; product_id toto_a;
3	protein_coding	CDS	113063355	113063561	.	-	0	gene_id toto; 
transcript_id toto_a; product_id toto_a; exon_number 2;
[nthierry at timc-bcm-07 Test_VEP_80]$

[nthierry at timc-bcm-07 Test_VEP_80]$ cat snp.vcf
#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO
3	113063450	.	G	A	.	PASS
[nthierry at timc-bcm-07 Test_VEP_80]$


[nthierry at timc-bcm-07 Test_VEP_80]$ perl gtf2vep.pl -i 
transcript_reorder.gff -f hs.chrom_3.fasta --dir cache/ -d 1 -s homo_sapiens
2015-06-18 12:26:23 - Checking/creating FASTA index
2015-06-18 12:26:23 - Processing chromosome 3
2015-06-18 12:26:23 - All done!

[This seems to work, cache/ now contains a single file: 
cache/homo_sapiens/1/3/113000001-114000000.gz ]


[nthierry at timc-bcm-07 Test_VEP_80]$ perl variant_effect_predictor.pl 
--offline -i snp.vcf --vcf --dir_cache cache/ --cache_version 1 --force 
--no_stats
2015-06-18 12:26:29 - Starting...

-------------------- EXCEPTION --------------------
MSG: Can not find internal name for species 'homo_sapiens'
STACK Bio::EnsEMBL::Registry::get_adaptor 
/home/nthierry/Test_VEP_80/Bio/EnsEMBL/Registry.pm:985
STACK Bio::EnsEMBL::Variation::Utils::VEP::get_version_data 
/home/nthierry/Test_VEP_80/Bio/EnsEMBL/Variation/Utils/VEP.pm:6120
STACK main::get_out_file_handle variant_effect_predictor.pl:1862
STACK main::main variant_effect_predictor.pl:227
STACK toplevel variant_effect_predictor.pl:145
Date (localtime)    = Thu Jun 18 12:26:29 2015
Ensembl API version = 80
---------------------------------------------------



The same test when using VEP v76 worked well:

[nthierry at timc-bcm-07 Test_VEP_76]$ perl gtf2vep.pl -i 
transcript_reorder.gff -f hs.chrom_3.fasta --dir cache/ -d 1 -s homo_sapiens
2015-06-18 12:45:50 - Checking/creating FASTA index
2015-06-18 12:45:50 - Processing chromosome 3
2015-06-18 12:45:50 - All done!
[nthierry at timc-bcm-07 Test_VEP_76]$ perl variant_effect_predictor.pl 
--offline -i snp.vcf --vcf --dir_cache cache/ --cache_version 1 --force 
--no_stats
2015-06-18 12:45:55 - Starting...
2015-06-18 12:45:55 - Detected format of input file as vcf
2015-06-18 12:45:55 - Read 1 variants into buffer
2015-06-18 12:45:55 - Reading transcript data from cache and/or database
[===============================================]  [ 100% ]
2015-06-18 12:45:55 - Retrieved 1 transcripts (0 mem, 1 cached, 0 DB, 0 
duplicates)
2015-06-18 12:45:55 - Analyzing chromosome 3
2015-06-18 12:45:55 - Analyzing variants
[===============================================]  [ 100% ]
2015-06-18 12:45:55 - Calculating consequences
2015-06-18 12:45:55 - Processed 1 total variants (1 vars/sec, 1 vars/sec 
total)
2015-06-18 12:45:55 - Finished!
[nthierry at timc-bcm-07 Test_VEP_76]$


Any ideas?

Regards,
Nicolas


-- 
------------------------------------------------------------
Nicolas Thierry-Mieg
Laboratoire TIMC-IMAG/BCM, CNRS UMR 5525
Pavillon Taillefer, Faculte de Medecine
38706 La Tronche cedex, France
tel: (+33)456.520.067, fax: (+33)456.520.055
------------------------------------------------------------