[ensembl-dev] problem with VEP and bigwig

[Pierre Lindenbaum]

Hi all,

I cannot make VEP working with a bigwig file.

In the following script, I

* create a VCF with one variation
* create a wig file with one range overlaping the variation
* convert the wig to bigwig
* annotate the variation with VEP:

     echo "fixedStep chrom=22 start=38823000 step=1 span=1000" > test.wig
     echo "99" >> test.wig
     echo "22    48823000">> chrominfo.txt
     /path/to/ucsc/wigToBigWig test.wig chrominfo.txt test.bw
     /path/to/ucsc/bigWigSummary test.bw 22 38823170 38823190 1
     99
     echo "##fileformat=VCFv4.1" > test.vep.vcf
     echo "#CHROM    POS    ID    REF    ALT    QUAL    FILTER INFO" >> 
test.vep.vcf
     echo "22    38823180    .    G    T    100.0    .    ." >> 
test.vep.vcf
     /path/to/variant_effect_predictor/variant_effect_predictor.pl 
--write_cache  --cache --dir cache  \
              --fasta human_g1k_v37.fasta \
              --format vcf --force_overwrite -\
              --custom  test.bw,MYBIGWIG,bigwig,overlap,0 \
              -i  test.vep.vcf -o test.vep.txt

Here is the output:


2013-02-26 16:31:56 - Checking/creating FASTA index
2013-02-26 16:31:56 - Read existing cache info
2013-02-26 16:31:57 - Starting...
2013-02-26 16:31:57 - Read 1 variants into buffer
2013-02-26 16:31:57 - Reading transcript data from cache and/or database
[===============================================]  [ 100% ]
2013-02-26 16:31:57 - Retrieved 271 transcripts (0 mem, 271 cached, 0 
DB, 0 duplicates)
2013-02-26 16:31:57 - Analyzing chromosome 22
2013-02-26 16:31:57 - Caching custom annotations
[===============================================]  [ 100% ]
2013-02-26 16:31:57 - Retrieved 2 custom annotations (2 MYBIGWIG)
2013-02-26 16:31:57 - Analyzing custom annotations
[>                                              ]    [ 0% ]Argument 
"fixedStep chrom=22 start=38823000 step=1 span=1000" isn't numeric in 
numeric ge (>=) at 
/path/to/variant_effect_predictor/Bio/EnsEMBL/Variation/Utils/VEP.pm 
line 1915.
[===============================================]  [ 100% ]
2013-02-26 16:31:57 - Analyzing variants
[===============================================]  [ 100% ]
2013-02-26 16:31:57 - Calculating consequences
2013-02-26 16:31:57 - Processed 1 total variants (1 vars/sec, 1 vars/sec 
total)
2013-02-26 16:31:57 - Finished!


and the file test.vep.txt

## ENSEMBL VARIANT EFFECT PREDICTOR v2.8
## Output produced at 2013-02-26 16:31:57
## Connected to homo_sapiens_core_70_37 on ensembldb.ensembl.org
## Using cache in /commun/data/pubdb/ensembl/vep/cache/homo_sapiens/70
## Using API version 70, DB version 70
## Extra column keys:
## CELL_TYPE : List of cell types and classifications for regulatory feature
## DISTANCE : Shortest distance from variant to transcript
## MYBIGWIG    : test.bw (overlap)
#Uploaded_variation    Location    Allele    Gene    Feature 
Feature_type    Consequence    cDNA_position    CDS_position 
Protein_position    Amino_acids    Codons    Existing_variation Extra
22_38823180_G/T    22:38823180    T    ENSG00000228620 
ENST00000433230    Transcript 
non_coding_exon_variant,nc_transcript_variant    395    -    - -    --
22_38823180_G/T    22:38823180    T    ENSG00000168135 
ENST00000303592    Transcript    missense_variant    1217    958 320    
P/T    Cct/Act    -


while I'm here :-) when I use --custom with a VCF indexed with tabix, 
VEP only shows the range where here found the data (e.g:" --custom 
/path/to/ALL.wgs.phase1_release_v3.20101123.snps_indels_sv.sites.vcf.gz,1KGRel3,vcf,exact,1 
"  )


rs3887390    22:46136619    T    -    -    - intergenic_variant    -    
-    -    -    -    - 1KGRel3=22:46136619-461366


Is it possible to display something else (e.g. a component of the INFO 
field) or should I write a plugin ?


Thank you,

Pierre