On 18 Dec 2006, at 21:52, Amir Karger wrote:
Re my previous email, I'd like to download all human exon sequences.
Since there are a couple hundred thousand, I often find Ensembl or
Biomart timing out on me. I tried telling biomart to make a compressed
web file and email me about it, but 6 hours later, it still hasn't done
so.
I'm happy to do this one chromosome at a time, but I'd prefer as
lightweight a solution as possible.
Unfortunately, I can't get the web services thing working right. Am I
right in thinking that I just need to post the right query to the right
URL? I tried to do so, following some instructions from the Biomart PDF
docs. I pulled an XML query off of biomart:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE Query>
<Query virtualSchemaName = "default" Header = "1" count = ""
softwareVersion =
"0.5" >
<Dataset name = "hsapiens_gene_vega" interface = "default" >
<Attribute name = "gene_stable_id" />
<Attribute name = "chrom_name" />
<Attribute name = "transcript_stable_id" />
<Attribute name = "hugo" />
<Filter name = "hugo" value = "ABL1,ABL2,AXL"/>
</Dataset>
</Query>
I put it into y.xml. Then I said:
POST "http://www.biomart.org/biomart/martservice"; < y.xml
(POST is just a link to lwp-request.)
When I try it, I get:
Query ERROR: caught BioMart::Exception: non-BioMart die(): Can't
use an undefined value as an ARRAY reference at
/ebi/www/main/biomart/www/biomart-perl/lib/BioMart/Query.pm line 1635.
Am I totally on the wrong track here? Do I need to use the Biomart Perl
API, even for something this simple?
nope, I just tried it and your query works fine for me:
bigmac: ~/websites/biomart-perl/scripts[arek] perl webExample.pl
../verynewQuery.xml
OTTHUMG00000020813 9 OTTHUMT00000054686 ABL1
OTTHUMG00000020813 9 OTTHUMT00000054684 ABL1
OTTHUMG00000020813 9 OTTHUMT00000054685 ABL1
OTTHUMG00000035199 1 OTTHUMT00000085174 ABL2
OTTHUMG00000035199 1 OTTHUMT00000085175 ABL2
OTTHUMG00000070513 19 OTTHUMT00000150517 AXL
OTTHUMG00000070513 19 OTTHUMT00000150516 AXL
I ran it with the webExample.pl script
(http://cvs.sanger.ac.uk/cgi-bin/viewcvs.cgi/biomart-perl/scripts/
webExample.pl?view=markup)
Can you try it again and see if it works? We have been updating the
server today
and it is possible that you were just unlucky and sent your request at
the wrong time.
If you experience this problem again please drop us a line
a.
Thanks in advance,
- Amir Karger
Research Computing
Life Sciences Division
Harvard University
------------------------------------------------------------------------
-------
Arek Kasprzyk
EMBL-European Bioinformatics Institute.
Wellcome Trust Genome Campus, Hinxton,
Cambridge CB10 1SD, UK.
Tel: +44-(0)1223-494606
Fax: +44-(0)1223-494468
------------------------------------------------------------------------
-------
