Hacking is believing in English: June 2007

Saturday, June 23, 2007

First, I make ActiveRecord model script (ensembl.rb) for Agilentprobe
table in EnsemblMart.

#!/usr/bin/env ruby

require 'rubygems'
require_gem 'activerecord'

ActiveRecord::Base.pluralize_table_names = false

ActiveRecord::Base.establish_connection(
:adapter  => 'mysql',
:database => 'ensembl_mart_40',
:host     => 'ensembldb.ensembl.org',
:username => 'anonymous',
:password => ''                                   
)

class Agilentprobe <> 

Next, I make ID translation program (affyprobe2moe4302.rb) using above script.

#!/usr/bin/env ruby

require 'ensembl'

agilentid = 'A_52_P654624'
agilentprobe = Agilentprobe.find_by_dbprimary_id('A_52_P654624')
moe4302 = Moe4302.find_all_by_gene_id_key(agilentprobe.gene_id_key)
moe4302.each do |entry|
puts [aglentid,
        entry.gene_stable_id, entry.transcript_stable_id, entry.translation_stable_id,
        entry.dbprimary_id].join("\t")
end

Let's run it!

$ ruby affyprobe2moe4302.rb

A_52_P654624    ENSMUSG00000058140      ENSMUST00000081249      ENSMUSP00000080006      1429452_x_at
A_52_P654624    ENSMUSG00000058140      ENSMUST00000081249      ENSMUSP00000080006      1435353_a_at
A_52_P654624    ENSMUSG00000058140      ENSMUST00000071829      ENSMUSP00000071732      1429452_x_at

Modeling Ensembl Mart using ActiveRecord

First, I make a Ruby script for modeling ensembl mart.

#!/usr/bin/env ruby

require 'rubygems'
require_gem 'activerecord'

ActiveRecord::Base.pluralize_table_names = false

ActiveRecord::Base.establish_connection(
:adapter  => 'mysql',
:database => 'ensembl_mart_40',
:host     => 'ensembldb.ensembl.org',
:username => 'anonymous',
:password => ''                                    
)

class Agilentprobe < ActiveRecord::Base
set_table_name "mmusculus_gene_ensembl__xref_agilentprobe__dm"
set_primary_key "gene_id_key"
end

Second, I make a searching program for Agilentprobe talbe in
Ensembl Mart using above modeling code.

#!/usr/bin/env ruby

require 'ensembl'

agilentprobe = Agilentprobe.find_by_dbprimary_id('A_52_P654624')

puts agilentprobe.dbprimary_id
puts agilentprobe.gene_id_key
puts agilentprobe.gene_stable_id
puts agilentprobe.transcript_id_key
puts agilentprobe.transcript_stable_id
puts agilentprobe.translation_id

Let's run it!

$ ruby run_ensembl.rb
A_52_P654624
209556
ENSMUSG00000058140
263491
ENSMUST00000081249
251942
ENSMUSP00000080006

Free clustring software for gene expression data

GenePattern (Maybe freesoft)

One-liner for mapping Agilent Microarray ID to Affymetrix Genechip ID

echo "
SELECT
mmusculus_gene_ensembl__xref_agilentprobe__dm.dbprimary_id,
mmusculus_gene_ensembl__xref_agilentprobe__dm.gene_stable_id,
mmusculus_gene_ensembl__xref_affy_mouse430_2__dm.dbprimary_id
FROM
mmusculus_gene_ensembl__xref_agilentprobe__dm,
mmusculus_gene_ensembl__xref_affy_mouse430_2__dm
WHERE
mmusculus_gene_ensembl__xref_agilentprobe__dm.dbprimary_id = 'A_52_P654624' AND 
mmusculus_gene_ensembl__xref_agilentprobe__dm.gene_id_key = /
mmusculus_gene_ensembl__xref_affy_mouse430_2__dm.gene_id_key;" /
| mysql -h ensembldb.ensembl.org -A -uanonymous -D ensembl_mart_40

You need mysqlclient.

Sunday, June 03, 2007

One-liner for mapping Affymetrix Probe ID to NCBI Entrez Gene ID :-)

You can map Affymetrix IDs to NCBI Entrez Genes by one-liner command.
echo 'library(mouse4302)\nwrite.table(unlist
(as.list(mouse4302GENENAME)),
file="affyid2genename.txt", sep="\\t", eol="\\n",
quote=F, col.names=F)' | R --no-save
The result was written to "affyid2genename.txt" file:
less affyid2genename.txt
1459626_at      NA
1457754_at      RIKEN cDNA 4930430F08 gene
1452692_a_at    NADH dehydrogenase (ubiquinone) flavoprotein 2
:
:
You need R, Bioconductor and mouse4302 package.

Making a gene expression database within 11 min using Ruby on Rails

Ruby on Rails is very powerful web application framework. This site shows that bioinformatics scientist can make a simple viewer for gene expression data within 11 min using Ruby on Rails.

See following the movie:
http://itoshi.tv/data/geneexp/RoR.html

Saturday, June 02, 2007

We need big memory for exon array analysis

In addition, the oligo package will require you to make a PDEnv
with the HuEx-l_OST-V2.cdf file, which I am unable to do on
a 64 bit Linux box with 8 Gb RAM (I think 16 Gb is sufficient).
via Re: Human exon array chip

Exonmap (Bioconductor package for exon array analysis)
requires big memory.

via Affy EXON Array: iterPLIER or RMA

It is based on the affy exon probeset mappings. It can run on
a machine with 4Gb of memory, but can easily use 8-12Gb depending
on what your trying to do. There are other probeset environments
available for this platform at Bioconductor (refseq, ensembl, etc.)
which are much more memory friendly.

via Re: Affymetrix exon arrays?

The oligo/makePlatformDesign packages can be used for the
exon arrays. However, you should know that these arrays are
incredibly huge, so unless you have a 64 bit OS and around
16 - 32 Gb RAM you won't be able to do anything with them.
via Re: Bioconductor package for Affymetrix Exon arrays.