[BioC] genefilter findLargest and Illumina ids

[Pan Du]

Hi Lavinia

Illumina Array_Address_Ids is not the regular IDs used in public. You need
to first convert Array_Address_Ids as Illumina ID or directly convert as
Entrez Gene IDs, and then do functional analysis.

You can use lumiMouseIDMapping for ID mapping. Here is some code:
 # convert Array_Address_Ids to nuID first
 nuIDs = IlluminaID2nuID(addressIDs, lib='lumiMouseIDMapping')
 # then convert back to regular IlluminaID
 IlluminaIDs = nuID2IlluminaID(nuIDs, lib='lumiMouseIDMapping')
 # or map to Entrez ID based on Illumina manifest file (this might be old)
 entrezIDs = nuID2EntrezID(nuIDs, lib='lumiMouseIDMapping')



Pan

On 4/14/10 5:00 AM, "bioconductor-request at stat.math.ethz.ch"
<bioconductor-request at stat.math.ethz.ch> wrote:

> Message: 21
> Date: Wed, 14 Apr 2010 15:25:44 +1000
> From: Lavinia Gordon <lavinia.gordon at mcri.edu.au>
> To: bioconductor at stat.math.ethz.ch
> Subject: [BioC] genefilter findLargest and Illumina ids
> Message-ID: <4BC551D8.60909 at mcri.edu.au>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> 
> Dear All
> 
> I am trying to replicate the example from the Category vignette, but
> with Illumina data
> (e.g. from 
> http://bioconductor.org/packages/2.5/bioc/vignettes/Category/inst/doc/Category
> .R)
> 
> My data is MouseWG-6_V1
> where the featureNames(mydata) are Array_Address_Ids (with the data read
> in as a LumiBatch/ExpressionSet object)
> and it is BeadStudio output.
> 
> Using the annotation package illuminaMousev1BeadID.db gives:
>> fL = findLargest(featureNames(mydata), abs(ttests$statistic),
> "illuminaMousev1BeadID")
> Loading required package: org.Mm.eg.db
> Error in .checkKeys(value, Lkeys(x), x at ifnotfound) :
>    value for "580022" not found
> 
> How can I confirm which annotation package to use, as I have BeadStudio
> data but with Array_Address_Ids?
> I assume it would be possible to substitute featureNames for
> featureData, where
>> head(featureData(mydata)[[2]])
> [1] "0610005I04"    "0610006I08RIK" "0610007C21RIK" "0610007C21RIK"
> "0610007J10RIK" "0610007L01RIK"
> but I haven't had any success in matching these to available keys.
> 
> Any advice greatly appreciated.
> 
> with regards
> 
> Lavinia Gordon.
> 
> -- 
> Senior Bioinformatics Officer
> Murdoch Childrens Research Institute
> Royal Children's Hospital
> Flemington Road
> Parkville
> Victoria 3052
> Australia
> www.mcri.edu.au