[BioC] loading processed data into eset class

[Kasper Daniel Hansen]

You are putting your data into an ExpressionSet.  However, rma takes
and AffyBatch as input and outputs an ExpressionSet, so you are using
the wrong class.

Having said that, you can usually find CEL files on ArrayExpress,
which will make your life much easier.

Kasper

On Thu, Aug 5, 2010 at 11:19 AM, Zhe Liu <zhliu.tju at gmail.com> wrote:
> Thanks a lot for your soon reply.
>
> I think this processed data in arrayexpress is not background corrected and
> normalized, though it should be as normalized in MIAME data format. Because
> the CHP:singal amount vary greatly. For example as below, some of them are
> less than 100 but some of them are over thousands.
>
> 51.2
> 78
> 64.6
> 136.3
> 191.2
> 530.7
> 589.4
> 2594.6
> 2853.1
>
> I think they should be normalized. Also I checked the idf file and there is
> not indication about the method about normalization. So is there any method
> applicable to eset?
>
> Thanks.
>
> Zhe
>
> -----Original Message-----
> From: Alex Gutteridge [mailto:alexg at ruggedtextile.com]
> Sent: 05 August 2010 15:59
> To: Zhe Liu
> Cc: bioconductor at stat.math.ethz.ch
> Subject: Re: [BioC] loading processed data into eset class
>
> On Thu, 5 Aug 2010 15:11:34 +0100, "Zhe Liu" <zhliu.tju at gmail.com> wrote:
>> I have a problem with loading matrix data into eset class object. The
>> data.frame is used to load tab-delimited processed data. But when I want
> to
>> run rma() method, there is always a error.
>>
>> Eset=new("ExpressionSet", phenoData=phenoData, featureData=featureData,
>> experimentData=experimentData, exprs=as.matrix(arraydata_pro))
>>
>> rEset=rma(Eset)
>>
>> when I run the final line, it reported:
>>
>> Error in function (classes, fdef, mtable)  :
>>
>> unable to find an inherited method for function "probeNames", for
> signature
>> "ExpressionSet"
>
> rma() works on AffyBatch objects and creates ExpressionSet objects - hence
> the error.
>
> If the data is already processed why do you want to run rma?
>
> --
> Alex Gutteridge
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>