[BioC] avereps problem

[Valerie Obenchain]

Hi Benoit,

When you post a question please include the results from sessionInfo() 
and provide a small sample of code that reproduces the error.

You are using an argument called 'channel' in read.maimages which I 
believe has been replaced by 'green.only'. This makes me think you have 
an outdated version of limma. The current release version of R is 2.12, 
BioC is 2.7 and limma is 3.6.1, please make sure you are using the 
current versions.

The limma documentation states that the normalization and exploratory 
data analysis functions are for two-color and the linear model and 
differential expression functions apply to all micoarrays. The 
normalization step of a one-color array could be your problem. In the 
analysis outlined here, dummy data are entered for the red channel.
http://matticklab.com/index.php?title=Single_channel_analysis_of_Agilent_microarray_data_with_Limma

This approach may be more robust and allow you to use more functions in 
the limma package. If this example does not clarify things, please 
provide some code that reproduces the problem.  If you know of any 
publicly available Agilent one-color data files for testing that would 
be helpful too.

Valerie



On 10/25/10 02:02, Benoit Loup wrote:
> Hi all,
>
> I have some troubles using the avereps function on Limma.
>
> I process and analyse agilent one-color data without any problems but
> when I average replicate probes with avereps, the function create a good
> matrix but the new object is only a matrix containing the new
> intensities. Thus, I lose the Elist format with all associated
> informations (weights, probe name, gene name, etc...).
> So my question is how to obtain a complete Elist (or RGlist or MAlist)
> object using this function.
>
> Here is my script:
>
> /library(limma)
> targets=readTargets("target.txt", sep="\t")
> Eraw=read.maimages(files=targets$FileName,source="agilent",names=targets$SampleName,channels=1)
>
> E=new("EList")
> E$E=Eraw$E
> E$Eb=Eraw$Eb
> E$weights=Eraw$weights
> E$targets=Eraw$targets
> E$genes=Eraw$genes
> E$source=Eraw$source
> E$printer=Eraw$printer
>
> Enorm=E
> Enorm$E=log2(normalizeBetweenArrays(Enorm$E,method="quantile"))
>
> EnormAvRep=avereps(Enorm,ID=Enorm$genes$ProbeName)
> /
> It's here that the new object is a matrix containing only intensities
> values.
>
> Thanks for your help.
>
>
>    
>
>
> _______________________________________________
> 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