[BioC] best package for cDNA single channel microarrays

[Wolfgang Huber]

Dear Artur,

you could use

- vsn for background correction and normalization (it just takes a 
matrix of features x arrays)
- genefilter or limma for the subsequent differential expression 
analysis, which should then not be qualitatively different from that of 
e.g. Affymetrix arrays


Watch out for spot and batch effects though, if these are "home-made" 
spotted arrays.


Best wishes
   Wolfgang

------------------------------------------------------------------
Wolfgang Huber  EBI/EMBL  Cambridge UK  http://www.ebi.ac.uk/huber


Naomi Altman wrote:
> For some things you might want to do, putting a matrix of 1's in R 
> will solve the problem.  But this will not work for 
> normalization.  If the array manufacturer does not provide a 
> normalization routine, I usually use quantile normalization which is 
> available in limma.  You probably want to look at the background 
> correction methods, to see what might be right for your array.
> 
> --Naomi
> 
> At 10:33 AM 3/2/2007, Artur Veloso wrote:
>> Hi All,
>>
>> I am working with cDNA microarrays that were run only on one channel and I
>> have been trying to analyze them using the limma package. By changing the
>> read.maimages fucntion I was able to upload my data, but most functions
>> won't work with that RGList created due to the absence of a red component.
>>
>> Is there a package that is more fit for such an analyses? Am I overlooking
>> an important function? Any ideas on how should I get past this?
>>
>> Thanks very much and sorry for such a basic question.
>> Artur
>