Hi,
Well, first off you need to figure out how to normalize your starting samples--cell number would be reasonable. 
Apart from that,  it's probably unwise to use any global normalization methods, so you'll need to use Affy's external controls.  I don't think any Bioconductor package currently does this, but it's straightforward with GCOS.
Elenita Kanin 
PhD Student 
UW-Madison Department of Biochemistry 
eikanin@wisc.edu 


----- Original Message ----- 
From: Pedro Botías <pbotias@bio.ucm.es> 
Date: Monday, April 3, 2006 3:54 am 
Subject: [BioC] Global extreme changes 
To: bioconductor@stat.math.ethz.ch 

> Dear all, 
> We are studying the changes on yeast gene expression under 
> conditions that lead to extreme changes in the transcriptome (we 
> think to global repression) and finally to cell death. We are 
> using Affymetrix technology and finding a great discrepancy 
> between MAS and GCRMA analysis, much more than in other studies 
> with less transcriptional changes. 
> Does anybody work with data in a situation of global severe 
> repression? From these two analysis, what could fit better to our 
> experimental conditions? 
> Thanks for your help. 
> 
> Pedro Botías Talamantes 
> Unidad de Genómica 
> Parque Científico de Madrid-UCM 
> Facultad CC Biológicas- Planta sótano 
> Universidad Complutense de Madrid 
> 28040 Madrid 
> 
> Tel: 91 3945036 
> FAX 91 3944883 
> e-mail: pbotias@bio.ucm.es 
> Web: http://www.ucm.es/info/gyp/genomica/ 
> [[alternative HTML version deleted]] 
> 
>

	[[alternative HTML version deleted]]