[BioC] Fold Change values after RMA

[Ben Bolstad]

Two points:

1. In general estimates of FC off microarrays tend to be smaller than
the truth, irrespective of processing algorithm. 

2. There is no specific reason why RMA should limit to FC values of 2.0
(and it does not do this in general, as you have observed with your own
dataset). 

In the case of Choudary et al they are studying gene expression changes
in the brain and my understanding is that these fold changes are
typically small, perhaps explaining the comment.

Ben



On Tue, 2006-02-14 at 10:30 +0100, kfbargad at ehu.es wrote:
> Dear List,
> 
> I have come across an article (Choudary et al. 2005, PNAS 102,15653-
> 15658) where they state that the FC values after RMA 
> preprocessing "always" remain below a maximum of 2.0 Is this right?
> 
> I have performed some analyses using RMA, quantile normalisation and 
> limma and am getting M values higher than 2, and if M = log2(FC), then 
> FC values are higher than 4. What am I missing? Any comments on this?
> 
> Thanks in advance,
> 
> David
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor