[BioC] backgroundCorrect offset value

[Gordon K Smyth]

Dear Prasad,

Just to add to James MacDonald's replies.  You can read more about offsets 
in Ritchie et al (Bioinformatics 2007) and Shi et al (NAR 2010).  Ritchie 
et al (2007) recommend offset=50 as a sensible choice for GenePix on a 
range of two color microarray platforms.

The purpose of the offset is to stabilize the variance as a function of 
the mean.  You can see the effect by typing

   plotSA(fit)

for a 'fit' object created by lmFit().  Generally speaking, you don't want 
to see excess variability at the low intensity range.  Obviously this also 
depends on how many probes you filter, and filtering probes not expressed 
in any condition is generally recommended.

Another way to see the success of variance stabilization is through the 
prior degrees of freedom fit$df.prior after running eBayes().  Generally 
speaking, better variance stabilization gives higher values for df.prior.

Best wishes
Gordon

> Date: Fri, 26 Aug 2011 15:00:20 +0000
> From: Prasad Siddavatam <siddavatam at gmail.com>
> To: <bioconductor at stat.math.ethz.ch>
> Subject: [BioC] backgroundCorrect offset value
>
>
>
> Hello Users,
>
> I have a question regarding the usage of backgroundCorrect function in LIMMA.
>
> when I do the following with offset 50, I am getting 2900 differentially
> expressed genes
> RG.b <- backgroundCorrect(RG, method = "normexp", offset = 50);
>
> where as, when I do the following with offset 1,
> I am getting 1300 differentially expressed genes
> RG.b <- backgroundCorrect(RG, method = "normexp", offset = 1);
>
> Please advise which offset value to be used? Why is offset value making
> so much difference?
>
> I am using this for TWO channel data, which is read by "genepix".
>
> Greatly appreciate your help.
>
> Prasad
>

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}