

Dear Simon,

Thanks a lot for your response.

----- Forwarded Message -----
From: Simon Anders <anders@embl.de>
To: bioconductor@r-project.org 
Sent: Wednesday, 8 August 2012 6:55 PM
Subject: Re: [BioC] deseq within and between group variation
 
Dear Sudeep

On 2012-08-07 18:31, sudeep s wrote:

> Is there a way to calculate within and between group variability
> between two sample groups (with biological replicates) within DESeq ?
> I went through DESeq manual, but couldn't find any reference.  So is
> the only way is to calculate it directly from the normalized read
> count data ?

Count data is not homoscedastic and hence, there is no point in looking at within and between group variance or using other methods from standard linear regression methods. What you want is within and between group _deviance_, and these are calculated by the 'fitNbinomGLMs' function. Please read up in on generalized linear models (GLMs) and analysis of deviance (ANODEV).

  Simon

_______________________________________________
Bioconductor mailing list
Bioconductor@r-project.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
	[[alternative HTML version deleted]]

