[BioC] RNA-Seq with multiple factors [was Informations about DESeq package]

[Gordon K Smyth]

Dear Roman,

I know that your question was specifically about the DESeq package, but 
you have previously asked about the edgeR, bayseq and DESeq packages on 
SEQanswers:

  http://seqanswers.com/forums/showthread.php?t=4349

so I thought you might appreciate a more complete answer.

I think that edgeR is the only package on Bioconductor that analyses 
RNA-Seq experiments with multiple explanatory factors.

In your SEQanswers post, you asked about a paired comparison problem 
(young males and females, old males and females).  The edgeR Vignette 
gives in Section 11 a worked example of an analysis with paired 
comparisons (tumour vs normal for the same patients) that is analogous to 
your problem.  edgeR does not require you to have replicates of any of the 
four conditions.  You can find tags that are differentially expressed 
between Males and Females, adjusting for any age differences, by following 
the analysis in the vignette.  You just need to substitute Age for Patient 
and Sex for Group.

edgeR also has some special features to accommodate comparisons without 
replicates, for example comparing two unreplicated libraries, but I don't 
think you need this here.  For the record, this is done with 
estimateGLMCommonDisp(method="deviance",robust=TRUE).

Try out the paired comparison analysis and let us know if you have any 
problems with it.

Best wishes
Gordon

---------------------------------------------
Professor Gordon K Smyth,
NHMRC Senior Research Fellow,
Bioinformatics Division, 
Walter and Eliza Hall Institute of Medical Research, 
1G Royal Parade, Parkville, Vic 3052, Australia.
smyth at wehi.edu.au
http://www.wehi.edu.au
http://www.statsci.org/smyth



> Date: Thu, 21 Apr 2011 10:04:07 +0200
> From: Bruno Roman <roman.bruno at hotmail.fr>
> To: simon.anders at embl.de, bioconductor at r-project.org
> Subject: [BioC] Informations about DESeq package
>
> Dear Dr. Anders,
>
> First I'd like to thank you for the great work you've done with DESeq
> package.
>
> Then I was wondering if you could give me some suggestions about the
> multiple RNAseq libraries comparison (i.e more than 2 conditions with or
> without replicates) and if this feature is supported by DESeq.
>
> Best regards
> Roman

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