[BioC] understanding of makeContrats() in Limma

[Stephen Henderson]

Hi if I understand you rightly you want to know how to get each contrast
6H-0H and 24H-6H. or A and B as you say.

contrasts.fit will have worked out both. The easiest way to access is
using toptable and either select A or B using coef=1 or coef=2 e.g.

toptable(fit2, coef=1, adjust.method="fdr")
or
toptable(fit2, coef=2, adjust.method="fdr")

Is that what you mean?


Stephen Henderson
Wolfson Inst. for Biomedical Research
Cruciform Bldg., Gower Street
University College London
United Kingdom, WC1E 6BT
+44 (0)207 679 6827


-----Original Message-----
From: bioconductor-bounces at stat.math.ethz.ch
[mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of gregory
voisin
Sent: 13 June 2006 15:14
To: bioconductor
Subject: [BioC] understanding of makeContrats() in Limma

Hi Bioconductorians,
 
  In this time ,as reference documentation, i use a web doc: 23. Limma :
Linear Models for microarray data ( Of course , written by G.Smith) 
In section 23.11 , Time Course experiments, treated analyse the same way
that estrogen data.. but with some difference :
 
Gordon write :" which gene respond at either the 6 h OR 24 h. times in
the wt?"
 
code: 
>cont.wt = makeContrasts("wt.6h-wt.0h", "wt.24h-wt.6r", levels =design) 
#build a contraste matrice.
> fit2 = contrasts.fit (fit , cont.wt)
>fit2 = eBayes (fit2)
  
To Choose Interesting Genes
 
sel.wt = p.adjust(fit2$F.p.values, method = "fdr") < 0.05

PROBLEM OF UNDERSTANDING HERE:
in the question "which gene respond at either the 6 h OR 24 h. times in
the wt", it's set A OR set B
and to choose interesting genes , there is not distinction between A and
B.
 
Hence : May be that the OR ( use here) is inclusive ( Include the AND)
=>  logical OR, by consequently, sel.wt give me the set of DEG
(differential Expressed Genes) at 6 h and 24 h...
 
Am I OK or OutSide.???
 
 
Thamks for your Help..
 
Greg Voisin 
Bioinformatician 
Centre de recherche du CHUM . Montreal
 
 
	[[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives:
http://news.gmane.org/gmane.science.biology.informatics.conductor

**********************************************************************
This email and any files transmitted with it are confidentia...{{dropped}}