[BioC] Questions about model matrix, logFC and adjusted P.value of t-test

[Maciej Jończyk]

> Message: 37
> Date: Wed, 27 Oct 2010 00:52:22 +0200
> From: Mingkwan.Nipitwattanaphon at unil.ch
> To: bioconductor at stat.math.ethz.ch
> Subject: [BioC] Questions about model matrix,	logFC and adjusted
> 	P.value of t-test
> Message-ID: <4cc75ba6.64.1ec.786541259 at unil.ch>
> Content-Type: text/plain; charset="iso-8859-1"
>

> My result file contains coefficients of all the 16 contrasts
> that I asked for and
>
> the p-value of each contrast (each t-test) but NOT the
> adjusted p-value. It also
>
> gives me the F value and the p-value of the F-test and again
> NOT the adjusted p-
>
> value of the F-test.
>
> I can get the adjusted p-value from the F-test by using the
> command "p.adjust"
>
> but not with the t-test. When I used the command "topTable"
> with coeff=1 (until
>
> 16 each time for all of my 16 contrasts), I can get the
> adjusted p-value of each
>
> contrast.
>
> My questions are:
>
> 1. Why does not the command "eBayes" give adjusted p-value?
> Is there an easier or
>
> more direct way to get adjusted p-value of the t-test?
>
Hi Mingkwan,

try "decideTests" command, it allows you to choose multiple contrast
correction method and p-value cutoff. Below example, but read about it
in limma's manual.


new_object=decideTests(result_of_eBayes,method="global",adjust.method="BH",p.value=0.05)

Best Regards,

Maciej Jończyk, MSc
Department of Plant Molecular Ecophysiology
Institute of Plant Experimental Biology
Faculty of Biology, University of Warsaw
02-096 Warszawa, Miecznikowa 1



___________________________________
NOCC, http://nocc.sourceforge.net