[BioC] limma decideTests

Gordon K Smyth smyth at wehi.EDU.AU
Thu Jun 1 14:33:11 CEST 2006

> Date: Wed, 31 May 2006 20:47:50 +0200
> From: Georg Otto <georg.otto at tuebingen.mpg.de>
> Subject: [BioC] limma decideTests
> To: bioconductor at stat.math.ethz.ch
> Hi,
> there seems to be a problem with decideTests (limma package):
> results<-decideTests(fit2, adjust.method="BH")
> gives a table with probe IDs as row names if the methods "separate" or
> "global" are used. With methods "heirarchical" and "nestedF" however,
> row names are missing. Is this a feature the purpose of which I do not
> understand?
> This is limma 2.6.2 and R 2.3.0
> Best,
> Georg

Whether or not decideTests() adds row names isn't actually specified in the documentation so,
strictly speaking, both behaviours are correct.  But it would be better to be consistent.  So I'll
update "heirarchical" and "nestedF" to preserve row names when this is possible.

Best wishes

More information about the Bioconductor mailing list