[BioC] RMA-bimodality:

Claus Mayer claus at bioss.ac.uk
Wed Jun 7 12:08:48 CEST 2006

Hi Philip,

the bimodality we discuss here is ACROSS genes. Normality/Unimodality 
that is assumed for parametric tests concerns the error distribution 
WITHIN the replicates for the same gene, i.e. the bi-modality of GCRMA 
normalized data across gene does not have any direct implications on 
whether a parametric or non-parametric test is preferable.



Phguardiol at aol.com wrote:
> Hi,
> these are comments/questions from a neophyte who asked questions about  this, 
> one or 2 years ago on this list, and did not get clear answer !
> is this bimodal distribution observed with GCRMA against the use of  
> parametric tests, does it favor the use of non parametric tests ? Said  differently, 
> should this distribution affect our choice regarding  statistical tests for 
> subsequent analyses ?

 Dr Claus-D. Mayer                    | http://www.bioss.ac.uk
 Biomathematics & Statistics Scotland | email: claus at bioss.ac.uk
 Rowett Research Institute            | Telephone: +44 (0) 1224 716652
 Aberdeen AB21 9SB, Scotland, UK.     | Fax: +44 (0) 1224 715349

More information about the Bioconductor mailing list