[BioC] limma design matrix question

[Gordon Smyth]

At 06:05 AM 16/11/2003, Marcelo Luiz de Laia wrote:
>Hi All,
>
>I have this targets:
>
> > ka.rest.targets
>   SlideNumber          FileNames      Cy3     Cy5
>1           1 Ka_Rest_Rep_1.datr Control Treated
>2           2 Ka_Rest_Rep_2.datr Control Treated
>3           3 Ka_Rest_Rep_3.datr Control Treated
>
>and I proceed
>
> > design <- designMatrix(ka.rest.targets,ref="Control")
> > design
>   Treated
>1       1
>2       1
>3       1
> > ?designMatrix
>
>Is this correct??

Yes.

>I am confused because I read this in help:
>
>"There is a need for design matrices for direct two-color design
>  and for single-channel experiments but, this version is only for
>  reference designs."
>"Examples:
>   targets <-
>data.frame(Cy3=c("Ref","Control","Ref","Treatment"),Cy5=c("Control","Ref","T
>reatment","Ref"))
>   designMatrix(targets, "Ref")"
>
>In my case, "REF" is my control (Control)?
>
>I don't understand the difference between the Reference and Control!

In the function designMatrix(), the "reference" is any RNA source which 
appears on all the arrays and which therefore acts as a "common reference". 
Your "Control" sample obviously fits this criterion.

You have a simple replicated experiment. This can be viewed both as a 
direct design and as a reference design. Anything works fine for this 
simple design.

Gordon

>Thanks
>
>Marcelo Luiz de Laia, M.Sc.
>Dep. de Tecnologia, Lab. Bioquímica e de Biologia Molecular
>Universidade Estadual Paulista - UNESP
>Via de Acesso Prof. Paulo Donato Castelane, Km 05
>14.884-900 - Jaboticabal, SP, Brazil
>PhoneFax: 16 3209-2675/2676/2677 R. 202/208/203 (trab.)
>Phone res: 16 3203 2328 - www.lbm.fcav.unesp.br - mlaia at yahoo.com
>---