[BioC] limma design

[Aurelie Laugraud]

Hello, 
I would like to create a design matrix for an experiment but I am a bit
confused and I would appreciate to have some expert views.
We are talking about dye-swaped microarrays (chicken genome,  but it
doesn't matter). 
We have two conditions : Vasa can be + or - 
and SF absent or present
first set of arrays(1 and 2): Vasa+/vasa- and the dye-swap (no SF at
all)
second(3 and 4): vasa-/vasa- SF  and the dye-swap
third(5 and 6): vasa-/vasa+ SF  and the dye-swap
fourth(7 and 8): vasa- SF/vasa+ SF  and the dye-swap

can I integrate this into one model matrix ? 
I have done the analysis separating the arrays because the only
comparaisons that interested the biologists were 1/2, 3/4, 5/6 and 7/8
but I don't think this is very nice and would like to make it better.

Thank you

Aurélie 





-- 
Aurélie Laugraud
PRABI
Bât.Mendel 2è étage
16 rue Raphael Dubois
69622 Villeurbanne Cedex
tel: 0472432826