[BioC] RE: Design/contrast matrix for limma

[Chris Wilkinson]

Hi Anthony,

> How do I set up a design and contrast matrix for the following experiment.
>
> I have unstimulated cells and stimulated cells for individuals with
> and without disease.
>
> sample groups
>
> 1	Disease		Unstimulated
> 2	Disease		Stimulated
> 3	No Disease	Unstimulated
> 4	No Disease	Stimulated
>
> Although I can set up design matrix for all two way comparisons, I am
> really interested in what genes are different between disease and no
> disease after stimulation of the cells.

I think what you have is a 2x2 factorial experiment. Check out the weaver
mutant data example in the limma example. In your case the first factor is
Disease (no disease & disease) and the second factor is Stimulation(unstim,
stim). What is of most interest is the interaction of disease and
stimulation (DxS).
            Unstimulated   Stimulated
No Disease  00 ----------- 0S
            |              |
            |              |
   Disease  D0------------ DS

The columns in the design matrix are Disease, Stim and (DiseaseXStim).
I would also suggest you look over Glonek and Solomon, Factorial and time
course designs for cDNA microarray experiments. Biostatistics 2004
Jan;5(1):89-111. (A copy is also available at
http://mag.maths.adelaide.edu.au/documents.html) It has a very nice
discussion on factorial designs, in particular 2x2 designs which I think you
would find helpful.

Cheers
Chris

Dr Chris Wilkinson

Research Officer (Bioinformatics)        | Visiting Research Fellow
Child Health Research Institute (CHRI)   | Microarray Analysis Group
7th floor, Clarence Rieger Building      | Room 121
Women's and Children's Hospital          | School of Applied Mathematics
72 King William Rd, North Adelaide, 5006 | The University of Adelaide, 5005

Math's Office (Room 121)        Ph: 8303 3714
CHRI   Office (CR2 52A)         Ph: 8161 6363

Christopher.Wilkinson at adelaide.edu.au

http://mag.maths.adelaide.edu.au/crwilkinson.html