[BioC] LIMMA: Help needed for limma analysis with a complex design

[Gordon K Smyth]

Dear Yiwen He,

See Sections 8.4 and 8.7 of the limma User's Guide:

http://www.bioconductor.org/packages/release/bioc/vignettes/limma/inst/doc/usersguide.pdf

I may be able to give you more detailed advice if you show the complete 
targets frame, so I can see how you have defined the experimental 
variables, and if you nominate which comparisons you are interested in 
testing.  I know you're interested in the effect of the treatment diet, 
but there many ways to interpret what you mean by an "effect".  For 
example, do you probably want to test for a difference between the special 
diet and the regular diet?  Are you interested in the difference between 
the before/after effects of the two diets?  On individual days?

Best wishes
Gordon

> Date: Thu, 1 Nov 2012 18:31:03 +0000
> From: "He, Yiwen (NIH/CIT) [C]" <heyiwen at mail.nih.gov>
> To: "bioconductor at r-project.org" <bioconductor at r-project.org>
> Subject: [BioC] LIMMA: Help needed for limma analysis with a complex
> 	design.
>
>
> Hi,
>
> I need help with analysis using the limma package. Here is the 
> experiment design:
>
> 20 patients, 10 on a regular diet and the other 10 on a special 
> treatment diet. For each patient, 4 samples were obtained: day 1, before 
> and after diet; day 6, before and after diet.
>
> Thus there are a total of 20x4=80 arrays. And the purpose of the study 
> is to find out the effect of the treatment diet.
>
> Here the treatment is nested with the patient (if that's the right way 
> to say it). It can be viewed as a paired analysis but the pair is for 
> both before/after treatment and day1/day6. I am at lost as how to create 
> a design matrix and call limma for such a study design. Can someone 
> please help?
>
> Any input will be highly appreciated!
>
> Yiwen He
> BIMAS/CIT/NIH

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}