[BioC] Limma and creating design matrices for complex loop designs

Naomi Altman naomi at stat.psu.edu
Fri Dec 9 15:53:49 CET 2005

In an experiment with 3 or more conditions (treatments, levels ...) 
it is usually necessary to have more contrasts than can be 
accommodated by the design matrix.
That is why there is fit.contrast as well as lmFit.

My approach is to loop designs is to use single channel analysis.  It 
is simplest to pick an arbitrary condition as the reference for the 
design matrix, or to leave out the intercept and fit all the 
conditions.  (This puts the condition mean in the "M" 
matrix.)  Either way, the main purpose of this step is to get the MSE 
for each gene, which is required for the contrasts.

Then I do all the contrasts of interest as calls to fit.contrast.

My paper on using single channel analysis for loop designs can be 
at  http://www.stat.psu.edu/~zhao/bcc/respapers/AltmanInterface04.doc 
but currently it does not discuss the limma implementation of this.


At 06:21 AM 12/9/2005, michael watson (IAH-C) wrote:
>I would like to respectfully request more examples of the creation of
>design matrices for complex (ie involving more than 3 samples) loop
>designs using two-colour microarrays in the limma documentation.
>Although section 20.5 does give an example, with 5 RNA samples, one of
>these is a pool which makes an obvious reference when creating the
>design matrix as it is a coefficent we are not interested in estimating.
>However, other loop designs may not contain such an obvious sample, and
>I am struggling to figure out a way to create a design matrix for my
>current design that allows all contrasts of interest.  In fact, at
>present, I have to create one design matrix to get some of the contrasts
>and another to get the rest.  Is this a valid approach?
>Many thanks
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch

Naomi S. Altman                                814-865-3791 (voice)
Associate Professor
Dept. of Statistics                              814-863-7114 (fax)
Penn State University                         814-865-1348 (Statistics)
University Park, PA 16802-2111

More information about the Bioconductor mailing list