[Bioc-devel] Splitting design matrix contrast pivoting logic into a separate function?

Dear Ryan,

See contrastAsCoef() in limma 3.17.23.

Best wishes
Gordon

---------------------------------------------
Professor Gordon K Smyth,
Bioinformatics Division,
Walter and Eliza Hall Institute of Medical Research,
1G Royal Parade, Parkville, Vic 3052, Australia.
http://www.statsci.org/smyth

On Thu, 29 Aug 2013, Ryan C. Thompson wrote:

Hi Gordon,

I am currently working on some code that would benefit from having
direct access to the code in edgeR::glmLRT that handles
reparametrizing the design matrix so that the N linearly independent
components of the specified contrasts are represented in the first N
columns of the design matrix (i.e. the part of glmLRT that does all
the QR decomposition stuff). Would you consider splitting this logic
into a separate function that takes a design matrix and contrast
matrix and returns the equivalent reparametrized design matrix and
vector of coefficients?

Thanks,

-Ryan Thompson