[BioC] Limma, dye-swaps and uneven numbers of replicate arrays

[Gordon K Smyth]

> Date: Thu, 1 Jun 2006 11:38:02 +0100
> From: "michael watson \(IAH-C\)" <michael.watson at bbsrc.ac.uk>
> Subject: [BioC] Limma, dye-swaps and uneven numbers of replicate
> 	arrays
> To: <bioconductor at stat.math.ethz.ch>
>
> Hi
>
> A simple question really - I have five biological replicate RNA samples.
> Each will have cells untreated vs treated, so that's five arrays.  I
> want to do dye-swaps, but that means I can have 2* Cy5/Cy3, and 3*
> Cy3/Cy5. Do I need to compensate for this "imbalance"?  How do I do it
> in limma?
>
> Thanks
> Mick

Such imbalance is accomodated automatically, just as is any imbalance in any linear model you
might fit.  It requires no special action from you.

Best wishes
Gordon