[BioC] Error when using limma contrasts.fit

[Adam Witney]

Hi,

I am going through the limma User Guide and am having a problem using
contrasts.fit for my data.

I have three biological replicates, two of which have duplicate arrays and
one has triplicate arrays. I have a targets file like so:

SlideNumber    FileName    Cy3    Cy5
  A1            File1      wt1    mu1
  A2            File2      mu1    wt1
  B1            File3      mu2    wt2
  B2            File4      wt2    mu2
  C1            File5      mu3    wt3
  C2            File6      mu3    wt3
  C3            File7      wt3    mu3

I am simply trying to follow userguide section 11.2, but am having problems:

> design <- modelMatrix(targets, ref="wt1")

Found unique target names:
 mu1 mu2 mu3 wt1 wt2 wt3

> design <- cbind(Dye=1, design)
> colnames(design)

[1] "Dye" "mu1" "mu2" "mu3" "wt2" "wt3"

> fit <- lmFit(MA, design)

Coefficients not estimable: wt2 wt3

> cont.matrix <- makeContrasts(MUvsWT=(mu1+mu2+mu3-wt2-wt3)/3, levels=design)
> fit2 <- contrasts.fit(fit, cont.matrix)

Error in contrasts.fit(fit, cont.matrix) :
    trying to take contrast of non-estimable coefficient
 

Am I doing this correctly?

Also I noticed that the userguide refers to designMatrix() which I believe
has been deprecated in favour of modelMatrix(), should this have been
changed in the userguide?

(R-2.1.1, limma 2.0.5, MacOSX 10.3.9)

Thanks for any help

Adam



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.