[Bioc-devel] Problem with normexp background correction in limma

[Edoardo Giacopuzzi]

Hi, 
I'm using limma package to analyze my .gpr files obtained from oligo
microarrays. I've performed 4 hybridization, with 2 dye-swap, and put
all data in the targets file.
After reading into limma the Targets files and the SpotType file, I
read the data from .gpr files using the function read.maimages as
follow:

> RG <- read.maimages(targets$FileName, source="genepix",
wt.fun=wtflags(0.01))

Then I read the GAL file using:
> RG$genes <- readGAL()

I also get the array print layout using:
> RG$printer <- getLayout(RG$genes) 

Now I want to correct background values using the backgroundCorrect
function present in limma and the normexp method, but it always
return me this error message:
> RGNormExp <- backgroundCorrect(RG, method="normexp", offset=25)
   Corrected array 1 
   Errore in optim(par = c(beta, log(sigma), log(alpha)), fn =  
normexp.m2loglik,  : 
   valore iniziale in 'vmmin' non finito

No problems instead if i use any other background correction
method...
What does this error message mean? 
Is there anything I can do to make normexp method work?
Thanks, Edoardo