[BioC] combineAffyBatch & GCRMA?

[Michael Seewald]

Dear Robert, dear Bioconductor people,

I would like to use combineAffyBatch to merge a dataset of Affymetrix
HG-U133Plus2 and HG-U133A arrays. The main point here is the comparison of the
two arrays - and possibly generation of a unified dataset that is valid for
further analysis. (Disposing half of the Plus2 information is not of concern
for now..)

So far I used the following to combine and normalize with RMA (works great!):

# DATA.133A, DATA.133Plus are generated by ReadAffy()
> DATA.combine <- combineAffyBatch( list(DATA.133A,DATA.133Plus),
+    c("hgu133aprobe","hgu133plus2probe"),
+    newcdf="newcdfenv", verbose=T
+ )
> newcdfenv <- DATA.combine$cdf
> DATA.A.RMA <- rma(DATA.combine$dat)

Then I run into trouble because (I believe) I have to work with the cdf
environment:
> DATA.A.GCRMA <- gcrma(DATA.A)
Computing affinitiesError in library(cdfpackagename, character.only = TRUE) :
        There is no package called 'newcdfenvcdf'

I tried also:
> ai <- compute.affinities(cdfname="newcdfenv",verbose=T)
Error in library(cdfpackagename, character.only = TRUE) :
        There is no package called 'newcdfenvcdf'

How can I create the correct "newcdfenvcdf"? And why doesn't
specifying "newcdfenv" not help?

Best wishes,
Michael