I am analyzing an Affymetrix array dataset. After RMA normalization the data is returned as an expressionSet object. However, I am not able to replace the expression value 'slot' with some new data anymore, although this did work before (BioC 3.3). See example below. Any suggestions on how to do this in BioC 3.4? Thanks.

--> The normalized data I get is REFSEQ-based, but I am averaging (collapsing) the data on the level of ENTREZG IDs, which I then would like to 'put back' in the expressionSet. The number of genes/features is reduced from ~60k to ~20k. All other (meta)data is not affected.

> library(Biobase)
> sample.data <- ExpressionSet(assayData=matrix(runif(1000), nrow=100, ncol=10))
> sample.data
ExpressionSet (storageMode: lockedEnvironment)
assayData: 100 features, 10 samples
  element names: exprs
protocolData: none
phenoData: none
featureData: none
experimentData: use 'experimentData(object)'
Annotation:  
> replacement <- matrix(runif(750), nrow=75, ncol=10)
> exprs(sample.data) <- replacement
Error in .validate_assayDataElementReplace(obj, value) :
  object and replacement value have different dimensions
>

> sessionInfo()
R version 3.3.1 Patched (2016-10-18 r71535)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=English_United States.1252
[2] LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
[1] Biobase_2.34.0      BiocGenerics_0.20.0
>

The ExpressionSet has 100 rows and 10 columns. The replacement data has only 75 rows and 10 columns. Which 75 rows and 10 columns are you interested in replacing, e.g.,

exprs(sample.data)[1:75,] <- replacement

Or perhaps as in your example there is no pData or fData, and you really just want to create

ExpressionSet(replacement)

?