I am trying to find a package to use for analysis of protein microarray data. The experiment involved one control and 4 treatment conditions. A ProteomicProfiler microarray for 55 proteins was used to detect the effects of each treatment on the panel of 55 proteins. This array uses protein-specific antibodies to capture and fluorescence secondary antibody to identify. Each sample was spotted twice for each protein. All the arrays were set up at the same time, and read at the same time (one batch). The arrays were read using a LiCor instrument, and the results outputted to an excel file. Within the excel file I can do the background subtract, and check the "standards" for reliability of the runs. But what I need is to normalize/standardize the data, and model them correctly to look at fold changes, etc. I have done a lot of searching through Bioconductor for protein microarray analytical packages, but everything I find is specific to a specific microarray type (like the PAA package), requiring input of raw data from a specific type of analyzer. Is there one which will work with the generic data out of an excel file? Here is an example of the data I have:

PROTEIN  CONTROL  TREATMENT 1  TREATMENT 2  TREATMENT 3  TREATMENT 4
      A       39           72          180           62           50
      A       32           75          149           66           44
      B       96           99          120          137          145
      B       89           99          122          142          140
      C       46           40           85           90           70
      C       43           37           80           88           77

These are just fictional values, not the real data, but they are representative of the range and changes that I have seen. The units are fluorescence per pixel on the blot. Thanks for suggestions of appropriate packages to use.
- hcnbox

The limma package is frequently used to analyse microarray data, including protein microarrays. limma accepts expression values in any format, e.g., a matrix of fluorescences as you seem to have.

The fluorescences have to be for spots (representing proteins or peptides) rather than for pixels. It seems unlikely to me that LiCor would output values per pixel.

The analysis can be conducted pretty much as for any microarray: log the flourescences, background correct and normalize, use duplicateCorrelation to handle the duplicate spots for each protein, etc etc.