virtualArrayBuildfData: Collapses expression values according to given identifiers
In virtualArray: Build virtual array from different microarray platforms

Description Usage Arguments Details Value Author(s) See Also Examples

Expression values are collapsed according to given identifiers, e.g. gene symbols. In the same run, a data.frame with 2 columns to fill the "fData" slot of the ExpressionSet is build. The data is pulled from the Bioconductor annotation package defined in the "annotation" slot of the ExpressionSet.

1	virtualArrayBuildfData(x, identifier = "SYMBOL", collapse_fun = median)

`x`	Name of an ExpressionSet as a character vector
`identifier`	Wich identifier to pull from the annotation (default="SYMBOL"). Theoretically it can be anything that is supported by the annotation package. Practically it is limited to identifiers giving a 1 to 1 mapping, such as ENTREZID, SYMBOL, GENENAME, UNIPROT, UNIGENE, etc. Identifiers giving 1 to 1+ mappings will be supported in future versions of the package.
`collapse_fun`	Which function to use to treat multiple lines targetting the same gene (default=median)

This function is normally only called from "virtualArray.ExpressionSet". You can use it, however, to collapse expression values referring to the same identifier together with their corresponding annotation in the "exprs" and "fData" slots by means of a user specified function (default is "median"). Note, that it is critical to define the correct Bioconductor annotation package in the "annotation" slot of the ExpressionSet and that the name of the ExpressionSet must be supplied as a character vector, due to the implementation in "virtualArray.ExpressionSet".

The value returned is an ExpressionSet whose "fData" slot has been filled with the selected identifiers. The rows in the expression matrix ("exprs" slot) have been collapsed to the selected identifiers. Therefore the size of the expression matrix has decreased during the process.

Andreas Heider

virtualArray-package, virtualArray.ExpressionSet, virtualArrayCompile

## first we need to load dummy data
library(affydata)
data(Dilution)
## we apply RMA to get an ExpressionSet
Dilution <- rma(Dilution,normalize=FALSE)
## now we collapse the expression values as stated in "Details"
Dilution_genesymbols <- virtualArrayBuildfData(x="Dilution")
## as you can see, we now have only one row per gene symbol
Dilution_genesymbols