I am trying to do background correction/ quantile normalization on a .txt file of expression values (.CEL data is unavailable to me). GSE_data is a 47,000 x 73 dataframe with headers of expression values. The data was gathered on an Illumina microarray.
I'm trying to use lumi like so:
library(lumi) testMat <- as.matrix(GSE_data) testMat <- testMat[,-1] # to get rid of probe IDs in first column eset<-new("ExpressionSet",exprs=testMat) # background correction/RMA normalization lumiObj <- lumiB(eset, method='bgAdjust',verbose=TRUE)
But I get this error:
error in evaluating the argument 'object' in selecting a method for function 'exprs': Error in bgAdjust(x.lumi, ...) : The object should be class "LumiBatch"!
Can anyone see what I'm doing wrong? What's the best way to do background correction/quantile normalization on .txt data?
Thank you so much!
EDIT: When I rewrite the eset definition as:
eset<-new('LumiBatch', exprs = test)
I then get the error:
Error in `featureNames<-`(`*tmp*`, value = c("1", "2", "3", "4", "5", : 'value' length (47308) must equal feature number in AssayData (47308)
EDIT 2: Also, using lumiR doesn't seem to work - inputting this (importme = file in my path that is a .txt file of the expression values only)
try.lumi <- lumiR('importme')
I get this:
Error in gregexpr("\t", dataLine1)[] : subscript out of bounds