# install necessary bioconductor packages # (only first time, see www.bioconductor.org) source("http://www.bioconductor.org/biocLite.R") biocLite("oligo") biocLite("pd.huex.1.0.st.v2") # normalize CEL files using oligo package (version 1.10.0 or newer) library(oligo) celFiles <- list.celfiles('/path/to/my/celfiles', full.name=TRUE) raw <- read.celfiles(celFiles, pkgname="pd.huex.1.0.st.v2") eset <- rma(raw, target="probeset") # alternatively, load normalized log2-expression data from text file # to obtain an ExpressionSet object with probeset expression levels: #library(affy) #eset <- readExpressionSet('/path/to/my/expression/file.txt') # load cosie functions source('cosie.R') # calculate corrected splicing indices cosieOut <- cosie(eset, '/path/to/cosie/parameter/file.txt') # write cosie results to tab-delimited text file with columns: # psId : probeset identifier # tcId : transcript cluster identifier # SI_* : corrected pre-splicing index # (one column per sample, sample name in '*', # subtract two of these columns in order to # obtain the differential splicing index) # tcLevel_* : trancript cluster expression level # (one column per sample, sample name in '*') write.cosie(cosieOut, 'cosieOut.txt')