R/read.gmt.R

In arcolombo/junk: qusage: Quantitative Set Analysis for Gene Expression

#' Simple function to read in a .gmt file and return a list of pathways
#' @param file  file gmt to read in
#' @export
#' @return gene matrix transposed set object 
read.gmt = function(file){
  if(!grepl("\\.gmt$",file)[1]){stop("Pathway information must be a .gmt file")}
  geneSetDB = readLines(file)                                ##read in the gmt file as a vector of lines
  geneSetDB = strsplit(geneSetDB,"\t")                       ##convert from vector of strings to a list
  names(geneSetDB) = sapply(geneSetDB,"[",1)                 ##move the names column as the names of the list
  geneSetDB = lapply(geneSetDB, "[",-1:-2)                   ##remove name and description columns
  geneSetDB = lapply(geneSetDB, function(x){x[which(x!="")]})##remove empty strings
  return(geneSetDB)
}