R/duplicateRemover.R

In HTSanalyzeR: Gene set over-representation, enrichment and network analyses for high-throughput screens

##This function gets rid of the duplicates in a gene list.

duplicateRemover <- function(geneList, method = "max") {
	##check arguments
	paraCheck("genelist", geneList)
	paraCheck("duplicateRemoverMethod", method)
	##Get the unique names and create a vector that will store the 
	##processed values corresponding to those names
	geneList.names <- names(geneList)
	datanames <- unique(geneList.names)
	data.processed <- rep(0, length(datanames))
	names(data.processed) <- datanames
	data.processed2 <- data.processed
	l.data.processed <- length(data.processed)	
	##If the absolute value of the min is bigger than the absolute 
	##value of the max, then it is the min that is kept		
	if(method=="max") {
		data.processed<-sapply(datanames, 
				function(i) {
					this.range<-range(geneList[which(geneList.names==i)])
					this.range[which.max(abs(this.range))]
				}
		)
	} else if(method=="min") {
		data.processed<-sapply(datanames, 
				function(i) {
					this.range<-range(geneList[which(geneList.names==i)])
					this.range[which.min(abs(this.range))]
				}
		)
	} else if(method=="average") {
		data.processed<-sapply(datanames, 
				function(i) {
					mean(geneList[which(geneList.names==i)])
				}
		)
	} else if(method=="fc.avg") {
		neg.fcs<-which(geneList < 1) 
		geneListRatios<-geneList
		#convert from fold change to ratio		
		geneListRatios[neg.fcs]<-abs(1/geneList[neg.fcs])
		#average the values across replicates		
		data.processed<-sapply(datanames,function(i) {
			mean(geneListRatios[which(geneList.names==i)])
		})	
		#convert back to fold change				
		neg.fcs<-which(data.processed<1)
		data.processed[neg.fcs]<-(-1/data.processed[neg.fcs]) 
	}	
	data.processed
}