R/weighted_down.R

In pcahan1/cancerCellNet: CellNet, for cancer

#' @title
#' Down Sample Gene Expressions
#'
#' @description
#' Weighted subtraction from mapped reads and simulate expression profile of  _total_ mapped reads
#' for normalization.
#'
#' @param expRaw matrix of total mapped reads per gene/transcript
#' @param total numeric post transformation sum of read counts
#' @param dThresh the threshold at which anything lower than that is 0
#'
#' @return matrix of downsampled read mapped to genes/transcripts
#'
#' @export
weighted_down<-function(expRaw, total=1e5, dThresh=0) {
  expCountDnW = apply(expRaw, 2, downSampleW, total=total, dThresh=dThresh)

  return(expCountDnW)
}

#' Weighted subtraction from mapped reades
#'
#' Simulate expression profile of  _total_ mapped reads as a way to normalize the counts
#'
#' @param vector a vector of total mapped reads per gene/transcript
#' @param total the sum of the read counts after transformation
#' @param dThresh the threshold at which anything lower is 0 after transformation. Usually 0
#'
#' @return vector of downsampled read mapped to genes/transcripts
downSampleW<-function(vector, total=1e5, dThresh=0){

  totalSignal = sum(vector) # get the sum of the vector
  wAve = vector/totalSignal 

  resid = totalSignal-total #num to subtract from sample
  residW = wAve*resid 

  ans = vector-residW
  ans[which(ans<dThresh)] = 0

  return(ans)
}