R/normalize_mat.R

In tinyheero/CHL26predictor: An R package for the CHL 26 Gene Predictor

#' Generate the Normalizer Values for each Sample
#'
#' \code{get_normalizer} returns the normalizer values for each sample. The 
#' normalizer value is based on the geometric mean of the housekeeper genes
#' ACTB, CLTC, and RPLP0.  
#'
#' You can change the housekeeper genes by specifying the genes.norm.factor
#' parameter.
#'
#' @param mat The expression matrix
#' @param genes.norm.factor Character vector containing the housekeeper genes 
#'   to use to calculate the normalizer
#' @return A numeric vector with the samples as names and the normalizer 
#'   as values
#' @export
#' @examples
#' normalizer <- get_normalizer(mat)
get_normalizer <- function(mat, 
                           genes.norm.factor = c("ACTB", "CLTC", "RPLP0")) {

  message("Generating the geometric mean of housekeeper genes")

  if (!length(genes.norm.factor) > 1) {
    stop("Must have more than 1 housekeeper gene")
  }

  if (!all(genes.norm.factor %in% rownames(mat))) {
    missing.ind <- !genes.norm.factor %in% rownames(mat)
    stop(paste0("Matrix is missing the genes: ", genes.norm.factor[missing.ind]))
  }

  normalizer.log <- log(mat[genes.norm.factor, ])
  normalizer.log.mean <- apply(normalizer.log, 2, mean)
  normalizer.log.mean.exp <- exp(normalizer.log.mean)
  normalizer.log.mean.exp
}

#' Normalize the Expression Matrix 
#'
#' This function normalizes the expression matrix.
#' 
#' @param mat The expression matrix
#' @param normalizer Normalizer values generated by \code{get_normalizer}
#' @return A normalized expression matrix
#' @export
#' @examples
#' normalizer <- get_normalizer(mat)
#' mat.norm <- normalizer_mat(mat, normalizer)
normalize_mat <- function(mat, normalizer) {

  message("Normalizing the expression matrix")
  mat.norm <- sweep(mat, 2, normalizer, "/")
  mat.norm.scaled <- mat.norm * 1000

  # Ensure that any values < 1 do not become negative
  mat.norm.scaled[mat.norm.scaled < 1] <- 1

  message("Log2 transforming")
  mat.norm.scaled.log <- log(mat.norm.scaled, 2)
  mat.norm.scaled.log
}