R/qc_rlsc.R

In ricoderks/Rcpm: General functions for the Center of Proteomics and Metabolomics (LUMC)

#' @title Perform QC-RLSC for batch correction of chromatographic signal
#' 
#' @description Perform QC-RLSC for batch correction of chromatographic signal.
#'
#' @param tab table N*K (row * column) with N samples and K variables.
#' @param colv vector N of numbers: 1 for QC samples and 2 for other samples.
#' @param or vector of measuring order (see details).
#' @param verbose print which variable has been corrected to monitor the process (default = FALSE).
#'
#' @return corrected table N*K
#' 
#' @details Make sure that everything is sorted in measurement order!!!
#' 
#' @export
#' @importFrom stats loess approx
#'
#' @author E. Nevedomskaya
#' @author Rico Derks
#' @references Dunn et al. Nature Protocols 6, 1060-1083 (2011)
qc_rlsc <- function(tab, colv, or, verbose = FALSE) {
	# create table of the same size as initial
	tab_corr <- tab
	
	# For each variable (columns) in the initial table
	for (i in 1:ncol(tab)) {
		# fit loess curve to the QCs
		ll <- loess(tab[which(colv == 1), i] ~ or[which(colv == 1)])
		
		# approximate the curve for all the samples
		aa <- approx(x = or[which(colv == 1)],
		             y = ll$fitted, 
		             xout = or)
		
		# correct the variable according to the curve for all the samples
		tab_corr[, i] <- tab[, i] / aa$y
		
		# print which variable has been corrected in order to monitor the progress
		if(verbose == TRUE) {
		  print(i)  
		}
		
	}
	
	return(tab_corr)
}