R/clean_profiles.R

In PrInCE: Predicting Interactomes from Co-Elution

#' Preprocess a co-elution profile matrix
#' 
#' Clean a matrix of co-elution/co-fractionation profiles by 
#' (1) imputing single missing
#' values with the average of neighboring values, (2) replacing missing values
#' with random, near-zero noise, and (3) smoothing with a moving average
#' filter. 
#' 
#' @param profile_matrix a numeric matrix of co-elution profiles, with proteins
#' in rows, or a \code{\linkS4class{MSnSet}} object
#' @param impute_NA if true, impute single missing values with the average of
#' neighboring values 
#' @param smooth if true, smooth the chromatogram with a moving average filter
#' @param smooth_width width of the moving average filter, in fractions 
#' @param noise_floor mean value of the near-zero noise to add 
#' 
#' @return a cleaned matrix
#' 
#' @examples
#' data(scott)
#' mat <- scott[c(1, 16), ]
#' mat_clean <- clean_profiles(mat)
#' 
#' @importFrom MSnbase exprs
#' @importFrom Biobase exprs<-
#' @importFrom methods is
#' 
#' @export
clean_profiles <- function(profile_matrix, impute_NA = TRUE, smooth = TRUE, 
                           smooth_width = 4, noise_floor = 0.001) {
  if (is(profile_matrix, "MSnSet")) {
    profile_matrix <- exprs(profile_matrix)
  }
  
  profile_matrix <- t(apply(profile_matrix, 1, clean_profile, 
                            impute_NA = impute_NA,
                            smooth = smooth, 
                            smooth_width = smooth_width, 
                            noise_floor = noise_floor))
  
  return(profile_matrix)
}