R/hs_SCAdiss.R

In CMET-UGent/MicroRaman: Phenotyping of microbial Raman spectroscopy data

#' Calculate the spectral contrast angle (SCA) between all cells in a hyperSpec object
#'
#'  The SCA measures the angle between two vectors corresponding to closely
#'  related spectra to measure whether they are the same or not
#'  (Wan *et al.* 2002). This function returns a dist object based on the SCA
#'  for downstream processing.
#'
#' @param x a hyperSpec object containing the cells in the rows, and wavenumbers (features) in columns
#' @importFrom stats as.dist
#' @return a distance object containing the SCA between each cell
#' @references Wan, K. X., Vidavsky, I., & Gross, M. L. (2002). Comparing similar spectra: from similarity index to spectral contrast angle. Journal of the American Society for Mass Spectrometry, 13(1), 85-88.
#' @examples
#' # Load hyperSpec object
#' data("hs_example")
#'
#' # Convert to MassSpectrum object
#' hs.x.proc <- hs_preprocess(hs_example)
#'
#' # cluster cells
#' disst <- hs_SCAdiss(hs.x.proc)
#' @export

hs_SCAdiss <- function(x) {
  if (!class(x) == "hyperSpec") {
    stop("You did not supply a valid hyperSpec object, and there is no default.")
  }
  matr <- x@data$spc
  diss <- matrix(nrow = nrow(matr), ncol = nrow(matr))
  # there has to be a way to get rid of this double for-loop
  # TODO: have a look at cov and see how it's done there (apparently with C calls)
  for (i in 1:nrow(matr)) {
    for (j in 1:nrow(matr)) {
      diss[i, j] <- SCA(matr[i, ], matr[j, ])
    }
  }
  row.names(diss) <- rownames(x@data$spc)
  diss <- as.dist(diss)
  attr(diss, "method") <- "hs_SCAdiss"
  return(diss)
}