R/map_receptor.R

In Dahaniel/DoOR.functions: Integrating Heterogeneous Odorant Response Data into a Common Response Model: A DoOR to the Complete Olfactome

#' map_receptor
#' 
#' Identifying the source of unknown response data by correlating it agains all 
#' DoOR responding units.
#' 
#' @param data data frame, containing two columns, one called "odorants" and one
#'   "responses" providing InChIKeys and odorant responses respectively.
#' @param response_matrix output is a numeric vector that contains the Pearson 
#'   Correlation Coefficient between given data and selected consensus data in
#' @param nshow numeric, if defined, only this number of results will be
#' @param sub character, a subset of responding units returned response matrix
#' @param threshold.p numeric, a p-value threshold, only correlations below will
#'   be returned
#' @param threshold.cor numeric, a correlation-coefficient threshold, only
#'   correlations above will be returned
#' @author Shouwen Ma <\email{shouwen.ma@@uni-konstanz.de}>
#' @author Daniel Münch <\email{daniel.muench@@uni-konstanz.de}>
#' @export
#' @importFrom stats na.omit cor.test
#' @aliases mapReceptor map_receptor
#' @examples
#' # load data
#' load_door_data(nointeraction = TRUE)
#' 
#' # pick example data
#' data <- data.frame(odorants  = Or22a$InChIKey,
#'                    responses = Or22a$Hallem.2004.EN)
#' data <- na.omit(data)
#' 
#' # find the corresponding receptor / responding unit
#' map_receptor(data = data)
map_receptor <- function(data,
                 response_matrix = door_default_values("door_response_matrix"),
                 sub,
                 threshold.p,
                 threshold.cor,
                 nshow) {
  if (!("odorants" %in% colnames(data)) |
      !("responses" %in% colnames(data)))
    stop("Please provide InChIKeys and responses in columns named 'odorants' and
         'responses'.")
  
  data$odorants   <- as.character(data$odorants)
  data            <- na.omit(data)
  res             <- data.frame()
  response_matrix <-
    response_matrix[match(data$odorants, rownames(response_matrix)),]
  
  if (!missing(sub))
    response_matrix <-
    response_matrix[, match(sub, colnames(response_matrix))]
  
  # remove n < 3
  n <- which(apply(!is.na(response_matrix), 2, sum) < 3)
  if (length(n) > 0) {
    message(paste(
      "skipped ",
      paste(names(response_matrix)[n], collapse = ", "),
      " as overlap (n) was < 3",
      sep = ""
    ))
    response_matrix <- response_matrix[,-n]
  }
  
  result <-
    apply(response_matrix, 2, function(x)
      cor.test(x, data$responses))
  
  result <- data.frame(
    responding.unit = names(result),
    n               = apply(!is.na(response_matrix), 2, sum),
    cor             = unlist(lapply(result, "[", "estimate")),
    p.value         = unlist(lapply(result, "[", "p.value"))
  )
  
  result <- result[order(result$cor, decreasing = TRUE), ]
  
  if (!missing(threshold.p))
    result <- result[which(result$p.value <= threshold.p), ]
  
  if (!missing(threshold.cor))
    result <- result[which(result$cor >= threshold.cor), ]
  
  if (!missing(nshow))
    result <- result[1:nshow,]
  
  return(result)
}