R/frapplot.R

In GuanqiaoDing/frapplot: Automatic Data Processing and Visualization for FRAP

#' Plot FRAP data of two selected groups
#'
#' @importFrom grDevices dev.off pdf
#' @importFrom graphics arrows axis legend lines par plot points
#'
#' @description Plot FRAP data of any two groups (e.g. control and mutant) in a consistent and publishable format.
#'
#' @param path Path of the output directory
#' @param control Name of the control.
#' @param mutant Name of the mutant.
#' @param info Returned information from [frapprocess()].
#'
#' @examples
#' info <- frapprocess(example_dataset, seq(0, 145, 5))
#' frapplot(tempdir(), "control", "mut2", info)
#'
#' @export

frapplot <- function(path, control, mutant, info){

  # retrieve information from frapprocess
  x <- info$time_points
  summary <- info$summary
  sample_means <- info$sample_means
  sample_sd <- info$sample_sd
  mod <- info$model

  # validate input
  if(!(control %in% summary$group_names)) {
    stop("Name of the control provided (first argument) is not found.")
  }
  else if(!(mutant %in% summary$group_names)) {
    stop("Name of the mutant provided (second argument) is not found.")
  }

  # create file
  filename <- paste(path, "/", control, "_", mutant, ".pdf", sep = "")
  pdf(filename, width = 8, height = 6)

  index1 <- which(summary$group_names == control)
  index2 <- which(summary$group_names == mutant)
  y1 <- sample_means[, index1]
  y2 <- sample_means[, index2]

  par(mar=c(5, 5, 3, 3))

  # plot data points and axis label
  plot(x, y1,
       pch = 21, bg = 'black', col = 'black',
       axes = FALSE, xlab = "Time (s)", ylab = "Fractional Recovery",
       cex.lab = 1.5, xlim = c(0, 150), ylim = c(0, 1),
       mgp = c(2, 1, 0), cex.axis = 1.2)
  points(x, y2, pch = 21, bg = 'blue', col = 'blue')

  # plot axes
  axis(1, seq(0, 150, 25), tck = -0.02,
       cex.axis = 1.2, mgp = c(2, 1, 0), pos = 0)
  axis(2, seq(0, 1, 0.2), c("0", "0.2", "0.4", "0.6", "0.8", "1.0"),
       tck = -0.02, cex.axis = 1.2,
       mgp = c(2, 1, 0), pos = 0, las = 1)

  # plot error bars
  upper <- sample_means + sample_sd
  lower <- sample_means - sample_sd
  arrows(x, upper[, index1], x, lower[, index1],
         length = 0.05, angle = 90, code = 3, lwd = 1.5, col = 'black')
  arrows(x, upper[, index2], x, lower[, index2],
         length = 0.05, angle = 90, code = 3, lwd = 1.5, col = 'blue')

  # plot fitted curves
  lines(x, predict(mod[[index1]], list(x)), lwd = 2, col = 'black')
  lines(x, predict(mod[[index2]], list(x)), lwd = 2, col = 'blue')

  # plot legend
  legend(x = 5, y = 1.05, legend = c(control, mutant),
         lty = c(1,1), lwd = 2, pch = c(20, 20), col = c('black', 'blue'),
         bty = 'n', cex = 1.5)

  dev.off()
}