R/plot_over_time.R

In thijsjanzen/GenomeAdmixR: Simulate Admixture of Genomes

#' Plot the frequencies of all ancestors over time
#' @description This function plots the frequency of all ancestors over time at
#' a specific location on the chromosome, after performing a simulation.
#' @param frequencies  A tibble containing four columns: \code{time},
#' \code{location}, \code{ancestor}, \code{frequency}. A fifth colum
#' \code{population} can be included if the tibble is the result of
#' \code{simulate_admixture_migration}.
#' @param focal_location Location (in Morgan) where to plot the allele
#' frequencies.
#' @return a ggplot2 object
#' @examples
#' pop <- simulate_admixture(
#'            module = ancestry_module(number_of_founders = 10,
#'                                     markers = 0.5),
#'            pop_size = 1000,
#'            total_runtime = 11)
#' require(ggplot2)
#' plot_over_time(frequencies = pop$frequencies,
#'                focal_location = 0.5)
#' @export
plot_over_time <- function(frequencies,
                           focal_location) {

  p1 <- c()
  to_plot <- dplyr::filter(frequencies, .data[["location"]] == focal_location)
  if (length(to_plot$time) == 0) {
    stop("No data to plot, are you sure that the focal location\nis within the
        vector of molecular markers?\n")
  }

  if ("population" %in% colnames(frequencies)) {
    p1 <- ggplot2::ggplot(to_plot)  +
      ggplot2::geom_step(ggplot2::aes(x = .data[["time"]],
                                      y = .data[["frequency"]],
                                      col = interaction(.data[["ancestor"]],
                                                        .data[["population"]]
                                                        )
                                      )
      )
  } else {
    p1 <- ggplot2::ggplot(to_plot,
                          ggplot2::aes(x = .data[["time"]],
                                       y = .data[["frequency"]],
                                       col = as.factor(.data[["ancestor"]]))) +
      ggplot2::geom_step()
  }

  p1 <- p1 +
    ggplot2::xlab("Time (Generations)") +
    ggplot2::ylab("Frequency") +
    ggplot2::labs(col = "Ancestor")

  return(p1)
}