R/nattswap_compare.R

In Epsian/nattswap: Generate Simulated Networks With Swapped Attributes

#' nattswap_compare
#'
#' @param get_att_swap_sig_results Result of \code{get_att_swap_sig}
#' @param metric_name Character. A vector of "degree", "evc", "bet", "n_bet", "closeness", "max_cohesion", or "nestedness". Defaults to "all" which will display all metrics.
#' @param swapped_att Character. Name of swapped attribute.
#' @param all_nrow How many rows do you want the plot to have?
#' @param all_ncol How many columns do you want the plot to have?
#' @param font_family Font family to use in plots.
#' @param yaxis Vector of 2 integers. Lower and upper limit of y axis.
#' @param xaxis List of with element names matching plot types \(e.g. degree\). List contains length 2 vectors specifying x axis limits for plot of that type.
#' @param bin_num The number of bins you want in the histograms.
#'
#' @return A ggplot2 plot showing real vs simulated means
#' @export
#'
#' @examples
nattswap_compare = function(get_att_swap_sig_results, metric_name = "all", swapped_att, all_nrow = 1, all_ncol = NULL, font_family = NULL, yaxis = c(NA, NA), xaxis = NA, bin_num = 50){

  # check if y-axis set is formatted correctly
  if(length(yaxis) != 2){stop("yaxis need to be vector with length 2!")}

  # make empty list for plots to be stored in
  plot_list = list()

  if("degree" %in% metric_name | "all" %in% metric_name){
    plot_list[["degree"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "degree_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_degree"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Degree Centrality", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_degree"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
    }

  if("evc" %in% metric_name | "all" %in% metric_name){
    plot_list[["evc"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "evc_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_evc"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Eigenvector Centrality", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_evc"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
    }

  if("closeness" %in% metric_name | "all" %in% metric_name){
    plot_list[["closeness"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "closeness_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_closeness"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Closeness", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_closeness"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
    }

  if("bet" %in% metric_name | "all" %in% metric_name){
    plot_list[["bet"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "bet_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_bet"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Betweenness Centrality", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_bet"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
    }

  if("n_bet" %in% metric_name | "all" %in% metric_name){
    plot_list[["n_bet"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "n_bet_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_n_bet"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Normalized Betweenness Centrality", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_n_bet"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
  }

  if("max_cohesion" %in% metric_name | "all" %in% metric_name){
    plot_list[["max_cohesion"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "max_cohesion_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_max_cohesion"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Max Cohesion", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_max_cohesion"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
  }

  if("nestedness" %in% metric_name | "all" %in% metric_name){
    plot_list[["nestedness"]] = ggplot2::ggplot(get_att_swap_sig_results[["means"]], ggplot2::aes(x=get_att_swap_sig_results[["means"]][, "nestedness_mean_true"])) +
      ggplot2::geom_histogram(color="black", fill="white", bins = bin_num) +
      ggplot2::geom_vline((ggplot2::aes(xintercept=get_att_swap_sig_results[["z-report"]][["r_nestedness"]])), color="black", linetype="dashed", size=1) +
      ggplot2::labs(x = "Mean Nestedness", y = "Frequency") +
      ggplot2::labs(caption = paste0("Z-Score = ", gsub("0\\.", "\\.", sprintf("%.3f", round(get_att_swap_sig_results[["z-report"]][["z_nestedness"]], digits = 3))))) +
      ggplot2::theme(text=ggplot2::element_text(family= font_family), plot.caption = ggplot2::element_text(size = 12))
  }

  # alter x axis of plot if given list
  # expects a list where each item matches a plot name (e.g. "degree")
  if(!is.na(xaxis[1])){
    # test if it is a list
    if(!is.list(xaxis)){stop("Expect a list with elements that match plot types (e.g. 'degree') whose value is a vector of 2 with min and max for x.")}

    # for each named element in xaxis list, apply to the x axis of output plot
    for(name in names(xaxis)){

      # apply the xlimits to plot matching name
      plot_list[[name]] = plot_list[[name]] + ggplot2::scale_x_continuous(limits = xaxis[[name]])
    }

  }

  # standardize y axis if numbers were given
  if(!all(is.na(yaxis))){
    plot_list = lapply(plot_list, function(iplot){
      iplot = iplot + ggplot2::scale_y_continuous(limits = yaxis)

      return(iplot)
    })
  }

  # combine plot for output
  out_plot = gridExtra::grid.arrange(grobs = plot_list, nrow = all_nrow, ncol = all_ncol, top = ggpubr::text_grob(paste0("Simulated vs. Observed for ", swapped_att), family = font_family))

  return(out_plot)
}