R/plotSampleInfo.R

In DavisLaboratory/standR: Spatial transcriptome analyses of Nanostring's DSP data in R

#' Plot the user-defined meta data using alluvium plot
#'
#' @param spe_object A SpatialExperiment object.
#' @param column2plot Which columns to plot.
#' @param textsize text size.
#' 
#' @importFrom ggalluvial StatStratum
#'
#'
#' @return A ggplot object
#' @export
#'
#' @examples
#' library(ggalluvial)
#' 
#' data("dkd_spe_subset")
#' plotSampleInfo(dkd_spe_subset, column2plot = c("SlideName", "disease_status", "region"))
#'
plotSampleInfo <- function(spe_object, column2plot, textsize = 3) {
  checkPackages("ggalluvial")

  stopifnot(all(column2plot %in% colnames(colData(spe_object))))
  
  # using alluvium plot to visualise user-defined metadata in the data.
  colData(spe_object) |>
    as.data.frame(optional = TRUE) |>
    dplyr::select(column2plot) |>
    ggalluvial::to_lodes_form() |>
    ggplot(aes(
      x = x, stratum = stratum,
      alluvium = alluvium, fill = stratum, label = stratum
    )) +
    ggalluvial::geom_flow(
      stat = "alluvium", lode.guidance = "frontback",
      color = "NA"
    ) +
    ggalluvial::geom_stratum(alpha = .5) +
    geom_text(stat = "stratum", size = textsize) +
    theme_light() +
    theme(legend.position = "none") +
    xlab("Metadata") +
    ylab("Frequency")
}

utils::globalVariables(c("x", "stratum", "alluvium"))