R/ggnetwork_plot.R

In ovrhuman/HPOExplorer: Analysis and Visualisation of the Human Phenotype Ontology

#' Generate network plot
#'
#' This creates a network plot which is compatible with plotly to make interactive
#' plots. It makes it possible to hover box that includes your results related to
#' each phenotype and/or a description of the phenotypes.
#'
#' @param phenoNet The network object created using create_network_object
#' @param colour_column column name of the variable to be mapped to colour <string>
#' @param colour_label A label for the colour figure legend <string>
#' @returns A network plot (compatible with interactive plotly rendering).
#' @import ggplot2
#' @export
ggnetwork_plot <- function(phenoNet,
                           colour_column = "fold_change",
                           colour_label = "Fold change") {
  message("ggnetwork_plot")
  network_plot <-  ggplot(phenoNet, aes(x=x,y=y,xend=xend,yend=yend,text=hover)) +
    ggnetwork::geom_edges(color = "darkgray")+
    geom_point(aes_string(colour = colour_column, size = "hierarchy")) +  #, text= hover)) +
    geom_text(aes(label = label), color = "black") +
    scale_colour_gradient2(low = "white", mid = "yellow", high = "red") +
    scale_size(trans = "exp") +
    guides(size = "none") +
    labs(colour = colour_label) +
    theme_void() #  use tooltip = "hover" with ggplotly for hover box
  return(network_plot)
}