Plot networks with identified communities — plot

A specialized function to plot network structures, overlaying community detection (clusters) via convex hulls and allowing for mapping of theoretical domains to node aesthetics. It utilizes ggplot2::ggplot() and ggforce::geom_mark_hull() for rendering.

Usage

plot_clusters(
  net,
  layout = "spring",
  original.categories = NA,
  pointsize = 10,
  concavity = 3,
  legend.title = "",
  labels = NA,
  labelsize = 3,
  radius = 0.04,
  shape = 21,
  outline = F,
  colored = F,
  alpha = 1,
  cluster.base = net,
  fill.colors = NA,
  show.clusters = T,
  ...
)

Arguments

net: Network produced by bootnet::estimateNetwork().
layout: Character string or matrix specifying the network layout. Passed to qgraph::qgraph(). Defaults to "spring".
original.categories: A data.frame with two columns, "label" (categories) and "L1" (nodes). Used for assigning nodes to theoretical domains.
pointsize: Numeric scalar or vector specifying node size(s). Defaults to 10.
concavity: Numeric value for the concavity of the cluster hulls. Passed to ggforce::geom_mark_hull(). Defaults to 3.
legend.title: Character string for the fill legend title. Defaults to "".
labels: Custom node labels, should be in order of the nodes in the net$graph. Defaults to NA.
labelsize: Numeric value specifying the size of node labels. Defaults to 3.
radius: Numeric value specifying the corner radius of cluster hulls. Passed to ggforce::geom_mark_hull(). Defaults to 0.04.
shape: Numeric value specifying the ggplot2::geom_point() shape for nodes. Defaults to 21.
outline: Logical. If TRUE, uses ggplot2::geom_label() to draw a box around text. If FALSE, uses ggplot2::geom_text(). Defaults to FALSE.
colored: Logical. If TRUE, edges are colored (e.g., red/blue for negative/positive). If FALSE, edge sign is represented by linetype. Defaults to FALSE.
alpha: Numeric value between 0 and 1 specifying edge transparency. Defaults to 1.
cluster.base: Network object to use for community detection. Defaults to the net argument.
fill.colors: Vector of character colors for the node fill aesthetic. If NA, defaults to a colorblind-friendly palette.
show.clusters: Logical. If TRUE, calculates communities and plots convex hulls around them. Defaults to TRUE.
...: Additional arguments passed to EGAnet::community.detection().

Value

A ggplot2::ggplot() object representing the network graph.

Examples

mtcars = datasets::mtcars
library(bootnet)
#> Loading required package: ggplot2
#> This is bootnet 1.9.1
#> For questions and issues, please see github.com/SachaEpskamp/bootnet.
library(EGAnet)
#> Error in library(EGAnet): there is no package called ‘EGAnet’

# Estimate network
network <- estimateNetwork(mtcars, default = "EBICglasso")
#> Estimating Network. Using package::function:
#>   - qgraph::EBICglasso for EBIC model selection
#>     - using glasso::glasso
#> Warning: A dense regularized network was selected (lambda < 0.1 * lambda.max). Recent work indicates a possible drop in specificity. Interpret the presence of the smallest edges with care. Setting threshold = TRUE will enforce higher specificity, at the cost of sensitivity.

# Plot with clusters
plot_clusters(network, layout = "spring", show.clusters = TRUE)
#> Error in loadNamespace(x): there is no package called ‘EGAnet’