Colour the nodes and edges of networks, or the clusters of an alluvial

color_networks() gives a colour to each category of a node column (typically a cluster column) and stores it in a color column of the nodes. Each edge gets a colour mixing the colours of its two ends, so that edges within a cluster take the colour of the cluster and edges between clusters a blend of both.

color_alluvial() does the same for the table of an alluvial plot created with networks_to_alluv().

Usage

color_networks(
  graphs,
  column_to_color,
  color = NULL,
  unique_color_across_list = FALSE,
  verbose = TRUE
)

color_alluvial(alluv_dt, column_to_color, color = NULL, verbose = TRUE)

Arguments

graphs

A tbl_graph or a list of tbl_graphs.

column_to_color

Character scalar, the column with the categories to colour, e.g. cluster_leiden (add_clusters()) or dynamic_cluster_leiden (merge_dynamic_clusters()).

color

The colours: NULL (default palettes, see the details), a character vector of colours, given to the categories by decreasing number of nodes, or a data frame whose first column holds the categories and the second column, named color, their colours.

unique_color_across_list

Logical, for a list of graphs. FALSE (default) gives a category the same colour in every graph: use it for dynamic clusters. TRUE treats the same value in two graphs as two categories (cluster "01" of two windows gets two colours): use it for clusters detected separately in each window.

verbose

Logical. If TRUE (default), reports when colours are recycled or truncated.

alluv_dt

A data frame created with networks_to_alluv().

Value

color_networks(): the same object as graphs with a color column for nodes and edges. color_alluvial(): alluv_dt with a color column.

Details

Giving the same colour to a cluster in every figure (network plots, alluvial, tables) is what lets a reader follow it. Pass the same data frame of colours to color_networks() and color_alluvial() for this.

Without color, the function uses the ggplot2 palette (without black) for up to 7 categories, completed by the Okabe-Ito palette (without black and grey) for up to 14; beyond 14 categories, the colours are recycled and several clusters share a colour. For many clusters, give a palette explicitly (e.g. from the scico or paletteer packages), or colour only the largest clusters and give the others a neutral grey.

A vector with fewer colours than categories is recycled, a longer one is truncated. A data frame of colours that misses some categories leaves their colour missing.

Examples

network <- build_network(
  nodes = subset(Nodes_stagflation, source_type == "Stagflation"),
  directed_edges = Ref_stagflation,
  source_id = "source_id",
  target_id = "target_id",
  method = "similarity",
  edges_threshold = 1
)
#> ℹ Coupling (`biblionetwork::biblio_coupling()`) with "similarity": ties
#>   weighted by the idf-weighted cosine of the two lists of targets, after Shen
#>   et al. (2019), between source_id entities sharing target_id entities.
#> ℹ Two source_id are linked when they share at least 1 target_id
#>   (`edges_threshold`).
#> ℹ Nodes without any edge are removed (`keep_singleton = FALSE`).
network <- add_clusters(
  network,
  clustering_method = "leiden",
  objective_function = "modularity",
  seed = 123,
  verbose = FALSE
)

network <- color_networks(network, column_to_color = "cluster_leiden")
unique(tidygraph::as_tibble(network, active = "nodes")[c("cluster_leiden", "color")])
#> # A tibble: 4 × 2
#>   cluster_leiden color  
#>   <chr>          <chr>  
#> 1 02             #00BA38
#> 2 01             #F8766D
#> 3 04             #00BFC4
#> 4 03             #619CFF