How to plot overlap clustering with a list of vertices of each group and the edge list by R? - r

Question

I have a file csv including edge list of graph. After implementing CONGA
(Clustering Overlap Girven-NewMan alorithm), result is a list of vertices of each group.
I don't know how to plot it so that each group has different color in graph by R.
I can plot graph with edge list in R, but I don't know how to mark vertices in each group.
Input: edge list file and list of vertices in each group.
Output: graph with different color for each group.
output nearly like this
My English isn't good. Thanks for your support.

Answer 1

You plot colors using $color of vertices. Try to assign a color like V(g)$color <- 'green'.
It is better if you give us some code.
You say you get a list of your group-members. Convert the list to a vector, and sign a new color to each unique group-member value. I wrote this example code. I think it shows what you're after.
library(igraph)
get_a_random_network <- function() {
# EN: Function to get some random data to use as an example
g <- erdos.renyi.game(100, 60, type="gnm", directed=F, loops=FALSE)
g <- g %>% delete_vertices( V(g)[degree(g)==0] )
(g)
}
# Get sample data
g <- get_a_random_network()
# Use a cluster algorythm to determine groups. You said you had a list. I use this to generate example data.
groups <- cluster_fast_greedy(g)
# Look at the vertecies
(V(g))
# Look at what groups they belong to:
(groups$membership)
# Here you write that you have "list of vertices of each group". You don't
# give us code, but I assume that you have data that looks like this:
CONGA_list <- lapply(1:max(groups$membership),function(x) V(g)[groups$membership ==x])
(CONGA_list)
# This is where you should really have provided a code example.
# You could convert a list like CONGA_list to a vector like this:
membership_groups <- rep(0, length(V(g)))
for(x in 1:length(CONGA_list)){
membership_groups[as.vector(CONGA_list[[x]])] <- x
}
(membership_groups == groups$membership)
# You give color to your network by first telling each vector which group it belongs to
V(g)$membership <- groups$membership
# Then we asign a color. I use a vector of R-colors which I get like this...
colors = grDevices::colors()[grep('gr(a|e)y', grDevices::colors(), invert = T)]
# ... and then I sample from them to give each vertecy a color.
colors <- sample(colors, max(V(g)$membership))
V(g)$color <- colors[V(g)$membership]
(V(g)$color)
# The plot will work with the colors in V(g)$color
plot(g, vertex.size=7, vertex.label=NA)
Good luck