CODEWITHSUNDEEP
rggplot2igraph
Biochimiste
Biochimiste

Reputation: 15

Change the representation of the graph

i would like to have help with my graph. I would like to be able to change the graph representation algorithm (LGL or other) but I can't. How can I do it? have tried several options that do not work...

library(NetworkToolbox)
library(dplyr)
library(igraph)
library(ggplot2)
library(ggnetwork)

M1 <- as_tibble(replicate(21,sample(1:3,100,rep=TRUE)))
colnames(M1) <- c("1st", "2nd", "3th", "4th", "5th", "6th","7th","8th","9th","10th",
                  "11th","12th","13th","14th","15th","16th","17th","18th","19th",
                  "20th","21th")

M2 <- as.matrix(round(cor(M1[,],method ="kendall"),2))

gr4ph <-  graph.adjacency(M2, mode = "undirected",weight=TRUE)
MAST <- MaST(M2, normal = False)
gr4ph <-  graph.adjacency(MAST , mode = "lower",weight=TRUE) 
ggplot(gr4ph, aes(x = x, y = y, xend = xend, yend = yend)) +
  geom_edges(color = "grey", alpha = 1) +
  geom_nodes(aes(color = name)) +  theme_blank() +
  geom_nodetext(aes(label = name), color = "black") +
  geom_edgetext(aes(label = weight))+
  theme(legend.position = "none")

Upvotes: 0

Views: 76

Answers (1)

Biochimiste
Biochimiste

Reputation: 15

You can use two solution. The layout argument is the key of the graph's representation : The following exemple are coded with "Large graph layout" With ggplot :

ggplot(gr4ph1, aes(x = x, y = y, xend = xend, yend = yend),layout = layout_with_lgl(gr4ph1))+
  geom_edges(color = "grey", alpha = 1,size=1.5,) +
  geom_nodes(aes(color = name), size = 3) +  theme_blank() +
  geom_nodetext(aes(label = name), color = "black", size = 3) +
  geom_edgetext(aes(label = weight), size = 3,label.padding=unit(0.01, "lines"))+
  theme(legend.position = "none")

With ggraph :

ggraph(gr4ph, layout = "lgl") +
  geom_edge_link(aes(label = weight), angle_calc = 'along',label_dodge = unit(2.5, 'mm')) +
  geom_node_point(aes(size=3,color= name))  +
  theme(legend.position = "none")+
  geom_node_text(aes(label = name), repel = TRUE)

Upvotes: 1

Related Questions