Skip to content
Related Articles

Related Articles

Plot Dendrogram with R and ggraph

Improve Article
Save Article
  • Last Updated : 22 Feb, 2022
Improve Article
Save Article

In this article, we are going to see how to customize dendrogram. It is also known as a tree diagram and it is a visual representation of the hierarchical relationship between items. A dendrogram’s main purpose is to figure out the best approach to assign objects to clusters. 

Creating Basic Dendrogram

To create a dendrogram we will igraph package which is used for creating and manipulating graphs and analyzing networks graph. So for creating a dendrogram, we will create a dataframe into the hierarchical data structure and then transform it into an edge list. To plot a Dendrogram we will use the following syntax:

ggraph(dataset, layout)

Parameters:

  • Dataframe: Hierarchical dataset/vector
  • layout: Dimension of layout
  • Circular: True/False

Syntax to install and import igraph package in the R working console:

install.package(‘igraph’)

library(igraph)

There are some optional functions to create edge into diagonal we will use geom_edge_diagonal() methods and to add nodes into a graph we will use geom_node_point() methods.

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
 
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# Now we can plot that
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram', circular = FALSE) +
  geom_edge_diagonal() +
  geom_node_point() +
  theme_void()


 

 

Output:

 

Circular Layout Dendrogram

 

To create a circular layout we will use circular = True, which will plot it into the circular layout.

 

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# plot
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram', circular = TRUE) +
  geom_edge_diagonal()+
  geom_edge_diagonal() +
  geom_node_point() +
  theme_void()


 

 

Output:

 

Edge style Dendrogram

 

To make the straight edge dendrogram we will use geom_edge_diagonal() methods.

 

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
 
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# Now we can plot that
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram') +
  geom_edge_link()+
  geom_node_point() +
  theme_void()


 

 

Output:

 

Labels in Dendrogram

 

To add the labels into the dendrogram we will use geom_node_text() which will add the labels into the graph into a hierarchical structure

 

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
 
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# Now we can plot that
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram') +
  geom_edge_diagonal() +
  geom_node_text(aes( label=name)) +
  theme_void()


 

 

Output:

 

Customize Labels in Dendrogram

 

To Customize Labels in Dendrogram we use geom_node_text() which will add the labels along with color into the graph into the hierarchical structure

 

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
 
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# Now we can plot that
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram') +
  geom_edge_diagonal() +
  geom_node_text(aes( label=name, color="red")) +
  theme_void()


 

 

Output:

 

Customize Nodes in Dendrogram

 

To Customize Labels in Dendrogram we use geom_node_point() which will add the nodes along with color into the graph into a hierarchical structure

 

R




# libraries
library(ggraph)
library(igraph)
library(tidyverse)
  
# create a data frame
data <- data.frame(
  level1="CEO",
  level2=c( rep("boss1",4), rep("boss2",4)),
  level3=paste0("mister_", letters[1:8])
)
  
# transform it to a edge list!
edges_level1_2 <- data %>% select(
  level1, level2) %>% unique %>% rename(
  from=level1, to=level2)
 
edges_level2_3 <- data %>% select(
  level2, level3) %>% unique %>% rename(
  from=level2, to=level3)
 
edge_list=rbind(edges_level1_2, edges_level2_3)
  
# Now we can plot that
graph_Data <- graph_from_data_frame( edge_list )
ggraph(graph_Data, layout = 'dendrogram') +
  geom_edge_diagonal() +
  geom_node_text(aes( label=name, color="red")) +
  geom_node_point(aes(color ="Blue"))+
theme_void()


 

 

Output:

 

 


My Personal Notes arrow_drop_up
Related Articles

Start Your Coding Journey Now!