4.1 Installation and loading

ggdendro can be installed as follow:

install.packages("ggdendro")

ggdendro requires the package ggplot2. Make sure that ggplot2 is installed and loaded before using ggdendro.

Load ggdendro as follow:

library("ggplot2")
library("ggdendro")

4.2 Visualize dendrogram using ggdendrogram() function

The function ggdendrogram() creates dendrogram plot using ggplot2.

# Visualization using the default theme named theme_dendro()
ggdendrogram(hc)

# Rotate the plot and remove default theme
ggdendrogram(hc, rotate = TRUE, theme_dendro = FALSE)

4.3 Extract dendrogram plot data

The function dendro_data() can be used for extracting the data. It returns a list of data frames which can be extracted using the functions below:

• segment(): To extract the data for dendrogram line segments
• label(): To extract the labels

# Build dendrogram object from hclust results
dend <- as.dendrogram(hc)

# Extract the data (for rectangular lines)
# Type can be "rectangle" or "triangle"
dend_data <- dendro_data(dend, type = "rectangle")
# What contains dend_data
names(dend_data)

## [1] "segments"    "labels"      "leaf_labels" "class"

# Extract data for line segments
head(dend_data$segments)

##           x         y     xend      yend
## 1 19.771484 13.516242 8.867188 13.516242
## 2  8.867188 13.516242 8.867188  6.461866
## 3  8.867188  6.461866 4.125000  6.461866
## 4  4.125000  6.461866 4.125000  2.714554
## 5  4.125000  2.714554 2.500000  2.714554
## 6  2.500000  2.714554 2.500000  1.091092

# Extract data for labels
head(dend_data$labels)

##   x y          label
## 1 1 0        Alabama
## 2 2 0      Louisiana
## 3 3 0        Georgia
## 4 4 0      Tennessee
## 5 5 0 North Carolina
## 6 6 0    Mississippi

dend_data can be used to draw a customized dendrogram using ggplot2:

# Plot line segments and add labels
p <- ggplot(dend_data$segments) + 
  geom_segment(aes(x = x, y = y, xend = xend, yend = yend))+
  geom_text(data = dend_data$labels, aes(x, y, label = label),
            hjust = 1, angle = 90, size = 3)+
  ylim(-3, 15)
print(p)

5.1 Chaining

The chaining operator (%>%) turns x %>% f(y) into f(x, y) so you can use it to rewrite multiple operations such that they can be read from left-to-right, top-to-bottom. For instance, the results of the two R codes below are equivalent.

Standard R code for creating a dendrogram:

data <- scale(USArrests)
dist.res <- dist(data)
hc <- hclust(dist.res, method = "ward.D2")
dend <- as.dendrogram(hc)
plot(dend)

R code for creating a dendrogram using chaining operator:

dend <- USArrests[1:5,] %>% # data
        scale %>% # Scale the data
        dist %>% # calculate a distance matrix, 
        hclust(method = "ward.D2") %>% # Hierarchical clustering 
        as.dendrogram # Turn the object into a dendrogram.
plot(dend)

5.2 Installation and loading

Install the stable version as follow:

install.packages('dendextend')

Loading:

library(dendextend)

5.3 How to change a dendrogram

The function set() can be used to change the parameters with dendextend.

The format is:

set(object, what, value)

Possible values for the argument what include:

5.4 Create a simple dendrogram

# Create a dendrogram and plot it
dend <- USArrests[1:5,] %>%  scale %>% 
        dist %>% hclust %>% as.dendrogram

dend %>% plot

# Get the labels of the tree
labels(dend)

## [1] "Alaska"     "Arizona"    "California" "Alabama"    "Arkansas"

5.5 Change labels

This section describes how to change label names as well as the color and the size for labels.

# Change the labels, and then plot:
dend %>% set("labels", c("a", "b", "c", "d", "e")) %>% plot

# Change color and size for labels
dend %>% set("labels_col", c("green", "blue")) %>% # change color
  set("labels_cex", 2) %>% # Change size
  plot(main = "Change the color \nand size") # plot

# Color labels by specifying the number of cluster (k)
dend %>% set("labels_col", value = c("green", "blue"), k=2) %>% 
          plot(main = "Color labels \nper cluster")
abline(h = 2, lty = 2)

In the R code above, the value of color vectors are too short. Hence, it’s recycled.

5.6 Change the points of a dendrogram nodes/leaves

# Change the type, the color and the size of node points
# +++++++++++++++++++++++++++++
dend %>% set("nodes_pch", 19) %>%  # node point type
  set("nodes_cex", 2) %>%  # node point size
  set("nodes_col", "blue") %>% # node point color
  plot(main = "Node points")

# Change the type, the color and the size of leave points
# +++++++++++++++++++++++++++++
dend %>% set("leaves_pch", 19) %>%  # node point type
  set("leaves_cex", 2) %>%  # node point size
  set("leaves_col", "blue") %>% # node point color
  plot(main = "Leaves points")

# Specify different point types and colors for each leave
dend %>% set("leaves_pch", c(17, 18, 19)) %>%  # node point type
  set("leaves_cex", 2) %>%  # node point size
  set("leaves_col", c("blue", "red", "green")) %>% #node point color
  plot(main = "Leaves points")

5.7 Change the color of branches

The color for branches can be controlled using k-means clustering:

# Default colors
dend %>% set("branches_k_color", k = 2) %>% 
  plot(main = "Default colors")

# Customized colors
dend %>% set("branches_k_color", 
             value = c("red", "blue"), k = 2) %>% 
   plot(main = "Customized colors")

It’s also possible to use the function color_branches().

5.8 Adding colored rectangles

Clusters can be highlighted by adding colored rectangles. This is done using the rect.dendrogram() function (modeled based on the rect.hclust() function). One advantage of rect.dendrogram over rect.hclust, is that it also works on horizontally plotted trees:

# Vertical plot
dend %>% set("branches_k_color", k = 3) %>% plot
dend %>% rect.dendrogram(k=3, border = 8, lty = 5, lwd = 2)

# Horizontal plot
dend %>% set("branches_k_color", k = 3) %>% plot(horiz = TRUE)
dend %>% rect.dendrogram(k = 3, horiz = TRUE, border = 8, lty = 5, lwd = 2)

5.9 Adding colored bars

This is useful for annotating the items in the clusters:

grp <- c(1,1,1, 2,2)
k_3 <- cutree(dend,k = 3, order_clusters_as_data = FALSE) 
# The FALSE above makes sure we get the clusters in the order of the
# dendrogram, and not in that of the original data. It is like:
# cutree(dend, k = 3)[order.dendrogram(dend)]

the_bars <- cbind(grp, k_3)

dend %>% set("labels", "") %>% plot
colored_bars(colors = the_bars, dend = dend)

5.10 ggplot2 integration

The following 2 steps are used:

1. Transform a dendrogram into a ggdend object using as.ggdend() function
2. Make the plot using the function ggplot()

dend <- iris[1:30,-5] %>% scale %>% dist %>% 
   hclust %>% as.dendrogram %>%
   set("branches_k_color", k=3) %>% set("branches_lwd", 1.2) %>%
   set("labels_colors") %>% set("labels_cex", c(.9,1.2)) %>% 
   set("leaves_pch", 19) %>% set("leaves_col", c("blue", "red"))
# plot the dend in usual "base" plotting engine:
plot(dend)

Produce the same plot in ggplot2 using the function:

library(ggplot2)
# Rectangle dendrogram using ggplot2
ggd1 <- as.ggdend(dend)
ggplot(ggd1) 

# Change the theme to the default ggplot2 theme
ggplot(ggd1, horiz = TRUE, theme = NULL) 

# Theme minimal
ggplot(ggd1, theme = theme_minimal()) 

# Create a radial plot and remove labels
ggplot(ggd1, labels = FALSE) + 
  scale_y_reverse(expand = c(0.2, 0)) +
  coord_polar(theta="x")

5.11 pvclust and dendextend

The package dendextend can be used to enhance many packages including pvclust. Recall that, pvclust is for calculating p-values for hierarchical clustering.

pvclust can be used as follow:

library(pvclust)
data(lung) # 916 genes for 73 subjects
set.seed(1234)
result <- pvclust(lung[1:100, 1:10], method.dist="cor", 
                  method.hclust="average", nboot=10)

## Bootstrap (r = 0.5)... Done.
## Bootstrap (r = 0.6)... Done.
## Bootstrap (r = 0.7)... Done.
## Bootstrap (r = 0.8)... Done.
## Bootstrap (r = 0.9)... Done.
## Bootstrap (r = 1.0)... Done.
## Bootstrap (r = 1.1)... Done.
## Bootstrap (r = 1.2)... Done.
## Bootstrap (r = 1.3)... Done.
## Bootstrap (r = 1.4)... Done.

# Default plot of the result
plot(result)
pvrect(result)

# pvclust and dendextend
result %>% as.dendrogram %>% 
  set("branches_k_color", k = 2, value = c("purple", "orange")) %>%
  plot
result %>% text
result %>% pvrect