Draw unbalanced grid of ggplot2 Plots in R

In this article, we are going to see how to Draw an Unbalanced Grid of ggplot2 plots in R Programming Language. We use mainly grid.arrange() and arrangeGrob() functions. Following is the brief information of these two functions. 

• grid.arrange(): It will calculate a different number of rows and columns to organize the plots and for forming nested layouts of plots

Syntax : grid.arrange(plot, nrow, ncol)

Parameter:

• plot: ggplot2 plot which we want to arrange
• nrow: Number of rows
• ncol: Number of columns

Return: Arrangement of plot

• arrangeGrob(): It function takes many parameters eg. layout_matrix, as.table, respect, top, bottom, left, right, and many more.

Syntax : arrangeGrob(…, heights)

Parameter:

• … : ggplot2 plots which we want to arrange on same page
• heights : Heights of those plots

Return: Arrange Multiple Plots on Same Page with different grid

First, we should load some R packages that are going to be in use further in this example. We will use ggplot2 package for draw plot and gridExtra for drawing various grids on plot.

install.packages("ggplot2")
library("ggplot2")
install.packages("gridExtra")
library("gridExtra")

For the Data of Scatter Plot, we pick some 50 random values for the X and Y axis both using rnorm() function which can generate random normal values. Then create a DataFrame and assign it to the ‘data’ data object.

data <- data.frame(x = rnorm(50), y = rnorm(50))

Now we create a Scatter Plot using ggplot() function and make it Scattered using geom_point() function.

Below is the implementation:

R

# Load Packages
library("ggplot2")
library("gridExtra")

# Create data for Plotting
data <- data.frame(x = rnorm(50), y = rnorm(50))

# Create Simple Scatter Plot
gplot <- ggplot(data, aes(x, y)) + geom_point(col = "green")
gplot

Output:

Simple Scatter Plot

Change Vertical Grid of plot

Here we are changing the vertical grid of the plot. We use grid.arrange() function of gridExtra package. Inside grid.arrange() package, we assign gplot (Our Scatter Plot) as a first parameter and ncol as the second parameter. ncol is just divided the plot area vertical in the value which we assign to ncol. Here we change the plot’s vertical grid to half of the plot space using “ncol = 2”.

Code:

R

# Load Packages
library("ggplot2")
library("gridExtra")

# Create Data for plotting
data <- data.frame(x = rnorm(50), y = rnorm(50))

# Create ggplot2 Scatter Plot
gplot <- ggplot(data, aes(x, y)) + geom_point(col = "green")

# Change the plot grid to half of the plot space.
grid.arrange(gplot, ncol = 2)

Output:

Scatter Plot with half vertical grid

Change Horizontal Grid of Plot

For changing the Horizontal grid of the plot, we simply replace ncol to nrow parameter of grid.arrange() function. nrow parameter takes the value in which we want to divide the plot horizontally. Here we change the plot’s Horizontal grid to half of the plot space using “nrow = 2”.

Code:

R

# Load Packages
library("ggplot2")
library("gridExtra")

# Create Data for plotting
data <- data.frame(x = rnorm(50), y = rnorm(50))

# Create ggplot2 Scatter Plot
gplot <- ggplot(data, aes(x, y)) + geom_point(col = "green")

# Change the plot Horizontal grid to half of the plot space.
grid.arrange(gplot, nrow = 2)

Output:

Scatter Plot with half horizontal grid

Draw Multiple plots of different grid size on the same Plot Space

For Draw Multiple plots of different sizes, we use arrangeGrob() as the second parameter of grid.arrange() function which will take the height and width of each graph as an argument. First, it will take gplot as an argument for each graph we want to create, and then it also specifies the heights of each plot using the heights parameter. 

Code:

R

# Load Packages
library("ggplot2")
library("gridExtra")

# Create Data for plotting
data <- data.frame(x = rnorm(50), y = rnorm(50))

# Create ggplot2 Scatter Plot
gplot <- ggplot(data, aes(x, y)) +
  geom_point(col = "green")

# Create Multiple Plots with different size of grids.
grid.arrange(gplot,
             arrangeGrob(gplot, gplot,
                         heights = c(0.4, 0.6)),
             ncol = 2)

Output:

Multiple Plots of different size of grids