GGPlot Colors Best Tricks You Will Love

Key ggplot2 R functions

This section presents the key ggplot2 R function for changing a plot color.

Set ggplot color manually:

• scale_fill_manual() for box plot, bar plot, violin plot, dot plot, etc
• scale_color_manual() or scale_colour_manual() for lines and points

Use colorbrewer palettes:

• scale_fill_brewer() for box plot, bar plot, violin plot, dot plot, etc
• scale_color_brewer() or scale_colour_brewer() for lines and points

Use grey color scales:

• scale_fill_grey() for box plot, bar plot, violin plot, dot plot, etc
• scale_colour_grey() or scale_colour_brewer() for points, lines, etc

Change the default ggplot gradient color:

• scale_color_gradient(), scale_fill_gradient() for sequential gradients between two colors
• scale_color_gradient2(), scale_fill_gradient2() for diverging gradients
• scale_color_gradientn(), scale_fill_gradientn() for gradient between n colors

Prerequisites

1. Load ggplot2 package and set the default theme:

library(ggplot2)
theme_set(
  theme_minimal() +
    theme(legend.position = "right")
  )

2. Initialize ggplots using the iris data set. Create a box plot (bp) and a scatter plot (sp) that we’ll customize in the next section:

# Box plot
bp <- ggplot(iris, aes(Species, Sepal.Length))

# Scatter plot
sp <- ggplot(iris, aes(Sepal.Length, Sepal.Width))

Specify a single color

Colors in R can be specified either by name (e.g.: “red”) or by hexadecimal color codes, such as “#FF1234”. You can find many examples of color names and codes at:

• Awesome List Of 657 R Color Names
• Awesome List of Hexadecimal Colors You Should Have

The following R script changes the fill color (in box plots) and points color (in scatter plots).

• Using hexadecimal color codes:

# Box plot
bp + geom_boxplot(fill = "#FFDB6D", color = "#C4961A")

# Scatter plot
sp + geom_point(color = "#00AFBB") 

• Using color names

# Box plot
bp + geom_boxplot(fill = "lightgray", color = "black")

# Scatter plot
sp + geom_point(color = "steelblue") 

Change colors by groups: ggplot default colors

You can set colors according to the levels of a grouping variable by:

• Mapping the argument color to the variable of interest. This will be applied to points, lines and texts
• Mapping the argument fill to the variable of interest. This will change the fill color of areas, such as in box plot, bar plot, histogram, density plots, etc.

In the following example, we’ll map the options color and fill to the grouping variable Species, for scatter plot and box plot, respectively.

Changes colors by groups using the levels of Species variable:

# Box plot
bp <- bp + geom_boxplot(aes(fill = Species)) +
  theme(legend.position = "top")
bp

# Scatter plot
sp <- sp + geom_point(aes(color = Species)) +
  theme(legend.position = "top")
sp

Set custom color palettes

It’s possible to set manually the color palettes by using the functions:

• scale_fill_manual() for box plot, bar plot, violin plot, dot plot, etc
• scale_color_manual() or scale_colour_manual() for lines and points

# Box plot
bp + scale_fill_manual(values = c("#00AFBB", "#E7B800", "#FC4E07"))

# Scatter plot
sp + scale_color_manual(values = c("#00AFBB", "#E7B800", "#FC4E07"))

You might also find interesting, the following list of colors:

custom.col <- c("#FFDB6D", "#C4961A", "#F4EDCA", 
                "#D16103", "#C3D7A4", "#52854C", "#4E84C4", "#293352")

Use a colorblind-friendly palette

When selecting a set of colors, it’s recommended to make sure that you choose color palettes that are robust to colorblindness. In the next sections, we’ll show you to check that your production figures are colorblind-friendly.

Here, we present two color-blind-friendly palettes, one with gray, and one with black (palettes source: http://jfly.iam.u-tokyo.ac.jp/color/).

# The palette with grey:
cbp1 <- c("#999999", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")

# The palette with black:
cbp2 <- c("#000000", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")

You can use these palettes as follow:

# To use for fills, add
bp + scale_fill_manual(values = cbp1)

# To use for line and point colors, add
sp + scale_colour_manual(values=cbp1)

Predefined ggplot color palettes

You can modify the default ggplot colors by using predefined color palettes available in different R packages. The most commonly used color scales, include:

• Viridis color scales [viridis package].
• Colorbrewer palettes [RColorBrewer package]
• Grey color palettes [ggplot2 package]
• Scientific journal color palettes [ggsci package]
• Wes Anderson color palettes [wesanderson package]

Learn more at: Top R Color Palettes to Know for Great Data Visualization

library(viridis)
# Gradient color
ggplot(iris, aes(Sepal.Length, Sepal.Width))+
  geom_point(aes(color = Sepal.Length)) +
  scale_color_viridis(option = "D")

# Discrete color. use the argument discrete = TRUE
ggplot(iris, aes(Sepal.Length, Sepal.Width))+
  geom_point(aes(color = Species)) +
  geom_smooth(aes(color = Species, fill = Species), method = "lm") + 
  scale_color_viridis(discrete = TRUE, option = "D")+
  scale_fill_viridis(discrete = TRUE) 

# Box plot
bp + scale_fill_brewer(palette = "Dark2")

# Scatter plot
sp + scale_color_brewer(palette = "Dark2")

library(RColorBrewer)
display.brewer.all(colorblindFriendly = TRUE)

# Box plot
bp + scale_fill_grey(start = 0.8, end = 0.2) 

# Scatter plot
sp + scale_color_grey(start = 0.8, end = 0.2) 

library("ggsci")
# Change area fill color using jco palette
bp + scale_fill_jco()


# Change point color 
sp + scale_color_jco()

library(wesanderson)
names(wes_palettes)

##  [1] "BottleRocket1"  "BottleRocket2"  "Rushmore1"      "Royal1"        
##  [5] "Royal2"         "Zissou1"        "Darjeeling1"    "Darjeeling2"   
##  [9] "Chevalier1"     "FantasticFox1"  "Moonrise1"      "Moonrise2"     
## [13] "Moonrise3"      "Cavalcanti1"    "GrandBudapest1" "GrandBudapest2"

library(wesanderson)
# Discrete color
bp + scale_fill_manual(values = wes_palette("GrandBudapest1", n = 3))

# Gradient color
pal <- wes_palette("Zissou1", 100, type = "continuous")
ggplot(heatmap, aes(x = X2, y = X1, fill = value)) +
  geom_tile() + 
  scale_fill_gradientn(colours = pal) + 
  scale_x_discrete(expand = c(0, 0)) +
  scale_y_discrete(expand = c(0, 0)) + 
  coord_equal() 

Gradient or continuous colors

sp2 <- ggplot(iris, aes(Sepal.Length, Sepal.Width))+
  geom_point(aes(color = Sepal.Length))

# Sequential color scheme. 
# Specify the colors for low and high ends of gradient
sp2 + scale_color_gradient(low = "blue", high = "red")

# Diverging color scheme
# Specify also the colour for mid point
mid <- mean(iris$Sepal.Length)
sp2 + scale_color_gradient2(midpoint = mid, low = "blue", mid = "white",
                            high = "red", space = "Lab" )

sp2 + scale_color_gradientn(colours = rainbow(5))

Conclusion

This article presents how to customize ggplot colors. The main points are summarized as follow.

• Create a basic ggplot. Map the color argument to a factor or grouping variable.

p <- ggplot(iris, aes(Sepal.Length, Sepal.Width))+
  geom_point(aes(color = Species))
p

• Set the color palette manually using a custom color scale:

p + scale_color_manual(values = c("#00AFBB", "#E7B800", "#FC4E07"))

• Use color blind-friendly palette:

cbp1 <- c("#999999", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")
p + scale_color_manual(values = cbp1)

• Use RColorBrewer palettes:

p + scale_color_brewer(palette = "Dark2")

• Use viridis color scales:

library(viridis)
p + scale_color_viridis(discrete = TRUE)