In christyray/crthemes: A Standardized ggplot2 Theme for Visualizations

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "figures/theme-",
  fig.width = 6,
  fig.asp = 2/3,
  out.width = "75%",
  dev = "ragg_png",
  dpi = 300
)

library(ggplot2)
library(crthemes)
library(palmerpenguins)
penguins <- tidyr::drop_na(penguins)

Base Theme

The core of the crthemes package is the theme_cr() function - a ggplot2 theme that can be applied to a ggplot object just like the default theme_grey() or theme_bw(). The theme_cr() function applies a full set of theme values to the plot, and those theme values can be modified with the theme() function if necessary.

The theme is applied by adding it to a ggplot object:

p <- ggplot(data = penguins, 
            aes(x = flipper_length_mm, y = bill_length_mm, color = species)) +
  geom_point(alpha = 0.5) +
  labs(
    x = "Flipper Length (mm)", 
    y = "Bill Length (mm)", 
    color = "Species",
    title = "Flipper and Bill Length",
    caption = "Data adapted from the palmerpenguins R package"
  ) +
  theme_cr()
p

Fonts

The default font used for the plot is "Roboto Regular", with bold text using "Roboto Medium". To see a list of included font families, use the font_names() function. The font_table() function prints a table of all of the included fonts and their properties. The font_preview() function generates a preview of a selected font:

font_names()

font_table()

font_preview("Roboto Light")

Any of the included fonts can be used with theme_cr() by passing their names to the font argument:

p + theme_cr(font = "Roboto Serif Regular")

Symbol Fonts

The ragg package uses the font named "symbol" for all symbols rendered in a plot. theme_cr() registers the provided font as the "symbol" font the first time it is called during an R session. To prevent this, use symbol = FALSE.

To remove a previously registered "symbol" font, use systemfonts::clear_registry() or restart the R session. The "symbol" font cannot be overwritten once it is set; to name a new "symbol" font, first remove the previously registered one, then run theme_cr() with font set to the font to be used for "symbol".

# Clear the previous symbol font
systemfonts::clear_registry()

p + labs(title = "Symbols Use Bold Font", 
         subtitle = bquote(alpha ~ beta ~ gamma ~ "are symbols")) +
  theme_cr(font = "Roboto Bold")

Base R Graphics Compatibility

By default, the packaged fonts are used with theme_cr(). If compatibility with base R graphics or cairographics is required, the argument cairo = TRUE can be used with font set to a system default font (or font = "" to use the system default sans serif font).

p + theme_cr(cairo = TRUE, font = "serif")

Colors

The pal data set contains several color palettes to use with the standard theme. The list of palette names can be printed with pal_names(), and pal_table() prints a table with the properties of all of the included palettes. The colors within a specific palette can be listed with pal_colors().

pal_names()

pal_table()

pal_colors("receptors")

Several of the palettes are Named, which means they have a list of molecule names matched to specific colors of the palette. The colors of a specific palette along with the color names and molecule names can be displayed with pal_preview():

# By default, pal_preview shows the color names
pal_preview(palette = "receptors")
# The label argument specifies what to use for the color labels
pal_preview(palette = "receptors", labels = "species")
# Specific species can be selected with the species argument
pal_preview(palette = "receptors", 
            species = c("IL6R", "IL8R", "IL6R-Ab", "IL8R-Ab"),
            labels = "species")
# The hex codes can also be displayed
pal_preview(palette = "highlight2", labels = "hex")
# Alpha transparency can be previewed
pal_preview(palette = "highlight2", alpha = 0.4)

Scale Functions

scale_fill_cr() and scale_color_cr() allow the custom palettes to be applied to a ggplot object. These functions are used like typical scale_*() functions.

They additionally accept the arguments:

palette: the palette to be applied
ncol: the range of colors to use from the palette
colors: the names of colors to use
species: the name-matched colors corresponding to particular molecules
alpha: the transparency value

# Apply a color palette
p + scale_color_cr(palette = "antibodies")
# Select specific colors from the palette, colors will be used in order given
p + scale_color_cr(palette = "antibodies", colors = c("blue", "orange", "green"))

When the data has includes species names from the project model, those names can be matched to color palettes with the Named trait:

set.seed(202)
mol <- c("IL6R", "IL8R", "IL6", "IL8")
df <- data.frame(x = rnorm(120), y = rnorm(120), il = as.factor(rep(mol,30)))
mol_p <- ggplot(data = df, aes(x, y, color = il)) +
  geom_point() +
  labs(x = "X Axis", y = "Y Axis", color = "Species", title = "Random Data") +
  theme_cr()

# Using the species names to call the colors
mol_p + scale_color_cr(palette = "receptors", species = levels(df$il))
# Alpha can also be applied directly in the scale_color_cr function
mol_p + scale_color_cr(palette = "receptors", species = levels(df$il), alpha = 0.5)

scale_fill_cr() works exactly the same as scale_color_cr():

library(dplyr)
# Using functions from dplyr to compute summary statistics
df <- df %>% group_by(il) %>% summarize(avg = mean(x))
mol_p_col <- ggplot(data = df, aes(x = il, y = avg, fill = il)) +
  geom_col() +
  labs(x = NULL, y = "Y Axis", title = "Random Data") +
  theme_cr()

# Using the species names to call the colors
mol_p_col + scale_fill_cr(palette = "receptors", species = levels(df$il), guide = NULL)