Visualization with ggplot2

(A few remarks and tips before the practical session)

R is the best technology for doing computational science

(Subjectively.)

ggplot2 is the most powerful visualization framework

(Objectively. 🙂)

ggplot2 is a core tidyverse package

“Grammar of Graphics”

A formal syntax and grammar for describing visualizations

What does this mean?

Let’s consider base R plotting


library(palmerpenguins)

head(penguins, 7)


species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g sex year
Adelie Torgersen 39.1 18.7 181 3750 male 2007
Adelie Torgersen 39.5 17.4 186 3800 female 2007
Adelie Torgersen 40.3 18.0 195 3250 female 2007
Adelie Torgersen NA NA NA NA NA 2007
Adelie Torgersen 36.7 19.3 193 3450 female 2007
Adelie Torgersen 39.3 20.6 190 3650 male 2007
Adelie Torgersen 38.9 17.8 181 3625 female 2007

A base R histogram

hist(penguins$body_mass_g)

A base R histogram

hist(penguins$body_mass_g, xlim = c(2000, 8000))

A base R histogram

hist(penguins$body_mass_g, xlim = c(2000, 8000),
     main = "Histogram of body mass of penguins")

A base R histogram

hist(penguins$body_mass_g, xlim = c(2000, 8000),
     main = "Histogram of body mass of penguins",
     xlab = "Body mass [grams]")

A base R histogram

hist(penguins$body_mass_g, xlim = c(2000, 8000),
     main = "Histogram of body mass of penguins",
     xlab = "Body mass [grams]",
     col = "black", border = "white")

A base R histogram

Code 
par(mfrow = c(1, 3))

species_1 <- filter(penguins, species ==  "Adelie")
hist(species_1$body_mass_g, xlim = c(2000, 8000),
     main = "Species 'Adelie'",
     xlab = "Body mass [grams]",
     col = "darkgreen", border = "white")

species_2 <- filter(penguins, species ==  "Chinstrap")
hist(species_2$body_mass_g, xlim = c(2000, 8000),
     main = "Species 'Chinstrap'",
     xlab = "Body mass [grams]",
     col = "darkblue", border = "white")

species_3 <- filter(penguins, species ==  "Gentoo")
hist(species_3$body_mass_g, xlim = c(2000, 8000),
     main = "Species 'Gentoo'",
     xlab = "Body mass [grams]",
     col = "darkorange", border = "white")

Base R plots use a single function and specify a (potentially large) number of parameters which change their “aesthetic” properties and visual elements.

Creating a new figure (even from the same data) usually requires a completely different command.

tidyverse provides “grammar for data manipulation”…

A “mini-language” with verbs like select(), filter(), etc.

(Helps us avoid “R bootcamp”-level nightmares.)

ggplot2 provides “grammar for visualizations”…

(To make visualizations easier in a similar sense.)

Layers in “Grammar of Graphics”

Layers in “Grammar of Graphics”

  • data — our data frame
  • aesthetics — “mapping” of columns to visual properties of a figure (x, or y axes, color, shape, etc.)
  • geoms — graphical elements to be plotted (histograms, points, lines, etc.).
  • many other customizations


A tiny example

A quick advertisement for the “layering” concept of ggplot2

What data to visualize?

ggplot(penguins)

How to visualize it? (x “aesthetic”)

ggplot(penguins, aes(x = body_mass_g))

How to visualize it? (x “aesthetic”)

ggplot(penguins, aes(x = body_mass_g)) +

On what dimensions?

ggplot(penguins, aes(x = body_mass_g)) + # the + is a bit like %>%
  xlim(2000, 8000)

What to visualize? (“geom”)

ggplot(penguins, aes(x = body_mass_g)) +
  xlim(2000, 8000) +
  geom_histogram()

How to visualize it? (fill “aesthetic”)

ggplot(penguins, aes(x = body_mass_g, fill = species)) +
  xlim(2000, 8000) +
  geom_histogram()

How to visualize it? (adding “facets”)

ggplot(penguins, aes(x = body_mass_g, fill = species)) +
  xlim(2000, 8000) +
  geom_histogram() +
  facet_wrap(~ species)

What to visualize? (“geom” again)

ggplot(penguins, aes(x = body_mass_g, fill = species)) +
  xlim(2000, 8000) +
  geom_density() +
  facet_wrap(~ species)

More “layers”…

ggplot(penguins, aes(x = body_mass_g, fill = species)) +
  xlim(2000, 8000) +
  geom_density() +
  facet_wrap(~ species) +
  theme_minimal()

Today we’re going to learn ggplot2 on our metadata and IBD data

A bit of catch-up work on IBDs from yesterday

We need to make a copy of an IBD processing pipeline.

Let’s get started!

  1. Go to www.bodkan.net/simgen
  2. Click on “Visualization with ggplot2” in the left panel
  3. “Cheatsheets and handouts” section in the left panel has a single-page version of these slides and the dplyr cheatsheet and ggplot2 cheatsheets
  4. Open your RStudio and start working!




ggplot2 version 4.0.0 was released just a few days ago. 🤞😬