04-ggplot2

Author
Affiliation

Professor Shannon Ellis

UC San Diego
COGS 137 - Fall 2024

Published

October 15, 2024

Data Visualization with ggplot2

Q&A

Q: I’m curious about the differences between base r pipe (|>) and the %>% operators! I’ve only learned about %>% and it’d be interesting looking at the differences between using either one.
A: There’s a blog post for that!

Q: The most confusing part of this lecture was getting started with dplyr functions, but it will become easier with practice.
A: Lots of people said something similar. And, this is definitely the right attitude! Lecture is for first exposure. Practice in lecture is to start underestanding. Labs are for guided practice where you have more time to practice. Case studies and homeworks are where we check our understanding! So, you’re not supposed to “get it” all the first time you see it in lecture.

Course Announcements

Due Dates:

  • Lab 02 due Thursday (11:59 PM)
  • Lecture Participation survey “due” after class

Notes:

  • HW02 now available; Lab03 posted Th

Suggested Reading

ggplot2 \(\in\) tidyverse

  • ggplot2 is tidyverse’s data visualization package
  • Structure of the code for plots can be summarized as
ggplot(data = [dataset], 
       mapping = aes(x = [x-variable], 
                     y = [y-variable])) +
   geom_xxx() +
   other options

Data: CS01 WB

WB <- read_csv("https://github.com/ShanEllis/datasets/raw/refs/heads/master/Blood.csv")

The Data

WB |>
  datatable()

The Cleaning

Note: this is the code from the end of 03-dplyr notes, combined into a single chunk.

WB <- WB |> 
  mutate(Treatment = fct_recode(Treatment, 
                                "5.9% THC (low dose)" = "5.90%",
                                "13.4% THC (high dose)" = "13.40%"),
         Treatment = fct_relevel(Treatment, "Placebo", "5.9% THC (low dose)")) |> 
  janitor::clean_names() |>
  rename(thcoh = x11_oh_thc,
         thccooh = thc_cooh,
         thccooh_gluc = thc_cooh_gluc,
         thcv = thc_v) |>
  mutate(timepoint = case_when(time_from_start < 0 ~ "pre-smoking",
                               time_from_start > 0 & time_from_start <= 30 ~ "0-30 min",
                               time_from_start > 30 & time_from_start <= 70 ~ "31-70 min",
                               time_from_start > 70 & time_from_start <= 100 ~ "71-100 min",
                               time_from_start > 100 & time_from_start <= 180 ~ "101-180 min",
                               time_from_start > 180 & time_from_start <= 210 ~ "181-210 min",
                               time_from_start > 210 & time_from_start <= 240 ~ "211-240 min",
                               time_from_start > 240 & time_from_start <= 270 ~ "241-270 min",
                               time_from_start > 270 & time_from_start <= 300 ~ "271-300 min",
                               time_from_start > 300 ~ "301+ min"))

  Why are there two mutates? Could they have all been in a single mutate?

A Plot

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(., mapping = aes(x = thc, y = thccooh,
                       color = treatment)) +
    geom_point() +
    labs(title = "THC and THC-COOH levels (0-30 min)",
         subtitle = "THC levels remain low in placebo group; THC-COOH is variable",
         x = "THC (ng/mL)", y = "THC-COOH (ng/mL)",
         color = "Treatment Group") +
    scale_color_viridis_d()

Coding out loud

Start with the WB data frame (filtering to only include first timepoint)

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot()

Start with the WB data frame, map thc levels to the x-axis

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc))

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc, 
                       y = thccooh))

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc, 
                       y = thccooh)) + 
  geom_point()

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point and map treatment group to the color of each point.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc, 
                       y = thccooh,
                       color = treatment)) + 
  geom_point()

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point and map treatment group to the color of each point. Title and subtitle the plot.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc, 
                       y = thccooh,
                       color = treatment)) + 
  geom_point() +
  labs(title = "THC and THC-COOH levels (0-30 min)",
       subtitle = "THC levels remain low in placebo group; THC-COOH is variable")

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point and map treatment group to the color of each point. Title and subtitle the plot, label the x and y axes

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(mapping = aes(x = thc, 
                       y = thccooh,
                       color = treatment)) + 
  geom_point() +
  labs(title = "THC and THC-COOH levels (0-30 min)",
       subtitle = "THC levels remain low in placebo group; THC-COOH is variable",
       x = "THC (ng/mL)", y = "THC-COOH (ng/mL)")

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point and map treatment group to the color of each point. Title and subtitle the plot, label the x and y axes, and title the legend.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(., mapping = aes(x = thc, y = thccooh,
                       color = treatment)) +
    geom_point() +
    labs(title = "THC and THC-COOH levels (0-30 min)",
         subtitle = "THC levels remain low in placebo group; THC-COOH is variable",
         x = "THC (ng/mL)", y = "THC-COOH (ng/mL)",
         color = "Treatment Group")

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis. Represent each observation with a point and map treatment group to the color of each point. Title and subtitle the plot, label the x and y axes, and title the legend. Finally, use a discrete color scale that is designed to be perceived by viewers with common forms of color blindness.

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(., mapping = aes(x = thc, y = thccooh,
                       color = treatment)) +
    geom_point() +
    labs(title = "THC and THC-COOH levels (0-30 min)",
         subtitle = "THC levels remain low in placebo group; THC-COOH is variable",
         x = "THC (ng/mL)", y = "THC-COOH (ng/mL)",
         color = "Treatment Group") +
    scale_color_viridis_d()

Coding out loud

WB |> 
  filter(timepoint=="0-30 min") |>
  ggplot(., mapping = aes(x = thc, y = thccooh,
                       color = treatment)) +
    geom_point() +
    labs(title = "THC and THC-COOH levels (0-30 min)",
         subtitle = "THC levels remain low in placebo group; THC-COOH is variable",
         x = "THC (ng/mL)", y = "THC-COOH (ng/mL)",
         color = "Treatment Group") +
    scale_color_viridis_d()

Start with the WB data frame, map thc levels to the x-axis and map thccooh levels to the y-axis.

Represent each observation with a point and map treatment group to the color of each point.

Title and subtitle the plot, label the x and y axes, and title the legend.

Finally, use a discrete color scale that is designed to be perceived by viewers with common forms of color blindness.

Argument names

Tip

You can omit the names of first two arguments when building plots with ggplot().

ggplot(data = WB, 
       mapping = aes(x = thc, y = cbn)) +
  geom_point()
ggplot(WB, 
       aes(x = thc, y = cbn)) +
  geom_point()

Your Turn

Generate a basic plot in ggplot2 using different filtering and/or variables than those in the last example (last example: thc & thccoooh, “0-30 min” timepoint).

Put a green sticky on the front of your computer when you’re done. Put a pink if you want help/have a question.

Aesthetics

Aesthetics options

Commonly used characteristics of plotting characters that can be mapped to a specific variable in the data are

  • color
  • shape
  • size
  • alpha (transparency)

Color

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color = treatment)) + 
  geom_point() +
  scale_color_viridis_d()

Shape

Mapped to a different variable than treatment

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color = treatment,
                                shape = group)) + 
  geom_point() +
  scale_color_viridis_d()

Shape

Mapped to same variable as color

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color = treatment,
                                shape = treatment)) + 
  geom_point() +
  scale_color_viridis_d()

Size

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color = treatment,
                                shape = treatment,
                                size = time_from_start)) + 
  geom_point() +
  scale_color_viridis_d()

Alpha

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color = treatment,
                                shape = treatment,
                                size = time_from_start,
                                alpha = cbd)) + 
  geom_point() +
  scale_color_viridis_d()

Mapping vs. setting

  • Mapping: Determine the size, alpha, etc. of points based on the values of a variable in the data
    • goes into aes()
  • Setting: Determine the size, alpha, etc. of points not based on the values of a variable in the data
    • goes into geom_*() (this was geom_point() in the previous example, but we’ll learn about other geoms soon!)

Mapping vs. Setting (example)

Mapping

ggplot(data = WB, 
       mapping = aes(x = thc, y = cbn,
                     size = time_from_start,
                     alpha = cbd)) + 
  geom_point() +
  scale_color_viridis_d()

Setting

ggplot(data = WB, 
       mapping = aes(x = thc, y = cbn)) + 
  geom_point(size = 2, alpha = 0.5) +
  scale_color_viridis_d()

Your Turn

Edit the basic plot you created earlier to change something about its aesthetics.

Put a green sticky on the front of your computer when you’re done. Put a pink if you want help/have a question.

Faceting

Faceting

  • Smaller plots that display different subsets of the data
  • Useful for exploring conditional relationships and large data

facet_grid

2d grid; `rows ~ cols

ggplot(data = WB, mapping = aes(x = thc, y = cbn)) + 
  geom_point() +
  facet_grid(treatment ~ group)

facet_wrap

“1d ribbon wrapped according to number of rows and columns specified or available plotting area”

ggplot(data = WB, mapping = aes(x = thc, y = cbn)) + 
  geom_point() +
  facet_wrap(~ timepoint)

Facet and color

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color=treatment)) + 
  geom_point() +
  facet_grid(treatment ~ group) +
  scale_color_viridis_d()

Face and color, no legend

ggplot(data = WB, mapping = aes(x = thc, y = cbn,
                                color=treatment)) + 
  geom_point() +
  facet_grid(treatment ~ group) +
  scale_color_viridis_d() +
  guides(color="none")

geoms

Common geoms

geom 1 Description 2
geom_point scatterplot
geom_bar barplot
geom_line line plot
geom_density densityplot
geom_histogram histogram
geom_boxplot boxplot

Your Turn

Generate a plot in ggplot2 using a different geom than what you did previously. Customize as much as you can before time is “up.”

Put a green sticky on the front of your computer when you’re done. Put a pink if you want help/have a question.

Recap

  • Can I explain the overall structure of a call to generate a plot in ggplot2?
  • Can I describe ggplot2 code? Can I create plots using ggplot2?
  • Can I explain the difference between mapping and setting?
  • Can I alter the aesthetics of a basic plot? (color, shape, size, transparency)
  • Am I able to facet a plot to generate a grid of figures
  • Can I describe what a geom is and do I know the basic plots available?

Footnotes

  1. ggplot2 geoms listed here↩︎

  2. When each visualization is appropriate here↩︎