ggplot2

Malie Lessard-Therrien and Etienne Low-Décarie
May 8, 2017

What is a graph?

  • Statistics and design combined
  • Explore and explain
  • Communicating results

ggplot2

Beautiful and flexible

Pretty ggplots

Outline (ggplot2)

Day I

Your first ggplot plot

  • Basic scatter plot
  • Exercise 1

Grammar of graphics

  • More advanced plots
  • Aesthetics, Geometrics
  • Exercise 2

Facets

  • Exercise 3

Saving your graphs

Day II

Pretty graphs for presentation

  • Fine tunning
  • Package cowplot
  • Exercise 4

Maps

Install/load ggplot2

if(!require(ggplot2)){install.packages("ggplot2")}
require(ggplot2)

Iris dataset

  • Edgar Anderson
  • RA Fischer

Iris flowers

head (iris)

  Sepal.Length Sepal.Width Petal.Length Petal.Width Species
1          5.1         3.5          1.4         0.2  setosa
2          4.9         3.0          1.4         0.2  setosa
3          4.7         3.2          1.3         0.2  setosa
4          4.6         3.1          1.5         0.2  setosa
5          5.0         3.6          1.4         0.2  setosa
6          5.4         3.9          1.7         0.4  setosa

What to use when

Explore your data using function “qplot”

  • For you and your colleagues
  • Quick
  • Default settings can't be changed
  • like the function “plot” in R-base

Explain a pattern (or lack of) in your data using function “ggplot”

  • For presentation or publication
  • All settings must be specified
  • Flexible (you have the control over all settings)

Your first graph with ggplot2

A basic scatter plot

qplot(data=iris,
      x=Sepal.Length,
      y=Sepal.Width)

plot of chunk unnamed-chunk-5

Change axis names 

qplot(data=iris,
      x=Sepal.Length,
      y=Sepal.Width,
      xlab="Sepal length (mm)",
      ylab="Sepal width (mm)")

plot of chunk unnamed-chunk-6

Exercice 1

produce a basic plot with built in data

data()
CO2
?CO2

WARNING: THERE ARE MULTIPLE CO2/co2 datasets (CASE SENSITIVE, use capitals)

Grammar of graphics (gg)

2 principles:

  • distinc layers of graphical elements
  • meaningful plots using aesthetic mapping

layers of a plot

Graph elements

3 essentials:

  • Data (what you collected, in the right format)
  • Aesthetics (aes)
  • Geometries (geom)

4 optionals:

  • Facets
  • Coordinates
  • Themes
  • Statistics

layers of a plot

Aesthetics

set with the aes() function

  • color (“outside” color)
  • fill (“inside” color)
  • shape (of points)
  • linetype
  • size (of points or line)
  • position (i.e., on the x and y axes)
  • group (that a point belongs to)
  • alpha (transparency of the point)

Geometic Objects

set with the “geom_…” command

Ex:

  • points (geom_point, for scatter plots, dot plots, etc)
  • lines (geom_line, for time series, trend lines, etc)
  • boxplot (geom_boxplot, for, well, boxplots!)
  • violins (geom_violin, region inside the violin contains all of the observed data)

A plot must have at least one geom; there is no upper limit. You can add a geom to a plot using the + operator

Available elements

http://ggplot2.tidyverse.org/reference/

How it works

  • 1. create a simple plot object
plot.object<-qplot() # to explore 
or
plot.object<-ggplot() # to explain
  • 2. add graphical layers/complexity
plot.object<-plot.object+layer()
  • 3. repeat step 2 until satisfied
  • 4. print your object to screen (or to graphical device)
print(plot.object)

Scatter plot as an R object with qplot

basic_graph<-qplot(data=iris,
                  x=Sepal.Length,
                  y=Sepal.Width,
                  xlab="Sepal length (mm)",
                  ylab="Sepal width (mm)")

print(basic_graph)

plot of chunk unnamed-chunk-11

Scatter plot as an R object with ggplot

Note: aes() and geom_point()

basic_graph <- ggplot(data=iris,
                    aes(x=Sepal.Length,
                        y=Sepal.Width))+
                      xlab("Sepal length (mm)")+
                      ylab("Sepal width (mm)")+
                      geom_point()

print (basic_graph)

plot of chunk unnamed-chunk-12

Geometric example

Scatter plot with colour and shape

basic_graph <- basic_graph+
              aes(colour=Species,
                  shape=Species)

print(basic_graph)

plot of chunk unnamed-chunk-13

Geometric example

Scatter plot with linear regression

Add a geom (eg. linear smooth)

basic_graph_lm <- basic_graph+
  geom_smooth(method="lm", se=F)

print(basic_graph_lm)

plot of chunk unnamed-chunk-14

Exercise 2

Produce a colorful plot containing linear regressions with built in data

CO2
?CO2
data()

WARNING: THERE ARE MULTIPLE CO2/co2 datasets (CASE SENSITIVE, use capitals)

Geometric example

Categorical x-axis

basic_graph_cat <- ggplot(data=iris,
                          aes(x=Species,
                              y=Sepal.Width))+
                              xlab("Species")+
                              ylab("Sepal width (mm)")+
  geom_point()

print (basic_graph_cat)

.
.
. 

basic_graph_cat_bx <- basic_graph_cat+
  geom_boxplot ()

print (basic_graph_cat_bx)

plot of chunk unnamed-chunk-17

plot of chunk unnamed-chunk-18

Geometric example

Categorical x-axis, geom_violin

Violin fruit type

Facets

Divide the data graphically

facet_grid(rows~columns)

iris_facets <- basic_graph + 
              facet_grid(. ~ Species)

print (iris_facets)

plot of chunk unnamed-chunk-20

Exercise 3

Create a colorful graph with the data you have used using geoms and facets

CO2
?CO2
data()

Facets

Using CO2 data

head (CO2)

  Plant   Type  Treatment conc uptake
1   Qn1 Quebec nonchilled   95   16.0
2   Qn1 Quebec nonchilled  175   30.4
3   Qn1 Quebec nonchilled  250   34.8
4   Qn1 Quebec nonchilled  350   37.2
5   Qn1 Quebec nonchilled  500   35.3
6   Qn1 Quebec nonchilled  675   39.2

Facets

CO2_graph_facets<-ggplot(data=CO2,
                aes(x=conc,
                    y=uptake,
                    colour=Treatment))+
  geom_point()+
  facet_grid(.~Type)

print(CO2_graph_facets)

plot of chunk unnamed-chunk-22

Groups

Problems when adding the geom_line

print(CO2_graph_facets+geom_line())

plot of chunk unnamed-chunk-23

Groups

Solution: specify groups

CO2_graph_fg<-CO2_graph_facets+geom_line(aes(group=Plant))

print(CO2_graph_fg)

plot of chunk unnamed-chunk-24

Help

Saving plots

pdf("./Plots/todays_plots.pdf")
  print(basic_graph)
  print(basic_graph_lm)
  print(basic_graph_cat_bx)
  print(iris_facets)
  print(CO2_graph_facets)
graphics.off()

all other R-base save functions available:
bmp(), jpeg(), etc

Saving plots

ggsave: saves last plot and guesses format from file name

ggsave("./Plots/todays_plots.jpeg", iris_facets)

Day 2

Pretty and meaningful graphs

  • Fine tunning (color, coordinates, themes)
  • Exercise 4
  • Package cowplot, fortify
  • Exercise 5

Maps

Fine tunning

Give new color to each iris species

Step 1: Create the reference graph

iris_color_graph <- ggplot (data=iris, 
                            aes (x= Sepal.Length,
                                 y=Sepal.Width, 
                                 color=Species))+
  xlab("Sepal lenght (mm)")+
  ylab("Sepal width (mm)")+
  geom_point(size = 3)+
  geom_smooth(method="lm", se=F)

print (iris_color_graph)

plot of chunk unnamed-chunk-27

Fine tunning

Give new color to each iris species

Step 2: Change color manually, using base color

iris_color_graph_manual <- iris_color_graph +
  scale_color_manual(values=c("setosa"="red",
                              "versicolor"="pink",
                              "virginica"="purple"), 
                     labels=c("Setosa",
                              "Versicolor", 
                              "Virginica"))

print (iris_color_graph_manual)

plot of chunk unnamed-chunk-28

Fine tunning

Give new color to each iris species

Step 2: Change color manually, using hexadecimal codes

iris_color_graph_manual_hexa <- iris_color_graph +
  scale_color_manual(values=c("setosa"="#6600CC",
                              "versicolor"="#990099",
                              "virginica"="#FF3399"), 
                     labels=c("Setosa",
                              "Versicolor", 
                              "Virginica"))

print (iris_color_graph_manual_hexa)

plot of chunk unnamed-chunk-29

Color options

http://www.cookbook-r.com/Graphs/Colors_(ggplot2)/

Exercise 4

Change the shape of data points manually with the dataset you have used and/or your own data

see: http://www.cookbook-r.com/Graphs/Shapes_and_line_types/

CO2
?CO2
data()

Coordinates

DF <- data.frame(variable = LETTERS[1:10], value = sample(10, replace = TRUE))

bar_graph <- ggplot(data=DF, 
                aes(x=variable,
                fill=variable,
                y=value))+
                geom_bar(stat="identity")

print(bar_graph)

plot of chunk unnamed-chunk-30

Coordinates

Change looks of graph by changing coordinates

polar_graph <- bar_graph +
  coord_polar()

print(polar_graph)

plot of chunk unnamed-chunk-31

Themes

Change font size

iris_classic_fs <- iris_color_graph_manual_hexa +
  theme(axis.title.x=element_text(size=30),
        axis.text.x=element_text(size=25))+
  theme(axis.title.y=element_text(size=30),
        axis.text.y=element_text(size=25))
print (iris_classic_fs)

plot of chunk unnamed-chunk-32

Themes

Set B&W using theme_classic ()

iris_classic <- iris_color_graph_manual_hexa+
           theme_classic(base_size = 20)

print (iris_classic)

plot of chunk unnamed-chunk-34

Graph with cowplot

  • simple add-on to ggplot2
  • provide a publication-ready theme for ggplot2
  • minimum amount of fiddling with sizes of axis labels, plot backgrounds, etc

see https://cran.r-project.org/web/packages/cowplot/vignettes/introduction.html

if(!require(cowplot)){install.packages("cowplot")}
require(cowplot)

Cowplot, B&W graph

With cowplot, your graph background is B&W by default

iris_cowplot <- iris_color_graph_manual_hexa

print (iris_cowplot)

plot of chunk unnamed-chunk-37

Exercise 5

  1. Create a B&W background graph with the dataset you have used and/or your own data

  2. Change the font of the axis titles, see http://www.cookbook-r.com/Graphs/Fonts/

CO2
?CO2
data()

Cowplot, multiple graphs

Label and align multiple graphs

fig.1 <- plot_grid(iris_color_graph_manual, iris_cowplot,
                   labels = c("a)", "b)"),
                   nrow = 2,
                   align = "v")
print (fig.1)

plot of chunk unnamed-chunk-39

Cowplot, multiple graphs

Label and align multiple graphs

fig.1 <- plot_grid(basic_graph_cat_bx, iris_cowplot,
                   labels = c("a)", "b)"),
                   nrow = 2,
                   align = "v")
print (fig.1)

plot of chunk unnamed-chunk-41

While modeling

Function “plot” in R-base for diagnostic plots

lm.iris <- lm(Petal.Length ~ Sepal.Length + Sepal.Width + Petal.Width,
            data = iris)
plot(lm.iris)

plot of chunk unnamed-chunk-42plot of chunk unnamed-chunk-42plot of chunk unnamed-chunk-42plot of chunk unnamed-chunk-42

Extending ggplot2 for modeling

Data visualization tools for statistical analysis results

ggfortify provides autoplotand
fortify for common models

require(ggfortify)
autoplot(lm.iris)

plot of chunk unnamed-chunk-44