Plotting with GGPLOT2

First part of this notebook based on Karthik Ram’s GGPLOT2 Lecture (CC-By 2.0)

#install.packages('ggplot2')
#library(ggplot2)

GOALS: Students should be able to use ggplot2 to generate publication quality graphics and understand and use the basics of the grammar of graphics.

• GGPLOT2 developed by Hadley Wickham based on a grammar-of-graphics
• Grammar of graphics consists of a dataset, coordinate system, and geoms – the visual representation of data points.
• Think about figure in layers: like you would in Photoshop, illustrator or Inkscape.

DataViz

• plotting our data is one of the best was to quickly explore it and the various relationships between variables
• 3 main plotting systems in R: the base plotting system, the lattice package, and ggplot2 *ggplot2 is built on the grammar-of-graphics:
• idea that any plot can be expressed from the same set of components: a data set, a coodinate system and a set of geoms– the visual representation of data points
• Key to understanding ggplot2 is thinking about a figure in layers – think of ArcGIS or programs like Photoshop

Terminology:

• ggplot - The main function where you specify the dataset and variables to plot
• geoms - geometric objects
• geom_point(), geom bar(), geom density(), geom line(), geom area()
• aes - aesthetics
• shape, transparency (alpha), color, fill, line-type.
• scales Define how your data will be plotted
• continuous, discrete, log

First Plots with GGPLOT

This famous (Fisher’s or Anderson’s) iris data set gives the measurements in centimeters of the variables sepal length and width and petal length and width, respectively, for 50 flowers from each of 3 species of iris. The species are Iris setosa, versicolor, and virginica.

library(ggplot2)
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

WE can use the data() function to show the available built-in data sets in R.

data()

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point()

Basic structure

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) + geom_point()

myplot <- ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width))
myplot + geom_point()

• Specify the data and variables inside the ggplot function.
• Anything else that goes in here becomes a global setting.
• Then add layers of geometric objects, statistical models, and panels.

Increase size of points

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point(size = 3)

Make it colorful

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point(size = 3)

Differentiate points by shape

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point(aes(shape = Species), size = 3)

Exercise 1

# Make a small sample of the diamonds dataset
d2 <- diamonds[sample(1:dim(diamonds)[1], 1000), ]

Then generate this plot below. (open 09-plot-ggplot2-ex-1-1.png)

ggplot(d2, aes(carat, price, color = color)) + geom_point() + theme_gray()

Switch to Gapminder Data

#gapminder <- read.csv("https://goo.gl/BtBnPg", header = T)
gapminder <- read.csv('data/gapminder-FiveYearData.csv', header=T)

ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp)) +
  geom_point()

NOTE:

• First we call the ggplot function -any arguments we provide the ggplot function are considered global options: they apply to all layers
• We passed two arguments to ggplot:
• data
• an aes function - which tells ggplot how variables map to aesthetic properties
• x & y locations

Alone the ggplot call isn’t enough to render the plot.

ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap))
## If run, would produce an error.

Need to tell ggplot how we want to present variables by specifying a geom layer. In the above example we used geom_point to create a scatter plot.

ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap)) +
  geom_point()

Challenge 1

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-1

Challenge 2

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-2

Box plots

See ?geom boxplot for list of options

library(MASS)
head(birthwt)

   low age lwt race smoke ptl ht ui ftv  bwt
85   0  19 182    2     0   0  0  1   0 2523
86   0  33 155    3     0   0  0  0   3 2551
87   0  20 105    1     1   0  0  0   1 2557
88   0  21 108    1     1   0  0  1   2 2594
89   0  18 107    1     1   0  0  1   0 2600
91   0  21 124    3     0   0  0  0   0 2622

ggplot(birthwt, aes(factor(race), bwt)) + geom_boxplot()

Histograms

See ?geom histogram for list of options

h <- ggplot(faithful, aes(x = waiting))
h + geom_histogram(binwidth = 30, colour = "black")

• adding color and adjusting the binwidth

h <- ggplot(faithful, aes(x = waiting))
h + geom_histogram(binwidth = 8, fill = "steelblue", colour = "black")

Line plots

Download data

download.file('https://raw.github.com/karthikram/ggplot-lecture/master/climate.csv', 'data/climate.csv')
#climate <- read.csv(text=RCurl::getURL(https://raw.github.com/karthikram/ggplot-lecture/master/climate.csv))

• In the climate data set, Anomaly10y is a 10-year running average of the deviation (in Celsius) from the average 1950–1980 temperature, and Unc10y is the 95% confidence interval. We’ll set ymax and ymin to Anomaly10y plus or minus Unc10y (Figure 4-25):

climate <- read.csv("data/climate.csv", header = T)
ggplot(climate, aes(Year, Anomaly10y)) +
  geom_line()

We can also plot confidence regions

• In the climate data set, Anomaly10y is a 10-year running average of the deviation (in Celsius) from the average 1950–1980 temperature, and Unc10y is the 95% confidence interval. We’ll set ymax and ymin to Anomaly10y plus or minus Unc10y (Figure 4-25):

ggplot(climate, aes(Year, Anomaly10y)) +
  geom_ribbon(aes(ymin = Anomaly10y - Unc10y, ymax = Anomaly10y + Unc10y),
              fill = "blue", alpha = .1) +
  geom_line(color = "steelblue")

Exercise 2

Modify the previous plot and change it such that there are three lines instead of one with a confidence band.

cplot <- ggplot(climate, aes(Year, Anomaly10y))
cplot <- cplot + geom_line(size = 0.7, color = "black")
cplot <- cplot + geom_line(aes(Year, Anomaly10y + Unc10y), linetype = "dashed", size = 0.7, color = "red")
cplot <- cplot + geom_line(aes(Year, Anomaly10y - Unc10y), linetype = "dashed", size = 0.7, color = "red")
cplot + theme_gray()

#theme_classic
#theme_bw()
#theme_minimal()

Gapminder line graph

Using scatter plot not the best way to visualize change over time. Let’s use line plot.

ggplot(data = gapminder, aes(x=year, y=lifeExp, by=country, color=continent)) +
  geom_line()

• used geom_line instead of geom_point

• added a by aesthetic to get a line per country and color by continent

• visualize both lines and points on the plot?

ggplot(data = gapminder, aes(x=year, y=lifeExp, by=country, color=continent)) +
  geom_line() + geom_point()

• note this is layered: so points have been drawn on top of the lines.
• example of this

ggplot(data = gapminder, aes(x=year, y=lifeExp, by=country)) +
  geom_line(aes(color=continent)) + geom_point()

• the aesthetic mapping of color has been moved from the global plot options in ggplot to the geom_line layer so it no longer applies to the points

Challenge 3

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-3

Bar Plots

ggplot(iris, aes(Species, Sepal.Length)) +
  geom_bar(stat = "identity")

library(tidyr)
#df <- melt(iris, id.vars = "Species")
df <- gather(iris, variable, value, -Species )
ggplot(df, aes(Species, value, fill = variable)) +
  geom_bar(stat = "identity")

The heights of the bars commonly represent one of two things: either a count of cases in each group, or the values in a column of the data frame. By default, geom_bar uses stat=“bin”. This makes the height of each bar equal to the number of cases in each group, and it is incompatible with mapping values to the y aesthetic. If you want the heights of the bars to represent values in the data, use stat=“identity” and map a value to the y aesthetic.

Dplyr and Tidyr

These two packages are the Swiss army knives of R. dplyr * filter * select * mutate * tidyr. * gather * spread * separate

Let’s look at iris again.

iris[1:2, ]

  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

df <- gather(iris, variable, value, -Species ) 

df[1:2, ]

  Species     variable value
1  setosa Sepal.Length   5.1
2  setosa Sepal.Length   4.9

ggplot(df, aes(Species, value, fill = variable)) +
  geom_bar(stat = "identity", position = "dodge")

Exercise 3

Using the d2 dataset you created earlier, generate this plot below. Take a quick look at the data first to see if it needs to be binned

d2 <- diamonds[sample(1:dim(diamonds)[1], 1000), ]
ggplot(d2, aes(clarity, fill = cut)) +
  geom_bar(position = "dodge")

Exercise 4

• Using the climate dataset, create a new variable called sign. Make it logical (true/false) based on the sign of Anomaly10y.
• Plot a bar plot and use sign variable as the fill.
• HINT: Look up ifelse function to create clim$sign

clim <- read.csv('data/climate.csv', header = TRUE)
clim$sign <- ifelse(clim$Anomaly10y<0, FALSE, TRUE)
# or as simple as
# clim$sign <- clim$Anomaly10y < 0
ggplot(clim, aes(Year, Anomaly10y)) + geom_bar(stat = "identity", aes(fill = sign)) + theme_gray()

Density Plots

ggplot(faithful, aes(waiting)) + geom_density()

ggplot(faithful, aes(waiting)) +
  geom_density(fill = "blue", alpha = 0.1)

ggplot(faithful, aes(waiting)) +
  geom_line(stat = "density")

Colors

aes(color = variable) 
aes(color = "black")
# Or add it as a scale
scale_fill_manual(values = c("color1", "color2"))

library(RColorBrewer) 
display.brewer.all() 

Using a color brewer palette

#df  <- melt(iris, id.vars = "Species")
ggplot(df, aes(Species, value, fill = variable)) +
  geom_bar(stat = "identity", position = "dodge") +
  scale_fill_brewer(palette = "Set1")

Manual color scale

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
geom_point() +
facet_grid(Species ~ .) +
scale_color_manual(values = c("red", "green", "blue"))

Transformations and statistics

• overlay statistical models over the data
• let’s use our first example

ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap, color=continent)) +
  geom_point()

• hard to see relationships b/t points because of strong outliers in GDP/cap
• We can change the scale of units on the y axis using the scale functions
• These control the mapping between the data values and visual values of an aesthetic.
• We’ll also use the alpha function, which is helpful when you have a large amount of data which is v. clustered

ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap)) +
  geom_point() + scale_y_log10()

• log10 function applied a transformation to the values of the gdpPercap column before rendering them on the plot
• each multiple of 10 now only corresponds to an increase in 1 on the transformed scale, e.g. a GDP per capita of 1,000 is now 3 on the y axis, a value of 10,000 corresponds to 4 on the y axis and so on

• This makes it easier to visualize the spread of data on the y-axis.

• We can fit a simple relationship to the data by adding another layer, geom_smooth:

ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap)) +
  geom_point() + scale_y_log10() + geom_smooth(method="lm")

• make the line thicker by setting the size aesthetic in the geom_smooth layer:

pwd <- ggplot(data = gapminder, aes(x = lifeExp, y = gdpPercap)) +
  geom_point() + scale_y_log10() + geom_smooth(method="lm", size=1.5)

• two ways an aesthetic can be specified:

1. Here we set the size aesthetic by passing it as an argument to geom_smooth.
2. use the aes function to define a mapping between data variables and their visual representation.

With iris data - smooth

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point(aes(shape = Species), size = 3) +
  geom_smooth(method = "lm")

Within facet

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point(aes(shape = Species), size = 3) +
  geom_smooth(method = "lm") +
  facet_grid(. ~ Species)

Challenge 4a

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-3#challenge-4a

Challenge 4b

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-3#challenge-4b

Multi-panel figures: FACEting

• we can split this out over multiple panels by adding a layer of facet panels:

starts.with <- substr(gapminder$country, start = 1, stop = 1)
az.countries <- gapminder[starts.with %in% c("A", "Z"), ]
ggplot(data = az.countries, aes(x = year, y = lifeExp, color=continent)) +
  geom_line() + facet_wrap( ~ country)

• facet_wrap layer took a “formula” as its argument, denoted by the tilde (~).
• tells R to draw a panel for each unique value in the country column of the gapminder dataset.

Modifying text

• would like to add text to elements in the graph
• do this by adding a few more layers:
• theme layer controls axis text & text size
• scales layer to change legend title

ggplot(data = az.countries, aes(x = year, y = lifeExp, color=continent)) +
  geom_line() + facet_wrap( ~ country) +
  xlab("Year") + ylab("Life expectancy") + ggtitle("Figure 1") +
  scale_colour_discrete(name="Continent") +
  theme(axis.text.x=element_blank(), axis.ticks.x=element_blank())

http://swcarpentry.github.io/r-novice-gapminder/08-plot-ggplot2#challenge-5

With iris along coloumns

#str(iris)
ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point() +
  facet_grid(Species ~ .)

### And along rows

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point() +
  facet_grid(. ~ Species)

Or wrap your panels

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
  geom_point() +
  facet_wrap( ~ Species)

Themes

Themes are a great way to define custom plots.

+theme() #### see ?theme() for more options

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
geom_point(size = 1.2, shape = 16) +
facet_wrap( ~ Species) +
theme(legend.key = element_rect(fill = NA),
legend.position = "bottom",
strip.background = element_rect(fill = NA),
axis.title.y = element_text(angle = 0)) 

#install.packages('ggthemes')
library(ggthemes)

Then add one of these themes to your plot

• theme_stata()
• theme_excel()
• theme_wsj()
• theme_solarized()

ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) +
geom_point(size = 1.2, shape = 16) +
facet_wrap( ~ Species) +
theme_solarized() +
theme(legend.key = element_rect(fill = NA),
legend.position = "bottom",
strip.background = element_rect(fill = NA),
axis.title.y = element_text(angle = 0)) 

How to save your plots

ggsave('~/path/to/figure/filename.png')
ggsave(plot1, file = "~/path/to/figure/filename.png")

ggsave(file = "/path/to/figure/filename.png", width = 6,
height =4)

ggsave(file = "/path/to/figure/filename.eps")
ggsave(file = "/path/to/figure/filename.jpg")
ggsave(file = "/path/to/figure/filename.pdf")

Resources:

This is just a taste of what you can do with ggplot2. RStudio provides a really useful cheat sheet of the different layers available, and more extensive documentation is available on the ggplot2 website. Finally, if you have no idea how to change something, a quick Google search will usually send you to a relevant question and answer on Stack Overflow with reusable code to modify!