ggplot.md

Data visualization with ggplot2

We will need

library("dplyr")
library("readr")
library("ggplot2")
library("hexbin")

Introduction

(Based on slides by Wolfgang Huber)

Base graphics

Uses a canvas model a series of instructions that sequentially fill the plotting canvas. While this model is very useful to build plots bits by bits bottom up, which is useful in some cases, it has some clear drawback:

• Layout choices have to be made without global overview over what may still be coming.
• Different functions for different plot types with different interfaces.
• No standard data input.
• Many routine tasks require a lot of boilerplate code.
• No concept of facets/lattices/viewports.
• Poor default colours.

The grammar of graphics

The components of ggplot2's of graphics are

1. A tidy dataset
2. A choice of geometric objects that servers as the visual representation of the data - for instance, points, lines, rectangles, contours.
3. A description of how the variables in the data are mapped to visual properties (aesthetics) or the geometric objects, and an associated scale (e.g. linear, logarithmic, rang)
4. A statistical summarisation rule
5. A coordinate system.
6. A facet specification, i.e. the use of several plots to look at the same data.

Plotting with ggplot2

Credit: This material is based on the Data Carpentry R for data analysis and visualization of Ecological Data material

We are going to use a complete version of the surveys data:

surveys <- read_csv("https://ndownloader.figshare.com/files/2292169")
surveys_complete <- surveys %>%
  filter(species_id != "",         # remove missing species_id
         !is.na(weight),           # remove missing weight
         !is.na(hindfoot_length),  # remove missing hindfoot_length
         sex != "")                # remove missing sex

To build a ggplot we need to:

• bind the plot to a specific data frame using the data argument

ggplot(data = surveys_complete)

• define aesthetics (aes), by selecting the variables to be plotted and the variables to define the presentation such as plotting size, shape color, etc.,

ggplot(data = surveys_complete, aes(x = weight, y = hindfoot_length))

• add geoms -- graphical representation of the data in the plot (points, lines, bars). To add a geom to the plot use + operator:

ggplot(data = surveys_complete, aes(x = weight, y = hindfoot_length)) +
  geom_point()

In practice, we prepare the data and aesthetics and store them in a plot ggplot variable that we can re-use during our data exploration using different geoms.

surveys_plot <-
    ggplot(data = surveys_complete,
           aes(x = weight, y = hindfoot_length))

surveys_plot + geom_point()

surveys_plot + geom_hex()

Building your plots iteratively

Building plots with ggplot is typically an iterative process. We start by defining the dataset we'll use, lay the axes, and choose a geom.

surveys_plot + geom_point()

Then, we start modifying this plot to extract more information from it. For instance, we can add transparency (alpha) to avoid overplotting.

surveys_plot +
    geom_point(alpha = 0.1)

We can also add colors for all the points

surveys_plot +
    geom_point(alpha = 0.1, color = "blue")

Or to color each species in the plot differently:

surveys_plot +
    geom_point(alpha = 0.1, aes(color = species_id))

Boxplot

Visualising the distribution of weight within each species.

surveys_bw <- ggplot(data = surveys_complete,
                     aes(x = species_id, y = hindfoot_length))
surveys_bw + geom_boxplot()

By adding points to boxplot, we can have a better idea of the number of measurements and of their distribution:

surveys_bw + 
    geom_boxplot(alpha = 0.6) +
    geom_jitter(alpha = 0.1, color = "tomato")

Notice how the boxplot layer is behind the jitter layer? What do you need to change in the code to put the boxplot in front of the points such that it's not hidden.

Challenges

Boxplots are useful summaries, but hide the shape of the distribution. For example, if there is a bimodal distribution, this would not be observed with a boxplot. An alternative to the boxplot is the violin plot (sometimes known as a beanplot), where the shape (of the density of points) is drawn.

• Replace the box plot with a violin plot; see geom_violin()

In many types of data, it is important to consider the scale of the observations. For example, it may be worth changing the scale of the axis to better distribute the observations in the space of the plot. Changing the scale of the axes is done similarly to adding/modifying other components (i.e., by incrementally adding commands).

• Represent weight on the log10 scale; see scale_y_log10()

• Create boxplot for weight.

surveys_bw + geom_violin()

surveys_bw + geom_boxplot() + scale_y_log10()

ggplot(data = surveys_complete,
       aes(x = species_id, y = weight)) +
    geom_boxplot()

Plotting time series data

Let's calculate number of counts per year for each species. To do that we need to group data first and count records within each group.

yearly_counts <- surveys_complete %>%
                 group_by(year, species_id) %>%
                 tally

Timelapse data can be visualised as a line plot with years on x axis and counts on y axis.

ggplot(data = yearly_counts, aes(x = year, y = n)) +
     geom_line()

Unfortunately this does not work, because we plot data for all the species together. We need to tell ggplot to draw a line for each species by modifying the aesthetic function to include group = species_id.

ggplot(data = yearly_counts,
       aes(x = year, y = n, group = species_id)) +
    geom_line()

We will be able to distinguish species in the plot if we add colors.

ggplot(data = yearly_counts,
       aes(x = year, y = n, group = species_id, colour = species_id)) +
    geom_line()

Faceting

ggplot has a special technique called faceting that allows to split one plot into multiple plots based on a factor included in the dataset. We will use it to make one plot for a time series for each species.

ggplot(data = yearly_counts,
       aes(x = year, y = n, group = species_id, colour = species_id)) +
    geom_line() +
    facet_wrap(~ species_id)

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

Now we would like to split line in each plot by sex of each individual measured. To do that we need to make counts in data frame grouped by year, species_id, and sex:

yearly_sex_counts <- surveys_complete %>%
    group_by(year, species_id, sex) %>%
    tally

We can now make the faceted plot splitting further by sex (within a single plot):

ggplot(data = yearly_sex_counts,
       aes(x = year, y = n, color = species_id, group = sex)) +
     geom_line() +
     facet_wrap(~ species_id)

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

Usually plots with white background look more readable when printed. We can set the background to white using the function theme_bw(). Additionally you can also remove the grid.

ggplot(data = yearly_sex_counts,
       aes(x = year, y = n, color = species_id, group = sex)) +
     geom_line() +
     facet_wrap(~ species_id) +
     theme_bw() 

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

Challenges

Modify the plotting code above to colour the time series by sex in the different facets.

To make the plot easier to read, we can color by sex instead of species (species are already in separate plots, so we don’t need to distinguish them further).

ggplot(data = yearly_sex_counts,
       aes(x = year, y = n, color = sex, group = sex)) +
    geom_line() +
    facet_wrap(~ species_id) +
    theme_bw()

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

The ggplot2 themes

In addition of theme_bw() that changes the plot background to white, ggplot2 comes with several other themes, which can be useful to quickly change the look and feel of your visualization. The complete list of themes is available at http://docs.ggplot2.org/current/ggtheme.html. theme_minimal() and theme_light() are popular, and theme_void() can be useful as a starting point to create a new hand-crafted theme.

Challenge

Use what you just learned to create a plot that depicts how the average weight of each species changes through the years.

yearly_weight <- surveys_complete %>%
                 group_by(year, species_id) %>%
                 summarise(avg_weight = mean(weight))
ggplot(data = yearly_weight,
       aes(x=year, y=avg_weight, color = species_id, group = species_id)) +
    geom_line() +
    facet_wrap(~ species_id) +
    theme_bw()

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

References

• ggplot2 documentation: http://docs.ggplot2.org/
• ggplot2 cheat sheet
• Graphs in the R cookbook, by Winston Chang.
• ggplot2: Elegant Graphics for Data Analysis by Hadley Wickham (book webpage). (This book is a bit outdated; I believe a new version is in preparation.)

Interactivity with ggvis

This section is based on the on-line ggvis documentation

The goal of ggvis is to make it easy to build interactive graphics for exploratory data analysis. ggvis has a similar underlying theory to ggplot2 (the grammar of graphics), but it’s expressed a little differently, and adds new features to make your plots interactive. ggvis also incorporates shiny’s reactive programming model and dplyr’s grammar of data transformation.

library("ggvis")
sml <- sample(nrow(surveys), 1e3)
surveys_sml <- surveys_complete[sml, ]

p <- ggvis(surveys_sml, x = ~weight, y = ~hindfoot_length)
p %>% layer_points()

surveys_sml %>%
    ggvis(x = ~weight, y = ~hindfoot_length,
          fill = ~species_id) %>%
    layer_points()

p %>% layer_points(fill = ~species_id)
p %>% layer_points(shape = ~species_id)

To set fixed plotting parameters, use :=.

p %>% layer_points(fill := "red", stroke := "black")
p %>% layer_points(size := 300, opacity := 0.4)
p %>% layer_points(shape := "cross")

Interactivity

p %>% layer_points(
          size := input_slider(10, 100),
          opacity := input_slider(0, 1))

p %>%
    layer_points() %>% 
    add_tooltip(function(df) df$weight)

• input_slider()
• input_checkbox()
• input_checkboxgroup()
• input_numeric()
• input_radiobuttons()
• input_select()
• input_text()

See the interactivity vignette for details.

Layers

Simple layers

• layer_points(), with properties x, y, shape, stroke, fill, strokeOpacity, fillOpacity, and opacity.
• layer_paths(), for paths and polygons (using the fill argument).
• layer_ribbons() for filled areas.
• layer_rects(), layer_text().

Compound layers, which which combine data transformations with one or more simple layers.

• layer_lines() which automatically orders by the x variable with arrange().
• layer_histograms() and layer_freqpolys(), which first bin the data with compute_bin().
• layer_smooths(), which fits and plots a smooth model to the data using compute_smooth().

See the layers vignette for details.

Like for ggplot2's geoms, we can overly multiple layers:

p %>%
    layer_points() %>%
    layer_smooths(stroke := "red")

More components

• scales, to control the mapping between data and visual properties; see the properties and scales vignette.

• legends and axes to control the appearance of the guides produced by the scales. See the axes and legends vignette.