Making figures in R¶

Useful resources and tutorials for visualising data using R.

General resources¶

The R Graph Gallery

If you want to make a graph in R, The R Graph Gallery is the place to start. It has lots of different ideas for plots, and simple to complex tutorials.
A ggplot2 tutorial for beautiful plotting in R

Lots of information and resources to help you make pretty plots.
Data Visualisation

Content from PMAP 8921: Data Visualization (Summer 2021) by Dr Andrew Heiss
The Evolution of a ggplot

Awesome blog post about how a simple ggplot can evolve into an informative and appealing figure.
R Base Graphics: An Idiot’s Guide

A blog on creating graphics in R using the preinstalled code, instead of packages such as ggplot2.
A Gentle Guide to the Grammar of Graphics with ggplot2

viridis

A pretty and colour-blind friendly palette!
ggsci

ggsci has some colour palettes based on various journals if you’re feeling fancy, or more excitingly some based on things like star trek…
Pride Palettes

Pride flag colour schemes!

Eulerr

If you are struggling with venn diagrams that aren’t proportional, Eulerr can provide a fix. It makes area-proportional diagrams, or if that is impossible, a good approximation with included goodness-of-fit stats! (Euler diagrams are very similar to venn diagrams, they just don’t require every set to interact.)
Heatmaps

A useful primer on what heatmaps actually tell us.
Upset plots

If you have lots of groups in your data, and venn diagrams turn into a massive confusing mess, that’s the time to try an upset plot. They use a matrix to represent different intersections, and a bar plot to visualise the number of elements in each intersection. You can also get fancy and introduce queries or extra plots that describe different features of your data.
ggstatsplot

This package runs statistical tests, and creates plots with them included. It also makes lovely violin/box plot combinations which are much more informative than a straightforward box plot.
ggbeeswarm

Are the points in your violin scatter plots overlapping so you can’t see all the data? Ta-da, ggbeeswarm fixed it! You can also find a great tutorial for ggbeeswarm here: Bee Swarm Charts in R.
ggplot2 extensions

All the extensions mentioned above and more!

ggplot2: elegant graphics for data analysis - A guide to the Grammar of Graphics used by ggplot2 - for understanding theory.
R Graphics Cookbook - The whole book isn’t available online, but some useful information is on the site about making data graphics in R.
Fundamentals of Data Visualisation - A guide to effectively communicating your data using figures.
Data visualisation using R, for researchers who don’t use R

Some of these pointers may not make sense until you’re trying to make a certain plot, they’re gathered here in the hopes they are useful.

If you use xlim/ylim or scale_y_continous() to zoom in on a boxplot, ggplot recalculates the range etc based on that limit. If you want to include the outliers but just zoom in without recalculation, use coord_cartesian() to change the limits.
In ggplot, discrete missing values are displayed. Adding scale_color_discrete(na.translate = FALSE) removes them. See tweet source for details.