Making a Multipanel Figure with a Single Bottom Legend

A way to create the common figure in medical journals with multiple panels and a single legend for all panels at the bottom.

knitr::include_graphics(
  here("_posts/2023-08-26-multipanel-figure-with-legend/multipanel_with_legend.png"))

The Multipanel Figure with a Single Legend

It is quite common for medical journals (which strictly limit the number of figures) to publish multipanel figures with a single legend for all panels of the figure at the bottom, in its own bounding box.

Note that this blog post is the result of several discussions at

https://community.rstudio.com/t/plot-in-a-box-with-long-text-legend-in-a-box/51962

and

https://stackoverflow.com/questions/60142452/multipanel-plots-with-spanning-figure-legend-in-r-with-ggtext-and-gridtext-in-r

Let’s walk through how to create this.

Data

First, let’s generate data for four dummy figures for the purposes of the layout.

d1 <- runif(500)
d2 <- rep(c("Treatment","Control"), each=250)
d3 <- rbeta(500, shape1=100, shape2=3)
d4 <- d3 + rnorm(500, mean=0, sd=0.1)
plotData <- data.frame(d1, d2, d3, d4)

Plots

Now we will generate the four dummy plots, p1 through p4.

p1 <- ggplot(data=plotData) + geom_point(aes(x=d3, y=d4)) +
  theme(plot.background = element_rect(color='black'))
p2 <- ggplot(data=plotData) + geom_boxplot(aes(x=d2,y=d1,fill=d2))+
  theme(legend.position="none") +
  theme(plot.background = element_rect(color='black'))
p3 <- ggplot(data=plotData) +
  geom_histogram(aes(x=d1, color=I("black"),fill=I("orchid"))) +
  theme(plot.background = element_rect(color='black'))
p4 <- ggplot(data=plotData) +
  geom_histogram(aes(x=d3, color=I("black"),fill=I("goldenrod"))) +
  theme(plot.background = element_rect(color='black'))

Legend Text

Now we will generate the legend text. There are some key pieces here that will turn out to be important later.

1. Setting the padding around the text in the element_textbox with the {ggtext} package. The plot_annotation function in {ggtext} allows some fine tuning.
2. Tweaking the margins on left and right, which are inherited (+5/+5) from the plot above. Note that margins and padding are expressed in TROUBLE (T,R,B,L) order.

fig_legend <- plot_annotation(
  caption = "**Figure 1.**  Testing Control vs. Treatment.   A. Scatterplot. 
  B. The outcomes in the Control Arm were slightly better than 
  the Treatment Arm. C. A Histogram. D. Yet Another Histogram.",
  theme = theme(
    plot.caption = element_textbox_simple(
      size = 11,
      box.colour = "black",
      linetype = 1,
      linewidth = 0.3,
      padding = unit(c(3, 3, 3, 3), "pt"),
      margin = unit(c(0, -5, 0, -5), "pt"), #note negative left and right margins because inherits margins from plots
      r = unit(0, "pt")
    )
  )
)

Multipanel Plot

Now, to put this all together with {patchwork}. I will add tags for general coolness.

p1 + {
  p2 + {
    p3 +
      p4 +
      plot_layout(ncol=1)
  }
} + fig_legend +
  plot_layout(ncol=1) +
  plot_annotation(tag_levels = 'A') & 
  theme(plot.tag = element_text(size = 11, face = 'bold'))