I have 16 samples and want to graph individual histograms along with the median of the samples for each observation. I would like the emphasis to be on the median, so I wanted to make it the largest (say 4x) and fit the remaining facets around it, like a reverse inset. My ideal layout would be:
med med a b c
med med d e f
g h i j k
l m n o p
Here's some code to make sample faceted histograms:
library(ggplot2)
tibble(
sample = factor(c("median", letters[1:16]), levels = c("median", letters[1:16])),
data = list(c(.1,.1,.1,.2,.8,.9,.9,.9))
) %>% unnest %>%
ggplot(aes(x = data)) +
geom_histogram() +
facet_wrap(~ sample, ncol = 4, scales='free', shrink = TRUE) +
scale_x_continuous(breaks = c(0,.5, 1.0))
I wondered if I could somehow work with scale or shrink but I'm not sure where to start with those. Perhaps I need to place filler data in the slots and then use grid to overlay a larger plot on top? In either case, I'm not sure how to begin and would appreciate any help.
