How to code this kind of extreme outliers presentation

[Kingpin]

Hey guys,

I found the following picture on stackexchange, but unfortunately it is without a code. Does anyone know how to code this kind of presentation for extreme outliers into a ggplot boxplot and violin plot?

Thanks!

[nirgrahamuk]

this looks to be a base r boxplot with some number annotations at the top ?

[FJCC]

Here is a start at doing that sort of thing with ggplot.

library(ggplot2)
DF <- data.frame(Source = c(rep(c("Case", "Control"),30), 
                            "Case", "Case", "Case"),
                 Value = c(rnorm(60), 10, 35, 124))
ggplot(DF, aes(Source, Value)) + geom_boxplot() +
  coord_cartesian(ylim = c(-5, 5)) +
  geom_segment(aes(x = "Case", y = 4, xend = "Case", yend = 5),
               arrow = arrow(length = unit(0.05, "npc"))) + 
  annotate(geom = "text", x = 1.2, y = 4, label = "10, 35, 124")

^{Created on 2022-02-21 by the reprex package (v2.0.1)}