Barplot ascending by group with partially rescaled y axis in R

December 11, 2021

I have a dataframe like the following:

group <- c("A", "A", "B","B","B","C","B","A","B","B","C","B","B","C","B","B","B","B","B","C")
name <- c("nam1","nam2","nam3","nam4","nam5","nam6","nam7","nam8","nam9","nam10","nam11","nam12","nam13","nam14","nam15","nam16","nam17","nam18","nam19","nam20")
obs <- c(7500,950,450,160,150,220,110,60,40,40,40,15,15,15,10,10,10,10,10,10)
stud <- c(10,2,3,5,6,2,2,2,2,1,2,2,2,3,2,2,2,2,1,1)
df <- data.frame(group,name,obs,stud)

I need to do a barplot to show the number of obs for each name, with number of stud on top of each bar, and with bars arranged in ascending order by group.

Here is what I’ve done:

ggbarplot(df, x = "name", y = "obs",
          fill = "group",               
          color = "white",            
          palette = "jco",            
          x.text.angle = 90,           
          label = stud, label.pos="out", lab.size = 3, order = c("nam20","nam14","nam11","nam6","nam19","nam18","nam17","nam16","nam15","nam13","nam12","nam10","nam9","nam7","nam5","nam4","nam3","nam8","nam2","nam1"),
          xlab = "name",
          ylab = "obs")

and I got this:

Now I need to collapse the y axis only for values of obs above, let’s say, 500, in order to make more visible all the bars, even those that correspond to the smaller values of obs.

Thanks in advance.

>Solution :

Couple of options: try scale_y_log10 or ggbreak::scale_y_break…

You’ll have to play around with the functions or other ggplot arguments for fine adjustments to get the appearance you need.

library(ggpubr)
library(ggplot2)
library(ggbreak)

ggbarplot(df, x = "name", y = "obs",
          fill = "group",               
          color = "white",            
          palette = "jco",            
          x.text.angle = 90,           
          label = stud, label.pos="out", lab.size = 3, order = c("nam20","nam14","nam11","nam6","nam19","nam18","nam17","nam16","nam15","nam13","nam12","nam10","nam9","nam7","nam5","nam4","nam3","nam8","nam2","nam1"),
          xlab = "name",
          ylab = "obs",
          title = "Log scale") + 
  scale_y_log10()


ggbarplot(df, x = "name", y = "obs",
          fill = "group",               
          color = "white",            
          palette = "jco",            
          x.text.angle = 90,           
          label = stud, label.pos="out", lab.size = 3, order = c("nam20","nam14","nam11","nam6","nam19","nam18","nam17","nam16","nam15","nam13","nam12","nam10","nam9","nam7","nam5","nam4","nam3","nam8","nam2","nam1"),
          xlab = "name",
          ylab = "obs",
          title = "ggbreak") +
  scale_y_continuous(limits = c(0, 7600))+
  scale_y_break(c(1000, 7200), scales = 0.3)