Question

I can do it with classic boxplot. Here we use the built-in data: PlantGrown as example.

attach(PlantGrowth)    
boxplot(weight~group,data=PlantGrowth,xaxt="n")
PlantGrowthSum=ddply(PlantGrowth,.(group),summarise,sum=length(weight))

> PlantGrowthSum
   group sum
1  ctrl  10
2  trt1  10
3  trt2  10

axis(1,1:3,paste(PlantGrowthSum$group,"(",PlantGrowthSum$sum,")",sep=""))

Here is a question, how about ggplot2?

library(ggplot2)
bp <- ggplot(data=PlantGrowth, aes(x=group, y=weight, fill=group))
    + geom_boxplot()
    +theme(axis.text.x=element_blank())
    +theme(axis.text.x=1:3)
bp

But it failed. Any clues about which parameter should be set up?

Answer 1

As in this case x values are discrete, you should use scale_x_discrete() to set labels for the x axis.

bp <- ggplot(data=PlantGrowth, aes(x=group, y=weight, fill=group))+
geom_boxplot()
bp+scale_x_discrete(labels=paste(PlantGrowthSum$group,"(",PlantGrowthSum$sum,")",sep=""))

More information and example about scales and other elements of ggplot2 plot can be found in ggplot2 documentation site.