R中的多个箱形图，同时按列和行将矩阵分组

Question

I am having trouble trying to figure out how to make a single figure containing multiple boxplots in R, while grouping my data frame/ matrix both by columns and rows.

I have a data frame in R with 10 rows and 500 columns. The columns are separated into 2 groups(factors - 1's and 2's) and now I want to have a single figure containing two boxplots for each row of my data frame subject to the column groups.

Ex.

    M1 N2 O1 P2 Q1 R2      # [The 1's and 2's refer to my two column groups]
 A  10 11 12 13 14 15
 B  15 14 13 12 11 10 
 C  20 21 22 23 24 25
 D  25 24 23 22 21 20

So for the above example I would like to have a single figure with "4 boxplots pairs" for each row such that each boxplot pair will represent values corresponding to 1's and 2's factors of my column.

Thanks in Advance !!!

Answer 1

Here on idea using reshape2 . Since You have more columns than rows, it natural to work on the transpose.

library(ggplot2)
library(reshape2)
dt <- read.table(text='
M1 N2 O1 P2 Q1 R2     
A  10 11 12 13 14 15
B  15 14 13 12 11 10 
C  20 21 22 23 24 25
D  25 24 23 22 21 20',header=TRUE)
dt.m <- melt(t(dt))
dt.m$Var1 <- gsub('[A-Z]','',dt.m$Var1)

Here 2 options to plot :

library(ggplot2)
library(gridExtra)
p1 <- ggplot(dt.m) +
   geom_boxplot(aes(x=Var2,y=value,fill=Var1))

p2 <- ggplot(dt.m) +
  geom_boxplot(aes(x=Var2,y=value,fill=Var2))+
  facet_grid(~Var1)

grid.arrange(p1,p2)