使用R的分类数据图

Question

I've a list of protein names(P1,P2,...,Pn) and they are categorized to three different expression levels High(H), medium(M) and Low(L) as measured in three experimental conditions (Exp1,Exp2, and Exp3).

I wish to make a plot as shown in the bottom part of the figure, with the name of the proteins at the left and name of experiments along the top and high, medium and low categories are indicated by Red,blue and green respectively.

I'm new to R, I would much appreciate any help.

Thanks in advance

Answer 1

You can create a file with data formatted like this (tab delimited):

pv   exp  val
1    1    H
2    1    L
3    1    L
4    1    M
1    2    H
2    2    H
3    2    M
4    2    H
1    3    L
2    3    L
3    3    L
4    3    M

And used the following commands to grab and plot them:

mat <- read.table(file.choose(),header=T) # read the file into memory

attach(mat) # map the header names to variable names

plot(pv~exp,col=val) # plot the categories against each other and use val (H,M,L) as the color array. R will assign those values to colors on its own. You can also create a color array using the val array to translate (H,M,L) to (Blue,Red,Green)... but there is other documentation out there for that.

Answer 2

Here is an approach that uses some of the magic of the ggplot2 and reshape2 packages.

First, recreate the data in the format you described:

df <- data.frame(
    P    = paste("P", 1:4, sep=""),
    Exp1 = c("L", "H", "L", "M"),
    Exp2 = c("M", "M", "L", "H"),
    Exp3 = c("H", "L", "L", "M"))

Next, load the add-on packages:

library(reshape2)
library(ggplot2)

Then, use melt() to convert your data from wide format to tall format. The id variable is "P", and we tell the function to rename the "variable" to "Exp":

mdf <- melt(df, id.vars="P", variable="Exp")

Because L - M - H has semantic order, we use the ordered parameter of factor() to inform R of this order:

mdf$value <- factor(mdf$value, levels=c("H", "M", "L"), ordered=TRUE)

Finally, we are ready to plot your data:

ggplot(mdf, aes(x=Exp, y=P, colour=value)) + 
    geom_point(size=3) + 
    scale_colour_manual(value=c("red", "green", "blue")) +
    xlab("") + 
    ylab("")