Plotting Data With Multiple Axes

Question

[pixie@bioinfo]

Hello, I have a couple of co-expressed gene clusters corresponding to each developmental stage of a plant, (total 3 such stages). I want to plot the distribution of genes in the modules for each of the stages, in a compact form since I have to put them in a poster. Pie charts are taking too much space.

Kindly suggest a resource where I can do this sort of a thing: Plots with Multiple axes

Example of my data:

Clusters x-stage y-stage    z-stage
black         91         190    127
green        163         564    258
turquoise    318         420    149
red          200         381     295

[PoGibas]

First of all, you can show us how your data looks like.
My answer probably would be ggplot.

You can (should) also try BoxPlotR (a web-tool for generation of box plots).

Answer 1

[Pavel Senin]

I think that you can achieve the goal by using ggplot in order to make individual plots and then combine those with grid & gridExtra packages as it is shown here. Yet another option would be to use facet-grid of ggplot. If you update your post with few examples of the data and figures that you have, I am sure that you'll be able to get more help here.

Here is an easy way of visualization of your data using facets and barplots, i.e. geom_hist:

require(ggplot2)
require(gridExtra)
require(reshape)

dat=read.table(header=T, text="Clusters x-stage y-stage    z-stage
black         91         190    127
green        163         564    258
turquoise    318         420    149
red          200         381     295")

dm=melt(dat,id.var=c("Clusters"))

p1 <- ggplot(dm, aes(variable, value, fill=variable)) + theme_bw() +
  geom_histogram(stat="identity") + facet_grid(. ~ Clusters)

p2 <- ggplot(dm, aes(Clusters, value, fill=Clusters)) +
  geom_histogram(stat="identity") + facet_grid(. ~ variable)

print(arrangeGrob(p1, p2, ncol=1))

You can adjust theme and use other visualizations (geom_boxplot, stat_ydensity) with different data.