How to annotate across or between plots in multi-p

问题:

If I have several plots in a single panel, I would like to be able to put text annotations that go across more than one plot area or that go in between plot areas. I'm sure there must be a way to do this, but I need some help figuring out how.
Here's some fake data to create four plots on one panel:

x1 = rnorm(100)
x2 = rnorm(100)
x3 = rnorm(100)
x4 = rnorm(100)

par(mfrow=c(2,2)) 
hist(x1, xlab="", main="Group A")
hist(x2, xlab="", main="Group B")
hist(x3, xlab="", main="")
hist(x4, xlab="", main="")

This creates the multi-plot panel below, except that I've also added text in red. I added the red text by annotating the PDF file after I created the graph panel in R and then saved it as a PDF. What I'd like to learn is how to annotate the graph panel programmatically within R, rather than having to save it and annotate the saved file.
Normally, I would just add an annotation using text() or mtext(). But that only works if you're annotating a single graph. Is there a way to add annotations relative to the coordinates of the entire panel of 4 graphs? I'm looking for a solution in base graphics right now, but if anyone knows how to do something similar in ggplot2 or lattice, I'd be interested in seeing that as well.

回答1:

If you truly want finer control over these kinds of layout issues, you can use the aptly named layout.

m <- matrix(c(1,2,3,3,4,5,6,6),ncol = 2,byrow = TRUE)
layout(m,widths = c(0.5,0.5),heights = c(0.45,0.05,0.45,0.05))

par(mar = c(2,4,4,2) + 0.1)
hist(x1, xlab="", main="Group A")
hist(x2, xlab="", main="Group B")

par(mar = c(0,0,0,0)) 
plot(1,1,type = "n",frame.plot = FALSE,axes = FALSE)
u <- par("usr")
text(1,u[4],labels = "Here",col = "red",pos = 1)

par(mar = c(2,4,2,2) + 0.1)
hist(x3, xlab="", main="")
hist(x4, xlab="", main="")

par(mar = c(0,0,0,0)) 
plot(1,1,type = "n",frame.plot = FALSE,axes = FALSE)
u <- par("usr")
text(1,u[4],labels = "Here",col = "red",pos = 1)

回答2:

You want to use mtext with outer=TRUE. For instance, mtext('Results2', side=1, line=-2, outer=TRUE) and mtext('Results1', side=3, line=-20, outer=TRUE).

回答3:

For this example the layout function is simplest as has been already shown, but for more general cases you should also learn the grconvertX and grconvertY functions (along with setting clipping using par(xpd=. The convert functions can give you the coordinates in the current user coordinates for different coordinate systems, so for example you could center something on the page by converting 0.5 from the normalized device coordinates.