draw line across in a ggplot2

Question

I've got a plot where I'd like to draw a line from (0,0) to (15,15) , with legend. How do I achieve that? Plot:

frame <- read.table('pastie_from_web', sep=",", header=TRUE)
colnames(frame) <- c("pos", "word.length")
plot <- ggplot(frame, aes(x=pos, y=word.length)) + scale_x_continuous(limits=c(1,15)) + scale_y_continuous(limits=c(1,15))+ geom_density2d(aes(color=..level..)) + scale_color_gradient(low="black", high="red") + opts(legend.position="none")
png(paste("graphs/", fname, ".png", sep=""), width=600, height=600)
print(plot)

Data: http://sprunge.us/gKiL or

structure(list(position = c(2, 2, 2, 2, 7, 8, 4, 5, 4, 9, 5, 
2, 7, 9, 9, 6, 5, 6, 9, 2, 6, 5, 5, 7, 7, 5, 6, 5, 5, 3, 2, 4, 
5, 2, 3, 2, 7, 5, 2, 5, 2, 6, 8, 7, 2, 8, 5, 4, 2, 5, 2, 2, 2, 
6, 8, 2, 2, 9, 5, 2, 4, 7, 3, 4, 9, 5, 5, 5, 5, 4, 7, 2, 7, 2, 
4, 4, 3, 2, 5, 6, 5, 5, 5, 5, 4, 4, 8, 7, 5, 7, 4, 3, 4, 5, 2, 
6, 6, 4, 4, 2, 2, 3, 2, 2, 6, 2), word.length = c(5L, 5L, 6L, 
4L, 9L, 11L, 5L, 8L, 8L, 10L, 8L, 9L, 8L, 10L, 10L, 7L, 9L, 10L, 
11L, 10L, 10L, 8L, 13L, 11L, 11L, 13L, 7L, 9L, 6L, 4L, 9L, 8L, 
9L, 6L, 4L, 5L, 11L, 13L, 13L, 13L, 10L, 9L, 11L, 8L, 4L, 10L, 
8L, 16L, 3L, 5L, 4L, 12L, 12L, 15L, 9L, 12L, 12L, 11L, 11L, 8L, 
16L, 9L, 8L, 7L, 10L, 11L, 6L, 13L, 5L, 8L, 8L, 5L, 8L, 5L, 6L, 
6L, 7L, 10L, 13L, 7L, 6L, 13L, 9L, 6L, 7L, 8L, 11L, 8L, 8L, 8L, 
8L, 8L, 7L, 6L, 5L, 9L, 9L, 5L, 5L, 6L, 7L, 8L, 8L, 10L, 8L, 
10L)), .Names = c("position", "word.length"), class = "data.frame", row.names = c(NA, 
-106L))

Answer 1

Here is an example data set to illustrate:

set.seed(42)
dat <- data.frame(x = runif(20, min = 0, max = 20),
                  y = runif(20, min = 0, max = 20))

p <- ggplot(dat, aes(x = x, y = y)) 
p + geom_point() + 
    geom_line(data = data.frame(x = c(0,15), y = c(0,15)),
              aes = aes(x = x, y = y), colour = "red")

Notice how we can specify different data argument to geoms, which allows us to plot different data objects on the same plot regions defined in the original ggplot() call. Note: If the second data frame (in the geom_line() call) has the same x and y axis mapping as the original plot, then you don't need a new aes() as I originally had the code (see revision history of the Answer). This may not have been clear and the comment from @Justin has prompted me to change the geom_line() to include a new aes() call to map the data to the aesthetics; it is not needed in my example but may well be needed in real world use.

The above gives:

If you want different arbitrary lines, consider geom_abline() to draw lines with give slope and intercept. geom_segment() is an alternative to the above geom_line() where you specify the start and end x and y coordinates. See the respective help pages for the geoms to determine which you prefer the use of.