[R] Spicing Up Native Circular Plot Using ggplot2

[Sidoti, Salvatore A.]

I have some angle data from an animal behavior study that I would like to plot for publication using ggplot2. What follows is my current workflow with some example data.

### BEGIN SCRIPT ###

### Create two data frames of random Cartesian coordinates ###
df1 <- data.frame(
x = sample(10, 11, replace = TRUE),
y = sample(10, 11, replace = TRUE))

df2 <- data.frame(
x = sample(10, 11, replace = TRUE),
y = sample(10, 11, replace = TRUE))

### Write a function that converts continuous Cartesian coordinates to velocities ###
get.polar <- function(df)
{
x <- diff(df$x)
y <- diff(df$y)
d <- complex(real = x, imaginary = y)
steps <- data.frame(speed = Mod(d), angle = Arg(d))
steps[-1,] # Deletes the first row as it does not contain an angle measurement
steps$time <- (1:nrow(steps))/30 # generates a time column in seconds (1 data point = 1/30 of a second)
return(steps)
}

df1_polar <- get.polar(df1)
df2_polar <- get.polar(df2)

require(circular)

### Convert angles into an object of type 'circular' ###
df1_rad <- circular(df1_polar$angle, type = 'angles', units = 'radians', zero=0, rotation = "counter")
df2_rad <- circular(df2_polar$angle, type = 'angles', units = 'radians', zero=0, rotation = "counter")

### Convert radians to degrees with a clockwise rotation and zero at "north" ###
df1_deg <- conversion.circular(df1_rad, type = "angles", units = "degrees", zero = pi/2, rotation = "clock")
df2_deg <- conversion.circular(df2_rad, type = "angles", units = "degrees", zero = pi/2, rotation = "clock")

### Convert negative rotations to positive ###
df1_deg[df1_deg < 0] <- df1_deg[df1_deg < 0] + 360 
df2_deg[df2_deg < 0] <- df2_deg[df2_deg < 0] + 360

par(pty = "s")
plot(df1_deg, units = "degrees")
ticks.circular(circular(seq(0,(11/6)*pi, pi/6)), zero = pi/2, rotation = "clock", tcl  = 0.075)
points(df2_deg, zero = pi/2, rotation = "clock", pch = 16, col = "darkgrey", next.points = -0.2)

### END SCRIPT ###

Some suggestions for turning this rough plot into something publishable using ggplot2?

Thank you!
Salvatore A. Sidoti