R plotting

Plotting in R

An introduction and brief reference to creating plots in R.

 

Types of plots

This lists the generic plots only. Many more specialized plot-types are available.

• plot()
• pie()
• hist()
• stripchart()
• stem()
• barplot()
• boxplot()

Examples:

# generate some data to plot
x <- rnorm(200)
y <- x^3 * 0.25 + rnorm(200, 0, 0.75)

# plot() ==========================================
?plot
# standard scatterplot
plot(x,y)


# barplot() =======================================
?barplot
barplot(table(round(y)))

# hist() ==========================================
?hist
hist(x, breaks=5)
# add a stripchart() of the actual values
stripchart(x, pch="|", add=TRUE, col="red3", xlim=c(-3, 3), at=-0.5)

# stem() ==========================================
# A stem-and-leaf plot is like a histogram, but output on the console.
# Useful e.g. when you need to send a distribution in the text
# of an e-mail
?stem
stem(x)

# pie() ===========================================
?pie
pie(table(abs(round(x*2))))

# boxplot() =======================================
?boxplot
boxplot(x,y)

Colour

Colours can be specified by number, by name, as hex-triplets as rgb or hsv values, and through color palettes.

Colours by number

The col=... parameter for plots is 1 by default and you can set it to the range 0:8.

• 0: white
• 1: black (the default)
• 2: red
• 3: green
• 4: blue
• 5: cyan
• 6: magenta
• 7: yellow
• 8: grey

barplot(rep(1,9), col=0:8, axes=FALSE, names.arg=c(0:8))

As you can see, these primary colors are decidedly garish and offend even the most minimal sense of aesthetics. Fortunately there are much more sophisticated ways to define colors in R.

Colours by name

You may have noticed that "red", "green", and "blue" work for the col=... parameter, but you probably would not have imagined that "peachpuff", "firebrick" and "goldenrod" are valid as well. In fact, there are 657 named colors in R. Access them all by typing:

colors()

pie(c(1,1,2,3,5,8),
    col=c("peachpuff", "firebrick", "goldenrod", "papayawhip", "whitesmoke", "moccasin")
    )

Read more about named colors (and related topics) at http://research.stowers-institute.org/efg/R/Color/Chart/

Colours as hex-triplets

Hex triplets in R work exactly as in HTML: a triplet of RGB values in two-digit hexadecimal representation. Read more e.g. at http://en.wikipedia.org/wiki/Web_colors

There are many tools on the Web that help to generate pleasing palettes.

Here is an example palette -"Creative Cloud"- taken from http://kuler.adobe.com/

CC <- c("#011640", "#024059", "#F2F0D0", "#BE6C5C", "#8C3037" )
hist(rnorm(1000), breaks=20 , col=CC)

The Alpha channel

R colors are actually specified as quartets: the fourth value the "Alpha channel" defines the transparency. Setting this to values other than "FF" (the default) can be useful for very crowded plots, or for creating overlays.

x <- rnorm(2000)
y <- x^3 * 0.25 + rnorm(2000, 0, 0.75)
# compare:
plot(x,y, pch=19, col="#990000") 
plot(x,y, pch=19, col="#99000008") # Alpha at ~ 10%

# or with multiple overlays of varying size ...
plot(x,y, pch=16, col="#AA330009")
points(x,y, pch=19, cex=2, col="#88554402")
points(x,y, pch=20, cex=0.3, col="#99004408")

Colour palettes

R has several inbuilt color palettes, or you can build your own.

# Inbuilt palettes
?rainbow
# view the palettes
opar <- par(mfrow=c(3,2))
n <- 20
sq <- rep(1, n)
barplot(sq, col=rainbow(n),        axes=F, main="rainbow(n)")
barplot(sq, col=cm.colors(n),      axes=F, main="cm.colors(n)")
barplot(sq, col=topo.colors(n),    axes=F, main="topo.colors(n)")
barplot(sq, col=terrain.colors(n), axes=F, main="terrain.colors(n)")
barplot(sq, col=heat.colors(n),    axes=F, main="heat.colors(n)")
par <- opar


# Example: We generate random points and calculate a "density"
# at each point. The we plot each point and color it according
# to its density. See the section on X-Y-Z plotting re. the 
# inbuilt functions supplied by R for this purpose.

n <- 1000
x <- rnorm(n)
y <- x^3 * 0.25 + rnorm(n, sd=0.75)
z <- rep(0, n) # initialize z
for (i in 1:n) { # calculate a density from the proximity of other points
	dx <- x-x[i]; dx <- dx * dx # square of the distance in x
	dy <- y-y[i]; dy <- dy * dy # square of the distance in y
	d <- dx + dy                # square of the sum
	d <- d[-i]                  # remove the self-element
	z[i] <- 1/sum(d)            # let density decay with 1/r^2
}
z <- z/max(z) - 0.00001 # normalize, but keep the max below 1.0
# now map each of the values of z into an interval of the palette
n <- 20                   # number of intervals
z <- floor(z * n) + 1     # convert to integers
pal <- rainbow(n)         # get a vector of colors from the palette
cz <- pal[z]              # apply the color for each density value
plot(x,y, col=cz, pch=16) # plot

# use a different palette
pal <- heat.colors(n); cz <- pal[z]
plot(x,y, col=cz, pch=16)

Custom palettes

"Cold" values should really be black, not red. Lets define a custom palette: colorRampPalette() is a function that returns a function. The returned function can be used to calculate a palette: a trajectory along a number of waypoints in colorspace. Try:

fc <- colorRampPalette(c("black", "red", "yellow", "white"), bias=0.5) # assigns a function to fc
fc # look at the function
fc(n) # use the function to get n values
pal <- fc(n); cz <- pal[z] # assign the colors according to z
plot(x,y, col=cz, pch=16)

Useful palettes have also been described specifically for cartography. http://colorbrewer2.org/ has palettes for seqential and qualitative diferences, and options for colorblind-safe and photocopy friendly palettes. You can use them via an R package:

install.packages("RColorBrewer")
library("RColorBrewer")
display.brewer.all()

# Here, we apply a Brewer palette to a Voronoi tesselation of a
# point set.

install.packages("voronoi")
library("voronoi")

p  <- pattern.sim(7, type='moon')  # create a point set
tess <- ve(p, rw=c(0,1,0,1))       # calculate the tesselation

# Let's color the cells by distance from a defined point
# using a Brewer palette
vec <- c(0.25, 0.4) # define a point
vDist <- function(x,v) { sqrt(sum((x-v)^2)) }  # calculates Euclidian distance
d <- apply(p,1,vDist, v=vec)                   # apply this to the point set

dCol <- floor(((d-min(d))/(max(d)-min(d)) * 10)) + 1 # map d into 10 intervals
dCol[which(dCol>10)] <- 10                           # demote the largest one

pal <- brewer.pal(10, "RdGy")   # create the palette
plot(tess,col = pal[ dCol ])    # plot the tesselation, color by palette

Other good sources of palettes include:

• Colors for data scientists
• HCL color picker (by Tristen Brown)
  • and an explanation of the superiority of the HCL system.
• color palettes from images

Lines

plot(c(0,10), c(0,10), type = "n", axes = FALSE, xlab = "", ylab = "")

# Line width
for (i in 1:10) {
	y <- 10-(i/2)
	segments(1,y,5,y, lwd=(0.3*i)^2)
	text(6, y, paste("lwd = ", (0.3*i)^2), col="grey60", adj=0, cex=0.75)
}

# Line type
for (i in 1:8) {
	y <- 4-(i/2)
	segments(1,y,5,y, lty=i)
	text(6, y, paste("lty = ", i), col="grey60", adj=0, cex=0.75)
}

• plot()
• axis()
• grid()

# For Details, see:
?plot.default

n <- 1000
x <- rnorm(n)
y <- x^3 * 0.25 + rnorm(n, sd=0.75)

plot(x,y)  # Default

# Axes
plot(x,y, xlim=c(-4, 4)) # fixed limits
plot(x,y, xlim=c(-4, 4), ylim=c(10, -10)) # reverse is possible
plot(x,y, log="xy")

# The axis parameters in the default plot are limited. 
# If you want more control, suppress the printing of an axis
# in the plot and use the axis() function instead. 
?axis

plot(x,y, axes=FALSE)
axis(1)
axis(2)

# Specifying
# - the length of the tickmarks (tcl) as a fraction
#   of the height of a line of text: positive values: 
#   inside, negative values: outside the plot area;
# - the positions of the tickmarks (at). The axis is
# only drawn to the end of the plot area, even if a
# tickmark position is defined beyond it. With this,
# you can make x and y axes intersect.
plot(x,y, axes=FALSE)
axis(1, at=-3:3, tcl=0.3)
axis(2, at=seq(-9, 6, 3), tcl=-0.3)

# Specifying
# - labels: the "labels" vector and the "at" vector must have
#   the same length, even if some tickmarks do not fall
#   inside the plot area and are not shown.
# - gridlines: use tck instead of tcl: (ticks as fraction of plot area)
plot(x,y, axes=FALSE, xlab="", ylab="")
axis(1, at=c(-10, -2, 0, 2, 10), 
        labels=c("", "minus two", "zero", "two", ""),
        tcl=0.3,
        cex.axis=0.75,
        col.axis="steelblue")
axis(2, at=seq(-6, 6, 3),
        labels=rep("", 5), # supress labels with empty string
        lwd=0,             # supress axis line
        tck=1.0,           # draw tickmarks full length of plot area
        lwd.ticks=0.5,     # line width for ticks
        lty=2,             # line type for axis and ticks
        col.ticks="red4"   # color for tickmarks
        )


# Axis-labels and title are straightforward parameters of plot
plot(x,y, xlab="rnorm(n)",
          ylab="x^3 * 0.25 + rnorm(n, sd=0.75)",
          cex.main=1.3,
          main="Sample\nPlot",
          cex.sub=0.75,
          col.sub="grey",
          sub="Scatterplot of noisy 3d-degree polynomial"
          )

# Add gridlines
?grid
grid()

Title, legends etc.

• mtext()
• legend()
• title()
• See also: box()

 

Plotting Characters

• points()

# pch
# id 1:20 are regular symbols
# create coordinates
plot(c(0,10), c(0,10), type = "n", axes = FALSE, xlab = "", ylab = "")

# coordinates for first 25 symbols
x1 <- rep(0.5:9.5, 2)[1:20]
y1 <- sort(rep(9.5:8.5, 10), dec=TRUE)[1:20]
points(x1, y1, pch=1:20)

# id 21:25 can have different border and fill colors
x2 <- 0.5:4.5
y2 <- rep(7.5,5)
 points(x2, y2, pch=21:25, col="slategrey", bg=rainbow(5))
#points(x2, y2, pch=21:25, col="slategrey", bg=c("red", "green", "blue"))

# ten extra symbols are defined as characters
x3 <- 0.5:9.5
y3 <- rep(6.5,10)
extra = c(".", "o", "O", "0","a","A", "*", "+","-","|")
points(x3, y3, pch=extra) # note: ext is a character vector

# The ASCII codes for characters 32 to 126 can also be used as plotting characters
x4 <- rep(seq(0.5,9.5,0.5), 5)[1:96]
y4 <- sort(rep(5.5:0.5, 19), dec=TRUE)[1:96]
points(x4, y4, pch=32:126, col="navyblue")

Plotting arbitrary text

Use the text() function to plot characters and strings to coordinates

# Example: add labels to the symbols
# first set: plain symbols (1 to 20)
text(x1-0.4, y1, paste(1:20), cex=0.75)
# symbols with separate background (21 to 25)
text(x2-0.4, y2, paste(21:25), cex=0.75)
# third set: special characters
text(x3-0.4, y3, extra, col="slateblue", cex=0.75, vfont=c("serif", "plain")) # change font for clarity

# a large set of Hershey vector fonts is available which gives access to many
# more plotting and labeling options via text() 
demo(Hershey)

===Plotting other symbols:===

In the most general way, Unicode characters can be plotted as text. The code is passed in hexadecimal, long integer, with a negative sign.

<source lang="rsplus">
# Here is a quarter note (Unicode: 266a) using plot()
plot(0.5,0.5, pch=-0x266aL, cex=5, xlab="", ylab="")

However, rendering varies across platforms since it depends on unicode support. It is safer to use the inbuilt Hershey vector fonts.

Drawing on plots

• points()
• abline()
• segments()
• lines()
• arrows() ... but to get a filled arrow use polygon()
• grid()
• rect()
• polygon()
• More: see the Index of functions for the graphics package

Plots of X-Y-Z coordinates

?contour
# Example (mostly taken from the help page)
x <- -6:16
op <- par(mfrow = c(2, 2))
contour(outer(x, x), method = "edge", vfont = c("sans serif", "plain"))
z <- outer(x, sqrt(abs(x)), FUN = "/")
image(x, x, z, col=colorRampPalette(c("black", "red", "yellow", "white"))(60))
contour(x, x, z, col = "pink", add = TRUE, method = "edge",
        vfont = c("sans serif", "plain"))
contour(x, x, z, ylim = c(1, 6), method = "simple", labcex = 1)
contour(x, x, z, ylim = c(-6, 6), nlev = 20, lty = 2, method = "simple")
par(op)

#see also persp()
?persp
demo(persp)

The plot window

Most parameters of the plot window can be set via the functions plot(), hist() etc., but some need to be set via the par() function. Calling par() without arguments lists the current sate of the plotting parameters. Calling it with arguments, returns the old parameters and sets new parameters. Thus setting new parameters and saving the old ones can be done in one step. The parameters that have to be set via par include:

• - multiple plots in one window (mfrow, mfcol, mfg)
• - margin layout (mai, mar mex, oma, omd, omi)
• - controlling position and size of a plot in the figure (fig, plt, ps, pty)
• - see par() for details.

See the section on X-Y-Z plots for an example of multiple plots in one window.

n <- 1000
x <- rnorm(n)
y <- x^3 * 0.25 + rnorm(n, sd=0.75)

# set window background and plotting axes via par
opar <- par(bg="steelblue", fg="lightyellow") 
# set axis lables and titles via plot parameters
plot(x,y, col.axis="lightyellow", col.lab="lightyellow")
par(opar)

plot(x,y) # confirm reset

Special packages

ggplot2
ggplot's new Theme system

# In the standard distribution ...
# ... use with library("package")
#   graphics

#   grid

#   lattice

# external, from CRAN
# ... use with install.packages("package"), then 
#              library("package")

#   hexbin

# external, from BioConductor
#   prada:
source("http://www.bioconductor.org/biocLite.R")
biocLite("prada")
#      Try:
n <- 1000
x <- rnorm(n)
y <- x^3 * 0.25 + rnorm(n, sd=0.75)
smoothScatter(x,y, nrpoints=200, pch=20, cex=0.5, col="#6633BB55")
plot (x, y, col=densCols(x,y), pch=20)

 

 

Further reading and resources