Ploting with ggplot

Learning objectives

• Questions:

  • How to plot in R using ggplot2 package

• Objectives:

  • Learn advanced plotting using ggplot2

• Keypoints:

  • ggplot2

ggplot

• ggplot2 is a graphics package, written by Hadley Wickham, a grad student at Iowa State, based on the ideas from the book “Grammar of Graphics” by Leland Wilkinson. Let’s install it (it will install multiple additional packages that it requires):

install.packages("ggplot2")
library(ggplot2)

Basic component of ggplot

• A data frame

• aes: aesthetic mappings showing how data are mapped to color, size

• geoms: geometric objects like points, lines, shapes.

• facets: for conditional plots.

• stats: statistical transformations like binning, quanti les, smoothing.

• scales: what scale an aesthetic map uses

• coordinate system

Type of ggplot

• Basic qplot

  • Same as plot in Base plot

  • Nicer graphics than Base plot

  • Difficult for customize

• Advanced ggplot

  • Flexible with many built-in function

• A quick way to get familiar with ggplot2 is the qplot function, which stands for quick plot. Let’s do a quick scatter plot from the iris dataset, plotting sepal length versus petal length:

qplot(Sepal.Length, Petal.Length, data=iris)

• Notice that it already looks nicer than the basic R plots we did in the last chapter. Now, let’s plot different species of iris with different colors and shapes:

qplot(Sepal.Length, Petal.Length, data=iris,
      color=factor(Species),
      shape=factor(Species))

• Now, let’s add smooth lines which show trends in the data:

qplot(Sepal.Length, Petal.Length, data=iris,
      color=factor(Species),
      shape=factor(Species),
      geom=c("point","smooth"))

• Let’s make these lines straight – that is, let’s fit a linear model to each species’ data:

qplot(Sepal.Length, Petal.Length, data=iris,
      color=factor(Species),
      shape=factor(Species),
      geom=c("point","smooth"), method="lm")

• Let’s make a density plot (a smoothed histogram) of sepal lengths of each species:

qplot(Sepal.Length,data=iris,geom="density",
      color=Species)

• There are many more ways to use ggplot2. Some useful (and beautiful) examples of code are here: http://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html

Basic qplot: Facets

qplot(Sepal.Length,Petal.Length,facets=.~Species, data=iris)

Advanced ggplot

• Sample plot

gp <- ggplot(mpg, aes(hwy, cty))

gp+geom_point(aes(color=cyl))
gp+geom_point(aes(color=factor(cyl)))
gp+geom_point(aes(color=factor(cyl)))+geom_smooth(method="lm")
gp+geom_point(aes(color=factor(cyl)))+geom_smooth(method="lm")
  +facet_grid(.~cyl)
# Save plot to file
ggsave("plot.png",width=5,height=5)

Annotation

• Labels: xlab(), ylab(), labs(), ggtitle()

• global annotation: use theme()

• Standard appearance: theme_bw()

gp+geom_point(aes(color=factor(cyl),
              size=factor(cyl)))+
  geom_smooth(method="lm")+
  xlab("Highway miles per gallon")+
  ylab("city miles per gallon")+
  ggtitle("Scatter plot for cty & hwy")+
  xlim(10,40)+ylim(10,40)+
  theme_bw(base_size = 15)

Some nice ggplots featuring

• Boxplot

ggplot(mpg,aes(x=manufacturer,y=hwy,
               fill=factor(manufacturer)))+
  geom_boxplot()+
  geom_jitter()+
  labs(title="Boxplot for Hwy per manufacturer",x="Manufacturer",y="Highway milage")+
  theme_bw()+coord_flip()+
  theme(legend.position = "none")

• Violin plot

g <- ggplot(mpg, aes(class, cty))
g + geom_violin(aes(fill=class)) + 
  labs(title="Violin plot", 
       subtitle="City Mileage vs Class of vehicle",
       caption="Source: mpg",
       x="Class of Vehicle",
       y="City Mileage")

• Histogram

g <- ggplot(mpg, aes(displ)) + scale_fill_brewer(palette = "Spectral")
g + geom_histogram(aes(fill=class), 
                   bins=10, 
                   col="black", 
                   size=.1) +   # change number of bins
  labs(title="Histogram with Fixed Bins", 
       subtitle="Engine Displacement across Vehicle Classes",
       x="enginer displacement (m)",
       y="Frequency count") 

• Scatter plot

data("midwest")
gg <- ggplot(midwest, aes(x=area, y=poptotal)) + 
  geom_point(aes(col=state, size=popdensity)) + 
  geom_smooth(method="loess", se=F) + 
  xlim(c(0, 0.1)) + 
  ylim(c(0, 500000)) + 
  labs(subtitle="Area Vs Population", 
       y="Population", 
       x="Area", 
       title="Scatterplot", 
       caption = "Source: midwest")
plot(gg)

• Density

g <- ggplot(mpg, aes(cty))
g + geom_density(aes(fill=factor(cyl)), alpha=0.8) + 
    labs(title="Density plot", 
         subtitle="City Mileage Grouped by Number of cylinders",
         caption="Source: mpg",
         x="City Mileage",
         fill="# Cylinders")+
    theme_bw()

• Density 2D

gg <- ggplot(faithful,aes(x=eruptions,y=waiting))
gg + stat_density_2d(aes(fill=..level..),
                     geom="polygon",color="black")+
     geom_smooth(method="lm",linetype=2,color="red")+
     scale_fill_continuous(low="green",high="red")+
     geom_point() +
     theme_bw()

• Geographic visualization with ggplot

library(maps)
states <- map_data("state")
ggplot(data = states)+
  geom_polygon(aes(x=long,y=lat,fill=region),
               color="black")+
  coord_fixed(1.3)+
  guides(fill=FALSE)

counties <- map_data("county")
SC_counties <- subset(counties,region == "south carolina")
ggplot(data = SC_counties)+
  geom_polygon(aes(x=long,y=lat,fill=subregion),
               color="black")+
  coord_fixed(1.3)+
  guides(fill=FALSE)

some.eu.countries <- c(
  "Portugal", "Spain", "France", "Switzerland", "Germany",
  "Austria", "Belgium", "UK", "Netherlands",
  "Denmark", "Poland", "Italy", 
  "Croatia", "Slovenia", "Hungary", "Slovakia",
  "Czech republic"
)
# Retrievethe map data
some.eu.maps <- map_data("world", region = some.eu.countries)

ggplot(some.eu.maps, aes(x = long, y = lat)) +
  geom_polygon(aes( group = group, fill = region))+
  scale_fill_viridis_d()+
  theme_void()+
  theme(legend.position = "none")

• Plot Shapefile for geography study

  • Download shape file data here

  • Store it in your folder: c:/R/GIS/ in Windows or /user/R/GIS in MacOS

  • Unzip it and rename all files to Countries_WGS84.* under C:/GIS/

• Install additional packages:

install.packages("rgdal")
install.packages("colorspace")

• Perform plotting

library(rgdal)
library(colorspace)
library(maps)

setwd('c:/R/GIS/')
gfile <- readOGR(dsn="Countries_WGS84.shp")
names(gfile)
gfile$CNTRY_NAME

plot(gfile)
plot(gfile,col=rainbow_hcl(50))
llgridlines(gfile,lty=5)

• Plot raster

  • Here we will plot a raster data base using Global land cover data set. The data can be downloaded from here.

• Unzip and put the raster data to working directory:

install.packages("raster")
library(raster)
library(rgdal)

setwd('c:/R/GIS/')
#import raster
Gcover <- raster("GLOBCOVER_L4_200901_200912_V2.3.tif")
#plot raster
plot(Gcover,main="GLobal Land cover")