GGplotTutorial.R

library(ggplot2)
##Here is the data
data(iris)


## Being Hasty? 
plot(iris$Sepal.Length, iris$Species)
plot(iris$Species, iris$Sepal.Length)

#Look at the Data
summary(iris)

#Base Plot; 
##Pros: quick, simple X-Y, simple colors
##Cons: simple, less than beautiful

plot(iris$Sepal.Width, iris$Petal.Length)
plot(iris$Sepal.Width, iris$Petal.Length, col='pink')
plot(iris$Sepal.Width, iris$Petal.Length, col='pink', pch=15)
plot(iris$Sepal.Width, iris$Petal.Length, col=iris$Species, pch=15)

class(iris$Species)
iris$Species
class(iris$Species)
str(iris)

quartz()

plot(
  iris$Sepal.Width,
  iris$Petal.Length,
  col = as.numeric(iris$Species),
  pch = 15,
  xlab = "Sepal Width",
  ylab = "Petal Length"
)


## Base Plot lets you add things onto the original plot, all within the plotting
## window. In general, this is great. But it can be a problem if you want to add
## or extract or edit deeper pieces

legend(
  "right",
  legend = levels(iris$Species),
  col = 1:length(levels(iris$Species)),
  pch = 15
)

lines(iris$Sepal.Width,
      iris$Petal.Length,type = 'p' )

title(main='Sepal vs Petal by Species') 
## But. instead, lets let the amazing magicians over at ggplot do it for us!



library(ggplot2)

## The first call, 'ggplot( ... )' is to set up the 'window' which you will plot
## in 
ggplot()

ggplot(data = iris, aes())

## WHAT?? NOTHING HAPPENED?

## Why yes, thats simply because we have to make it 'aesthetic;)'
ggplot(data = iris, aes(Sepal.Width, Petal.Length))

## Something almost happened. Can you tell? 


## Now we just need to add what we want the plot to DO. Scatter? Line? Box?

ggplot(data = iris,aes(Sepal.Width, Petal.Length))+geom_point()

## But kind sir, that is no different than the original

## I beg to differ; 

ggplot(data = iris,aes(x=Sepal.Width, y=Petal.Length, colour = Species))+geom_point()

# Look how easy it is to add fun things! all via AES (cant spell EASY without AES)
ggplot(data = iris,aes(x=Sepal.Width, y=Petal.Length, colour = Species))+
  geom_point()+geom_smooth()

ggplot(data = iris,aes(y=Sepal.Width, fill = Species, group = Species))+
  geom_boxplot()


#Now lets do something fun


## WARNING. WHICHES AHEAD
iris2=iris
iris2$SepalCat = NA
iris2$SepalCat[which(iris2$Sepal.Width>=3.25)]='Large'
iris2$SepalCat[which(iris2$Sepal.Width<3.25)]='Small'
table(iris2$SepalCat)

ggplot(data=iris2, aes(x=SepalCat,y=Petal.Length, fill = Species))+
  geom_boxplot()+
  ggpubr::stat_compare_means(method = 't.test',label = 'p.signif',
                             comparisons = list(c("Large","Small")))
                             



#####
# OWN DATA




library(ggpubr)
ggplot(data = iris,aes(x=Species, y=Sepal.Width, fill = Species, group = Species))+
  geom_boxplot()+
  ggtitle(label = "Boxplot Prettyness")+
  theme(plot.title = element_text(face='bold', hjust = 0.5))+
  scale_fill_manual(values = c('lightgoldenrod', 'salmon2', '#11E700'))+
  ggpubr::stat_compare_means(method = 't.test',label = 'p.signif',
                            comparisons = list(c("setosa","versicolor"),
                                               c("setosa", "virginica"),
                                               c("virginica", "versicolor")))
?ggpubr::stat_compare_means
?ggpubr::stat_pvalue_manual()

p_vals=data.frame('group1'=c('setosa', 'setosa', 'virginica'),
                  'group2'=c('versicolor', 'virginica', 'versicolor'),
                  'p'=c('***', 'N---S', 'NS'),'y.position'=c(5,6,4))


ggplot(data = iris,aes(x=Species, y=Sepal.Width, fill = Species, group = Species))+
  geom_boxplot()+
  ggtitle(label = "Boxplot Prettyness")+
  theme(plot.title = element_text(face='bold', hjust = 0.5))+
  ggpubr::stat_pvalue_manual(data=p_vals)+ggpubr::theme_pubr()








iris2$Species_SepalCat <- interaction(
  iris2$Species,
  iris2$SepalCat,
  sep = "_"
)

ggplot(iris2,
               aes(x = Species_SepalCat,
                   y = Petal.Length,
                   fill = Species)) +
          geom_boxplot() +
  stat_compare_means(
              comparisons = list(
                c("setosa_Small", "versicolor_Small"),
                c("setosa_Large", "setosa_Small")
              ),
              method = "wilcox.test",
              label = "p.signif"
            )+
  geom_boxplot() + 
  scale_x_discrete(
            labels = function(x) {
              ifelse(grepl("versicolor_Small", x), "Small",
                     ifelse(grepl("versicolor_Large", x), "Large", ""))
            }
          )





# Thats great for categorical, but maybe not so much for contionous: 


ggplot(data = iris,aes(y=Petal.Width, x=Sepal.Width, shape = Species))+
  geom_point(aes(colour=Petal.Length))+
  scale_colour_gradientn(colours = c('blue', 'white','red'))
  
ggplot(data = iris,aes(y=Petal.Width, x=Sepal.Width, shape = Species))+
  geom_point(aes(color=Petal.Length), size=3)+
    scale_color_gradient(low = 'chartreuse', high = 'violetred1')+
    scale_shape_manual(values = c(15,16,17))
colors()

"mediumvioletred"
"linen"
"#F0F"
"mediumaquamarine"
"#A33000"
"lavenderblush"