if else condition in ggplot to add an extra layer

Question

say I want to plot two layers in ggplot, one containing points and another one containing lines if a certain criteria is fulfilled.

The code without the criteria could look like this:

library("ggplot2")

# Summarise number of movie ratings by year of movie
mry <- do.call(rbind, by(movies, round(movies$rating), function(df) {
  nums <- tapply(df$length, df$year, length)
  data.frame(rating=round(df$rating[1]), year = as.numeric(names(nums)), number=as.vector(nums))
}))

p <- ggplot(mry, aes(x=year, y=number, group=rating))

p + 
geom_point()+
geom_line()

now the condition for plotting the points and not only the lines would be, that an object called tmp.data does not equal the expression "no value".

tmp.data<-c(1,2,3) # in this case the condition is fulfilled

# attempt to plot the two layers including the condition in the plotting function
p+ 
  if(tmp.data[1]!="no value"){ geom_point()+}
  geom_line()

fails....

Error: unexpected '}' in:
"p+ 
if(tmp.data[1]!="no value"){ geom_point()+}"

geom_line() geom_line:
stat_identity:
position_identity: (width = NULL, height = NULL)

Why not just change the order: `p + geom_line() + if(tmp.data[1]!="no value"){ geom_point()}` — shadow, Apr 07 '14 at 14:40
@shadow. this works only if my second layer is the last argument in the function. if I add further arguments like: `+ theme(panel.background = element_rect(fill = "lightskyblue2"))` it does not work anymore — joaoal, Apr 07 '14 at 14:54

score 92 · Answer 1 · answered May 23 '16 at 18:20

92

This was done using ggplot2 2.1.0. I think you can do exactly what the OP wished, just by switching the parenthesis so that they encompass the entire if statement.

Here is an example that add a horizontal line depending on if Swtich is T or F. First, where the condition is TRUE

library(ggplot2)

df<-data.frame(x=1:10,y=11:20)
Switch=T

ggplot(df,aes(x,y))+
{if(Switch)geom_hline(yintercept=15)}+
  geom_point()

Now, the same thing but the condition is FALSE

df<-data.frame(x=1:10,y=11:20)
Switch=F

ggplot(df,aes(x,y))+
{if(Switch)geom_hline(yintercept=15)}+
  geom_point()

answered May 23 '16 at 18:20

John Paul

12,196
6
55
75

2

it seems that you cannot use the `+` inside of the `{`, so if you want to conditionally add multiple things to a ggplot, I believe you need to add another `if` statement (inside of its own `{`) for each object you add to the plot. – filups21 Jan 08 '18 at 15:54
26

You cannot use `+` inside the `{`, but you can put everything into a `list` if you have multiple steps. `{if(Switch)list(geom_hline(yintercept=15), geom_hline(yintercept = 13))}` – StephenK Jun 24 '20 at 19:07
@StephenK in a million years, I would never have intuited that this was the right way to execute multiple steps inside `{}`, but thankfully I was able to find your comment! – C. Murtaugh Feb 13 '23 at 02:46

score 54 · Accepted Answer · answered Apr 07 '14 at 14:51

54

What you are seeing is a syntax error. The most robust way I can think of is:

tmp.data<-c(1,2,3) 
if(tmp.data[1]!="no value") {
   p = p + geom_point()
}
p + geom_line()

So you compose the object p in a sequence, only adding geom_point() when the if statements yields TRUE.

answered Apr 07 '14 at 14:51

Paul Hiemstra

59,984
12
142
149

1

this solution doesn't work if there trying to facet_wrap. I get the error: Error in seq.default(min, max, by = by) : 'from' must be a finite number – val Oct 16 '18 at 21:49

score 10 · Answer 3 · answered Feb 25 '20 at 14:34

Following the ggplot2 book, you can create a function which returns a list. Any NULL components will be ignored.

library(ggplot2)
library(ggplot2movies)

# Summarise number of movie ratings by year of movie
mry <- do.call(rbind, by(movies, round(movies$rating), function(df) {
    nums <- tapply(df$length, df$year, length)
    data.frame(rating=round(df$rating[1]), year = as.numeric(names(nums)), number=as.vector(nums))
}))

# create function to add points conditionally
# If the list contains any NULL elements, they’re ignored. 
my_plot <- function(point = FALSE){
    list(
        geom_line(),
        if (point) 
            geom_point()
    )

}

p <- ggplot(mry, aes(x=year, y=number, group=rating))
p + my_plot()

p + my_plot(point = TRUE)

^{Created on 2020-02-25 by the reprex package (v0.3.0)}

score 5 · Answer 4 · edited Feb 08 '17 at 14:52

library(ggplot2)

# Summarise number of movie ratings by year of movie
mry <- do.call(rbind, by(movies, round(movies$rating), function(df) {
  nums <- tapply(df$length, df$year, length)
  data.frame(rating=round(df$rating[1]), year = as.numeric(names(nums)), number=as.vector(nums))
}))

tmp.data<-c(1,2,3) # in this case the condition is fulfilled

p <- ggplot(mry, aes(x=year, y=number, group=rating))

# this won't "loop through" the data points but it's what you asked for
if (tmp.data[1]!="no value") {
  p <- p + geom_point() + geom_line()
} else {
  p <- p + geom_line()
}
p

but perhaps this is more like what you really want?

mry$rating <- factor(mry$rating)
p <- ggplot(mry, aes(x=year, y=number, group=rating))
p <- p + geom_line()
p <- p + geom_point(data=mry[!(mry$rating %in% tmp.data),], 
                    aes(x=year, y=number, group=rating, color=rating), size=2)
p <- p + scale_color_brewer()
p

score 0 · Answer 5 · answered Jul 15 '22 at 05:45

An alternative example using iris dataset creating a plot with conditional components, some with more than one line, and with multiple switches.

library("ggplot2")

p <- ggplot(iris, 
            aes(x=Sepal.Length, 
                y=Sepal.Width, 
                colour=Species))

# set plot switches to turn on or off various components
include_points = TRUE
include_lines = FALSE
include_density = TRUE
include_facet = TRUE

p + 
  list(
    if(include_points){geom_point()},
    if(include_lines){geom_line()},
    if(include_facet){
      
      # multi-line conditional elements can be included as a list
      list(
        facet_grid(rows = "Species"),
        labs(subtitle = "This plot has been faceted, and colour has been removed"),
        aes(colour = NULL)
      )
      
    },
    if(include_density){geom_density2d()}) +
  
  # additional elements common to all possible conditional combinations
  # can be added as well
  theme_minimal() + 
  labs(title = "Plot with conditional elements")

With the conditions as follows:

include_points = TRUE
include_lines = FALSE
include_density = TRUE
include_facet = TRUE

The following plot is generated:

Faceted plot with density and points

Setting include_facet = FALSE the following plot is generated:

Coloured plot with density - no facets

if else condition in ggplot to add an extra layer

5 Answers5

Linked

Related