How to assign colors using a variable containing color names in a grouped ggplot?

Question

In this simple example, I create a variable with the names of colors.

df <- mtcars %>%
  mutate(color = "green",
     color = replace(color, cyl==6, "blue"),
     color = replace(color, cyl==8, "red"))

Running the code below works as expected.

ggplot(df, aes(wt, mpg)) +
  geom_point(color = df$color)

What if I want to use geom_line to create three lines--green, blue, and red?

ggplot(df, aes(wt, mpg, group=cyl)) +
  geom_line(color = df$color)

Instead, I get three lines with the colors cycling throughout.

How can I use a variable with color names to assign the color of different lines?

Does this answer your question? [How to conditionally highlight points in ggplot2 facet plots - mapping color to column](https://stackoverflow.com/questions/15804504/how-to-conditionally-highlight-points-in-ggplot2-facet-plots-mapping-color-to) — tjebo, Jul 01 '20 at 13:43

score 5 · Accepted Answer · answered Oct 22 '18 at 17:06

5

I think you are looking for scale_color_identity

ggplot(df, aes(wt, mpg)) +
  geom_point(aes(color = color)) +
  scale_color_identity(guide = "legend") # default is guide = "none"

Here is the respective line plot

ggplot(df, aes(wt, mpg)) +
  geom_line(aes(color = color)) +
  scale_color_identity(guide = "legend")

answered Oct 22 '18 at 17:06

markus

25,843
5
39
58

1

Very nice. scale_identity() was my missing step. Much appreciated. – John J. Oct 22 '18 at 18:57

score 0 · Answer 2 · answered Oct 22 '18 at 17:01

0

You can use a custom color scale:

ggplot(df, aes(wt, mpg, group=cyl)) +
    geom_line(aes(color = color)) +
    scale_color_manual(values = c("blue"="blue","red"="red","green"="green"))

answered Oct 22 '18 at 17:01

Richard J. Acton

885
4
17

bob1 · Answer 3 · 2018-10-22T17:10:32.870

Short answer: you can't. You have set the variable to set the colors in the variable you created.

However, there is a way to do this in ggplot by default:

mtcars$cyl <-as.factor(mtcars$cyl) ## set mtcars$cyl as factors (i.e use exact values in column)

ggplot(mtcars, aes(x=wt, y= mpg, color = cyl)) +
       geom_point()+
       scale_color_manual(breaks = c("8", "6", "4"),
                    values=c("red", "blue", "green"))+ ## adjust these if you want different colors
       geom_line()

Left out the lines bit...

How to assign colors using a variable containing color names in a grouped ggplot?

3 Answers3