How to append lines in a plot with ggplot2

Question

I am trying to make a plot like this in an easy way with ggplot (or base R)

The idea is to consider the dark line as a single group with a gap between two subgroups (CAP and NC), and also to put a second x lab to tag times (as you can see, times 0,15 and 240 are common to both groups).

Thanks for your help,

EDIT: Here there is an example.

db0 <- structure(list(t = c(0, 5, 15, 30, 60, 0, 15, 60, 0, 5, 15, 
30,60, 0, 15, 60), g = c("C", "C", "C", "C", "C", "N", "N", "N", 
"C", "C", "C", "C", "C", "N", "N", "N"), v = c("T1", "T1", "T1", 
"T1", "T1", "T1", "T1", "T1", "T2", "T2", "T2", "T2", "T2", "T2", 
"T2", "T2"), y = c(10, 12, 15, 25, 33, 10, 13, 11, 11, 20, 28, 
14, 10, 11, 11, 12), x = c(1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 1L, 
2L, 3L, 4L, 5L, 6L, 7L, 8L)), row.names = c(NA, -16L), class = 
"data.frame")

db0$x <- rep(1:8, length.out=16)

library(ggplot2)
ggplot(db0, aes(x=as.factor(x), y=y, group=g, color=v)) + 
  geom_path() + 
  geom_point() +
  scale_x_discrete(labels=c("1" = "0",  "2" = "5", "3" = "15", "4" = "30",
                            "5" = "60", "6" = "0", "7" = "15", "8" = "60"))

The question here is how to do this plot in an easy way, without creating the x variable and the changing the x-lab names, but with ggplot2 commands. Also, Having problems with geom_path that links the end of one group with the beginning of the following.

What is your actual question? What about some toy data for us and the code you were runnig so far? — deschen, Dec 26 '20 at 19:05
Example aded. I also forgot to ask how to include the second x labels (NC and CAP in the first graph, and would be C and N for the example). Thanks for your help. — Manuel Ramón, Dec 26 '20 at 19:47
Why is the visual width of 0-60 on the left so much wider than 0-60 on the right? — r2evans, Dec 26 '20 at 20:50
The idea behind this plot is to reduce space. This is why the space between 0 and 60 was reduced. I understand this is probably not a right way to draw the plot, but I am doing that as an exercise to improve my ggplot2 skills. I think I can prepare the data (it would take a lot of time) so ggplot2 could draw the plot, but I would like to know if there is an elegant way to do it — Manuel Ramón, Dec 26 '20 at 22:00

score 0 · Answer 1 · answered Dec 26 '20 at 20:53

Use group = interaction(g, v) to prevent the lines from connecting.
I think you can use facet_grid to get similar results.

BTW, I'm using x=t here to get the actual value for the x-axis instead of attempting to map it with ordinals and such.

ggplot(db0, aes(x=t, y=y, group=interaction(g, v), color=v)) + 
  geom_path() + 
  geom_point() +
  facet_grid(. ~ g, space = "free_x")

I added space="free_x" here to attempt to get the x-axis on the right to compress, which is when I noticed that the x-range is the same for the left and the right ... theoretically, if they were different, the size of each pane might be proportional. (See https://stackoverflow.com/a/10455381/3358272)

There is a bit of formatting/labeling you'll need to do to get the "CAP" and "NC" labels you seek.

If you need similar x-ranges but dissimilar widths, then you will likely need to munge the x-axis of one of the groups and then attach a labeller somehow that distinguishes between the two groups.

A third option is to produce two separate plots (using different x-axis controls) and use a package like patchwork to combine them with dissimilar widths.

Thanks @r2evans for your answer. I know that using facet_grid I can get almost what I asking for, but in this case I prefer to get both lines in the same graph. The idea is to reproduce the first graph using ggplot2 instead of basic plots. Also, thank you for the interaction tip. — Manuel Ramón, Dec 26 '20 at 21:47

How to append lines in a plot with ggplot2

1 Answers1