How to add different lines for facets

Question

I have data where I look at the difference in growth between a monoculture and a mixed culture for two different species. Additionally, I made a graph to make my data clear.

I want a barplot with error bars, the whole dataset is of course bigger, but for this graph this is the data.frame with the means for the barplot.

plant           species     means
Mixed culture   Elytrigia   0.886625
Monoculture     Elytrigia   1.022667
Monoculture     Festuca     0.314375
Mixed culture   Festuca     0.078125

With this data I made a graph in ggplot2, where plant is on the x-axis and means on the y-axis, and I used a facet to divide the species.

This is my code:

    limits <- aes(ymax = meansS$means + eS$se, ymin=meansS$means - eS$se)
    dodge <- position_dodge(width=0.9)

    myplot <- ggplot(data=meansS, aes(x=plant, y=means, fill=plant)) + facet_grid(. ~ species) 
    myplot <- myplot + geom_bar(position=dodge) + geom_errorbar(limits, position=dodge, width=0.25)
    myplot <- myplot + scale_fill_manual(values=c("#6495ED","#FF7F50"))
    myplot <- myplot + labs(x = "Plant treatment", y = "Shoot biomass (gr)")  
    myplot <- myplot + opts(title="Plant competition")
    myplot <- myplot + opts(legend.position = "none")
    myplot <- myplot + opts(panel.grid.minor=theme_blank(), panel.grid.major=theme_blank())

So far it is fine. However, I want to add two different horizontal lines in the two facets. For that, I used this code:

    hline.data <- data.frame(z = c(0.511,0.157), species = c("Elytrigia","Festuca")) 
    myplot <- myplot + geom_hline(aes(yintercept = z), hline.data)

However if I do that, I get a plot were there are two extra facets, where the two horizontal lines are plotted. Instead, I want the horizontal lines to be plotted in the facets with the bars, not to make two new facets. Anyone a idea how to solve this.

I think it makes it clearer if I put the graph I create now:

enter image description here

No I want just a horizontal line, at a given y coordinate. However the value are different for each facet, eg facet Elytrigia has a horizontal line at 0.511 and facet Festuca a horizontal line at 0.157. — Marinka, Aug 07 '12 at 12:55
May I suggest you cut and paste your data set exactly as is from the console or use `dput(means)` to give us your data. — Tyler Rinker, Aug 07 '12 at 12:56

score 84 · Accepted Answer · answered Aug 07 '12 at 13:36

84

Make sure that the variable species is identical in both datasets. If it a factor in one on them, then it must be a factor in the other too

library(ggplot2)
dummy1 <- expand.grid(X = factor(c("A", "B")), Y = rnorm(10))
dummy1$D <- rnorm(nrow(dummy1))
dummy2 <- data.frame(X = c("A", "B"), Z = c(1, 0))
ggplot(dummy1, aes(x = D, y = Y)) + geom_point() + facet_grid(~X) + 
    geom_hline(data = dummy2, aes(yintercept = Z))

enter image description here

dummy2$X <- factor(dummy2$X)
ggplot(dummy1, aes(x = D, y = Y)) + geom_point() + facet_grid(~X) + 
    geom_hline(data = dummy2, aes(yintercept = Z))

enter image description here

answered Aug 07 '12 at 13:36

Thierry

18,049
5
48
66

Oh yeah, now I remembered that there was a space behind my species name, how annoying. At least there was nothing wrong with my code – Marinka Aug 07 '12 at 13:39
Doesn't seem to work for `geom_abline`, or did I mess something up? See my question http://stackoverflow.com/questions/43696806/r-ggplot-different-geom-ablines-in-facets – Boern Apr 29 '17 at 14:28
You need to specify the `data = dummy2` in the `geom_line`. Otherwise you will get the error `Error: mapping must be created by aes()`. – user3507584 May 30 '19 at 09:19

score 0 · Answer 2 · answered Apr 02 '23 at 08:41

Another option using geom_abline where you specify the intercept as your position of the horizontal line and you have to make sure the slope is 0 in your aesthetics. Here is some reproducible code (slightly adjusted from @Thierry):

library(ggplot2)
dummy1 <- expand.grid(X = factor(c("A", "B")), Y = rnorm(10))
dummy1$D <- rnorm(nrow(dummy1))
dummy2 <- data.frame(X = c("A", "B"), Z = c(1, 0))
ggplot(dummy1, aes(x = D, y = Y)) + 
  geom_point() + 
  facet_grid(~X) + 
  geom_abline(data = dummy2, aes(intercept = Z, slope = 0))

^{Created on 2023-04-02 with reprex v2.0.2}

As you can see the result is the same when using geom_hline.

How to add different lines for facets

2 Answers2

Linked

Related