legend for group of lines in ggplot - r

Question

I am trying to plot a group of lines and assign one label to these lines in the legend. Here's an example of the data (df2) that I am using.
structure(list(
true = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L),
.Label = c("model1", "model2"), class = "factor"),
test = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L),
.Label = c("false.model1", "false.model2", "false.model3", "true model"), class = "factor"),
est.interval = c(0, 5, 5, 7, 7, 10, 10, 11, 11, 0, 2, 2, 3, 3, 0, 2, 2, 4, 4, 0, 2, 2, 3, 3, 4, 4, 0, 7, 7, 10, 10, 13, 13, 0, 4, 4, 5, 5, 0, 5, 5, 6, 6, 7, 7, 9, 9, 0, 2, 2, 3, 3, 4, 4),
sens = c(1, 1, 0.75, 0.75, 0.5, 0.5, 0.25, 0.25, 0, 1, 1, 0.25, 0.25, 0, 1, 1, 0.5, 0.5, 0, 1, 1, 0.5, 0.5, 0.25, 0.25, 0, 1, 1, 0.75, 0.75, 0.25, 0.25, 0, 1, 1, 0.5, 0.5, 0, 1, 1, 0.75, 0.75, 0.5, 0.5, 0.25, 0.25, 0, 1, 1, 0.5, 0.5, 0.25, 0.25, 0)),
.Names = c("true", "test", "est.interval", "sens"),
class = "data.frame")
And here I plot the curves using ggplot.
ggplot(df2, aes(x=est.interval, y=sens, color=test)) +
ylim(0,1) +
geom_line(size = 0.6) +
facet_wrap(~true, nrow=1) +
scale_colour_manual(values=c(rep('#CCCCCC', 3), "#000000"), name="Estimation Model") +
guides(color=guide_legend(keywidth = 3, keyheight = 1))
I'd like to have the three grey "false.model" curves separately visible on the plot, but in the legend I just want one "false models" entry beside one grey line. Any thoughts on how to do this?

Answer 1

I had a problem with your dput. It seemed to be missing rownames which made it an invalid data.frame. Here a corrected dump
df2 <- structure(list(true = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("model1",
"model2"), class = "factor"), test = structure(c(1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L,
2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L,
4L, 4L, 4L), .Label = c("false.model1", "false.model2", "false.model3",
"true model"), class = "factor"), est.interval = c(0, 5, 5, 7,
7, 10, 10, 11, 11, 0, 2, 2, 3, 3, 0, 2, 2, 4, 4, 0, 2, 2, 3,
3, 4, 4, 0, 7, 7, 10, 10, 13, 13, 0, 4, 4, 5, 5, 0, 5, 5, 6,
6, 7, 7, 9, 9, 0, 2, 2, 3, 3, 4, 4), sens = c(1, 1, 0.75, 0.75,
0.5, 0.5, 0.25, 0.25, 0, 1, 1, 0.25, 0.25, 0, 1, 1, 0.5, 0.5,
0, 1, 1, 0.5, 0.5, 0.25, 0.25, 0, 1, 1, 0.75, 0.75, 0.25, 0.25,
0, 1, 1, 0.5, 0.5, 0, 1, 1, 0.75, 0.75, 0.5, 0.5, 0.25, 0.25,
0, 1, 1, 0.5, 0.5, 0.25, 0.25, 0)), .Names = c("true", "test",
"est.interval", "sens"), row.names = c(NA, -54L), class = "data.frame")
The easiest thing to do would be to create a new factor that you can use for the coloring. For example
df2$testcol<-factor(ifelse(df2$test=="true model",1,2),
levels=1:2,
labels=c("True Model","False Model"))
Now all the false models will share a common value. Then you can do
ggplot(df2, aes(x=est.interval, y=sens, group=test, colour=testcol)) +
ylim(0,1) +
geom_line(size = 0.6) +
facet_wrap(~true, nrow=1) +
scale_colour_manual(values=c("True Model"='#CC0000',
"False Model"="#999999"), name="Estimation Model") +
guides(color=guide_legend(keywidth = 3, keyheight = 1))
to get