Direct labels puts labels at wrong place

Question

I'm trying to do a plot with ggplot2 and trying to put labels at the right place on mean trajectories with confidence regions. Since I cannot share my data, I created a reproducible example:

set.seed(456)
library(ggplot2)
library(directlabels)
time<-1:25
df2<-data.frame(time,value=runif(2500),group=as.factor(c(rep('Alpha',1250),rep('Beta',1250))))
ggplot(df2, aes(x=time, y=value,group=group,colour=group))+
      stat_summary(geom="ribbon", fun.data=mean_cl_normal,fun.args=list(conf.int=0.95), fill="grey")+
      stat_summary(geom="line", fun=mean,size=2,show.legend=TRUE)+
      stat_summary(geom="point", fun=mean, color="red",size=3)+
      theme(legend.position='none',axis.title=element_text(size=40),
        axis.text=element_text(size=30),panel.background = element_blank(),axis.line = 
        element_line(colour = 'black'))+
      geom_dl(aes(label = group),method=list(dl.combine('first.qp','last.qp')),color='black')

I obtain the following plot:

As you can see, the labels are completely wrong. I'm trying to put the labels next to the first mean point and/or next to the last mean point. With my data, the result is worse: both labels are under the lower trajectory and both overlapped. I also get the following warning:

┌ Warning: RCall.jl: Warning in regularize.values(x, y, ties, missing(ties)) :
│   collapsing to unique 'x' values
│ Warning in regularize.values(x, y, ties, missing(ties)) :
│   collapsing to unique 'x' values
│ Warning in regularize.values(x, y, ties, missing(ties)) :
│   collapsing to unique 'x' values
│ Warning in regularize.values(x, y, ties, missing(ties)) :
│   collapsing to unique 'x' values

How can I solve my problem and put the labels next to the first/last mean values of each trajectory? I tried to change the method option but with no success... As the plotted points are means, the methods with .points don't work. Moreover, how can I control the size of labels?

I tried another ggplot2code:

ggplot(df2, aes(x=time, y=value,group=group,colour=group))+
       stat_summary(geom="ribbon", fun.data=mean_cl_normal,
                fun.args=list(conf.int=0.95), fill="grey")+
       stat_summary(geom="line", fun=mean,size=2,show.legend=TRUE)+
       stat_summary(geom="point", fun=mean, color="red",size=3)+
       theme(legend.position = 'none',axis.title=element_text(size=40),
                axis.text=element_text(size=30),panel.background = 
                element_blank(),axis.line = element_line(colour = 
                'black'))+
       stat_summary(aes(label=group), fun=mean, geom="text", size=8)

and I obtain the following plot:

I think I'm close with this last plot, if I can keep only the first and/or last labels.

You can file a bug report here: https://github.com/tdhock/directlabels/issues — M Turgeon, Jul 29 '20 at 23:40
@coolsv: do these help https://stackoverflow.com/a/53346829/786542 & https://stackoverflow.com/a/49602796/786542 & https://stackoverflow.com/a/50807185/786542? — Tung, Jul 30 '20 at 03:42
Seem too complicated to me... I tried to get the ideas of `ggrepel` but get awful results with the labels spread everywhere instead of just being plotted at the end of the line. I don't want to modify all my code since I took several hours before getting the above plots... A total nightmare just to put labels at the end of a line :-( — coolsv, Jul 30 '20 at 04:33

score 1 · Accepted Answer · answered Jul 30 '20 at 08:49

You were already on the right track with your last approach. To make it work you have to filter your df to include only the obs for the the first and last time. Try this:

set.seed(456)
library(ggplot2)
library(dplyr)

time<-1:25
df2<-data.frame(time,value=runif(2500),group=as.factor(c(rep('Alpha',1250),rep('Beta',1250))))

dlabs <- df2 %>%
  group_by(group) %>% 
  arrange(time) %>% 
  filter(time %in% c(first(time), last(time)))

ggplot(df2, aes(x=time, y=value,group=group,colour=group))+
  stat_summary(geom="ribbon", fun.data=mean_cl_normal,fun.args=list(conf.int=0.95), fill="grey")+
  stat_summary(geom="line", fun=mean,size=2,show.legend=TRUE)+
  stat_summary(geom="point", fun=mean, color="red",size=3)+
  theme(legend.position='none',axis.title=element_text(size=40),
        axis.text=element_text(size=30),panel.background = element_blank(),axis.line = 
          element_line(colour = 'black'))+
  stat_summary(data = dlabs, aes(label = group, hjust = ifelse(time == first(time), 1.2, -.2)), fun = mean, geom = "text", color = "black") +
  scale_x_continuous(expand = expansion(mult = .1))

Direct labels puts labels at wrong place

1 Answers1

Linked