I am running through a large dataset and creating a standard set of plots for a large number of species in that dataset. The premise is quite simple. For some species there will be data to create all the plots.

#produce some data
samples <- data.frame(
  GROUP = c("dog","dog","dog","cat","cat","cat"),
  VAR1 = c(1,3,5,1,3,5),
  VAR2   = c(2,4,6,2,4,6),
  VAR3   = c(2,4,6,NA,NA,NA),
  VAR4   = c(1,2,3,1,2,3),
  VAR5 = c(3,2,1,NA,NA,NA)
)

library(ggplot2)
library(cowplot)

#create 3 plots for a species that has data and join the plots with cowplot
plot1.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR1,y=VAR2))+
  geom_point()+
  theme_bw()

plot2.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR3,y=VAR5))+
  geom_point()+
  theme_bw()

plot3.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR4,y=VAR3))+
  geom_point()+
  theme_bw()

group.plot.1<-(plot1.1 + plot2.1 + plot3.1)
group.plot.1

But for others there will be no data to support some of the plots so an empty plot is returned. This still looks good and everything lines up fine.

plot1.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR1,y=VAR2))+
  geom_point()+
  theme_bw()

plot2.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR3,y=VAR5))+
  geom_point()+
  theme_bw()

plot3.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR4,y=VAR1))+
  geom_point()+
  theme_bw()

group.plot.2<-(plot1.2 + plot2.2 + plot3.2)
group.plot.2

For the plots with no data I would like to add a 'No Data' label. I do that using ggdraw but when I group the plots the empty plot with the label no longer lines up with the other plots.

#adding 'No Data' label to empty plot
plot2.2b <- ggdraw(plot2.2)+
  draw_label("No Data", color = "#C0A0A0", size = 10)

group.plot.2b<-(plot1.2 + plot2.2b + plot3.2)
group.plot.2b

How do I get the empty plot to line up with the others again? It feels like there should be an easy solution for this or a better way of adding the label but I have had no success with other approaches. Thanks for your help.

I am running through a large dataset and creating a standard set of plots for a large number of species in that dataset. The premise is quite simple. For some species there will be data to create all the plots.

#produce some data
samples <- data.frame(
  GROUP = c("dog","dog","dog","cat","cat","cat"),
  VAR1 = c(1,3,5,1,3,5),
  VAR2   = c(2,4,6,2,4,6),
  VAR3   = c(2,4,6,NA,NA,NA),
  VAR4   = c(1,2,3,1,2,3),
  VAR5 = c(3,2,1,NA,NA,NA)
)

library(ggplot2)
library(cowplot)

#create 3 plots for a species that has data and join the plots with cowplot
plot1.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR1,y=VAR2))+
  geom_point()+
  theme_bw()

plot2.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR3,y=VAR5))+
  geom_point()+
  theme_bw()

plot3.1<-ggplot(samples[samples$GROUP=="dog",],aes(x=VAR4,y=VAR3))+
  geom_point()+
  theme_bw()

group.plot.1<-(plot1.1 + plot2.1 + plot3.1)
group.plot.1

But for others there will be no data to support some of the plots so an empty plot is returned. This still looks good and everything lines up fine.

plot1.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR1,y=VAR2))+
  geom_point()+
  theme_bw()

plot2.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR3,y=VAR5))+
  geom_point()+
  theme_bw()

plot3.2<-ggplot(samples[samples$GROUP=="cat",],aes(x=VAR4,y=VAR1))+
  geom_point()+
  theme_bw()

group.plot.2<-(plot1.2 + plot2.2 + plot3.2)
group.plot.2

For the plots with no data I would like to add a 'No Data' label. I do that using ggdraw but when I group the plots the empty plot with the label no longer lines up with the other plots.

#adding 'No Data' label to empty plot
plot2.2b <- ggdraw(plot2.2)+
  draw_label("No Data", color = "#C0A0A0", size = 10)

group.plot.2b<-(plot1.2 + plot2.2b + plot3.2)
group.plot.2b

How do I get the empty plot to line up with the others again? It feels like there should be an easy solution for this or a better way of adding the label but I have had no success with other approaches. Thanks for your help.

• r
• ggplot2
• cowplot

• 1 cowplot::plot_grid(plot1.2, plot2.2, plot3.2, align = "hv", axis = "none", rows = 1) works for me – Michiel Duvekot Commented 2 days ago
• yes, that works for including plot2.2 but not for including plot2.2b (i.e. the one with the 'no data' label added to it) – Always_confused Commented 2 days ago

1 Answer 1

Do you need to create the empty plot with ggdraw()? You can create the same empty plot with {ggplot}, and it will work with {cowplot} as expected:

plot2.2 <- ggplot(samples[samples$GROUP=="cat",],aes(x=VAR3,y=VAR5))+
  annotate(geom = "text", label = "No data", x = 1, y = 1,
           color = "#C0A0A0", size = 5) +
  theme_bw() +
  theme(axis.text = element_blank(),
        panel.grid = element_blank())