How to identify a single point and label it in R

Question

I’m looking to identify and label a single point on a graph. I am working with a dispersion chart with an indicator on the y-axis and the years on the x-axis. My intention is to highlight Brazil and identify it with a legend inside the chart.

The data below can be used to illustrate the problem:

pibs <- tibble::tibble(
  posicao = c(1, 2, 3, 7, 4, 5, 6, 8, 9, 11, 13, 14, 10, 12, 17, 15, 16, 18,
              20, 19, 21),
  pais = c("Brasil", "México", "Argentina", "Venezuela", "Colômbia",
           "Chile", "Peru", "Equador", "Cuba", "República Dominicana",
           "Uruguai", "Costa Rica", "Guatemala", "Panamá", "El Salvador",
           "Bolívia", "Paraguai", "Honduras", "Haiti", "Nicarágua",
           "Santa Lúcia"),
  valor = c(2055000, 1149000, 637700, 210100, 309200,
            277000, 215200, 102300, 93790, 75020, 58420,
            58060, 75660, 61840, 28020, 37120, 29620,
            22980, 8608, 13730, 1686)
)

Answer 1

The gghighlight is a sensational package to do this:

Here’s an example of how to do with it:

library(tidyverse)
library(gghighlight)

ggplot(pibs, aes(x = posicao, y = valor)) +
  geom_point(color = "blue") +
  gghighlight(pais == "Brasil", label_key = pais, unhighlighted_colour = "black")

^{Created on 2019-02-28 by the reprex package (v0.2.1)}

Answer 2

First let’s create a dataset to use in the answer. I used the package datapasta to copy and paste the table of this page.

library(tidyverse)
pibs <- tibble(
  posicao = c(1, 2, 3, 7, 4, 5, 6, 8, 9, 11, 13, 14, 10, 12, 17, 15, 16, 18,
              20, 19, 21),
  pais = c("Brasil", "México", "Argentina", "Venezuela", "Colômbia",
           "Chile", "Peru", "Equador", "Cuba", "República Dominicana",
           "Uruguai", "Costa Rica", "Guatemala", "Panamá", "El Salvador",
           "Bolívia", "Paraguai", "Honduras", "Haiti", "Nicarágua",
           "Santa Lúcia"),
  valor = c(2055000, 1149000, 637700, 210100, 309200,
            277000, 215200, 102300, 93790, 75020, 58420,
            58060, 75660, 61840, 28020, 37120, 29620,
            22980, 8608, 13730, 1686)
)

pibs
#> # A tibble: 21 x 3
#>    posicao pais                   valor
#>      <dbl> <chr>                  <dbl>
#>  1       1 Brasil               2055000
#>  2       2 México               1149000
#>  3       3 Argentina             637700
#>  4       7 Venezuela             210100
#>  5       4 Colômbia              309200
#>  6       5 Chile                 277000
#>  7       6 Peru                  215200
#>  8       8 Equador               102300
#>  9       9 Cuba                   93790
#> 10      11 República Dominicana   75020
#> # ... with 11 more rows

^{Created on 2019-02-27 by the reprex package (v0.2.1)}

The form with which the ggplot2 work is that there should be a column in the table indicating which countries to show. We can do it like this:

pibs2 <- pibs %>% 
  mutate(mostrar = pais == "Brasil")
pibs2
#> # A tibble: 21 x 4
#>    posicao pais                   valor mostrar
#>      <dbl> <chr>                  <dbl> <lgl>  
#>  1       1 Brasil               2055000 TRUE   
#>  2       2 México               1149000 FALSE  
#>  3       3 Argentina             637700 FALSE  
#>  4       7 Venezuela             210100 FALSE  
#>  5       4 Colômbia              309200 FALSE  
#>  6       5 Chile                 277000 FALSE  
#>  7       6 Peru                  215200 FALSE  
#>  8       8 Equador               102300 FALSE  
#>  9       9 Cuba                   93790 FALSE  
#> 10      11 República Dominicana   75020 FALSE  
#> # ... with 11 more rows

^{Created on 2019-02-27 by the reprex package (v0.2.1)}

Now, with the column in hand it is possible to use it to achieve the goal. To highlight with the color, just use the new column to map the colors.

p <- ggplot(pibs2, aes(posicao, valor, col = mostrar)) +
  geom_point() +
  scale_color_manual(values = c("black", "darkgreen"))

p

^{Created on 2019-02-27 by the reprex package (v0.2.1)}

And after that it is possible to use this column to filter the table and use only the record that matters to the layer with the text.

p + 
  geom_text(aes(label = pais), data = filter(pibs2, pais == "Brasil"),
            position = position_nudge(1)) +
  theme(legend.position = "none")

^{Created on 2019-02-27 by the reprex package (v0.2.1)}