Confidence intervals ggplot2 with different colours based on preselection-CodePudding

I am new at the stackoverflow, so excuse me if my explanation is not as precise. I have plotted some confidence intervals from a dataset where y = discrete ascending numbers (1:31) to position the intervals into the y axis, x = mean, lower and upper 95% HPD intervals. I attach part of the dataset:

y	x	lower	upper
1	143,580	80,675	203,670
2	127,740	90,799	168,240
3	134,840	98,665	174,030
4	138,660	99,682	176,360

    ggplot(data, aes(x, y))         
      geom_point()  
      geom_errorbar(aes(xmin = upper, xmax = lower))   
      scale_y_continuous(position = "right")  
      scale_x_reverse()  
      ggtitle("Molecular Dating (95% HPD intervals)")  
      theme(plot.title = element_text(hjust = 0.5))  
      xlab("Time (years)")  
      theme(axis.title.y = element_blank(),
          axis.text.y = element_blank(),
          axis.ticks.y = element_blank())
    )

I want to apply 2 different colours on the confidence intervals e.g. for the 8th, 17th and 31st the colour red and for the 4th, 6th and 9th the colour gray. I tried a variety of things but I think I am missing something. Could someone please suggest a way that this could work?

Thank you very much in advance and I appreciate your feedback!

CodePudding user response：

You can create a dichotomous variable and map it to the color aesthetic.
In the code below I choose values at random using sample. That should be replaced by your colors assignment criterion. Then, the actual colors are set in scale_color_manual.

Also, your data has comma as the decimal point, that's why I read it with dec = ","

x <- 'y     x   lower   upper
1   143,580     80,675  203,670
2   127,740     90,799  168,240
3   134,840     98,665  174,030
4   138,660     99,682  176,360'
data <- read.table(textConnection(x), header = TRUE, dec = ",")

suppressPackageStartupMessages({
  library(ggplot2)
  library(dplyr)
})

set.seed(2022)
colors <- sample(c("this", "that"), nrow(data), replace = TRUE)
data$colors <- colors

ggplot(data, aes(x, y, color = colors))   
  geom_point()  
  geom_errorbar(aes(xmin = upper, xmax = lower))   
  scale_color_manual(values = c(this = "grey", that = "red"))  
  scale_y_continuous(position = "right")  
  scale_x_reverse()  
  xlab("Time (years)")  
  ggtitle("Molecular Dating (95% HPD intervals)")  
  theme(plot.title = element_text(hjust = 0.5))  
  theme(axis.title.y = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks.y = element_blank())

^{Created on 2022-08-06 by the reprex package (v2.0.1)}