r - 基于条件的部分粗体ggplot2指南标签

Question

我有一些要绘制的整洁数据，将两列合并为分组。

在图例中，我想打印其中一列中的条目之一大胆的脸。

这是一个最小的工作示例:

library(ggplot2)

plot.data <- data.frame(x=1:10, y=1:10,
                        class1=rep(letters[1:2], 10),
                        class2=rep(LETTERS[1:2], each=5))

# This class should be displayed in bold
# Might be subject to change; consider it a function parameter
class2.bold <- "A"

# This is the default
# Note that the used classes are automagically determined from the data
ggplot(plot.data, aes(x, y)) +
  geom_point(aes(color=paste(class1, class2, sep="/"))) +
  scale_color_discrete("class")

# This is what I want
# Note that it only works because I know the data used in this example
ggplot(plot.data, aes(x, y)) +
  geom_point(aes(color=paste(class1, class2, sep=" / "))) +
  scale_color_discrete("class",
                       label=c(expression(a / bold(A)),
                               expression(a / B),
                               expression(b / bold(A)),
                               expression(b / B)))

如何利用 class2.bold 和 plot.data 获得第二个图仅(即在编写代码时不进一步了解其内容)？

例如，当我将 class2.bold 更改为 "B" 并将更多数据行添加到plot.data 具有附加的 class1 和 class2 值/组合，我仍然希望绘图命令无需手动调整即可工作。

Answer 1

这对我有用。可能有更快的方法，但这可以实现。

您的代码:

  plot.data <- data.frame(x=1:10, y=1:10,
                            class1=rep(letters[1:2], 10),
                            class2=rep(LETTERS[1:2], each=5))
    
    # This class should be displayed in bold
    # Might be subject to change; consider it a function parameter
    class2.bold <- "A"
    
    # This is the default
    # Note that the used classes are automagically determined from the data
    ggplot(plot.data, aes(x, y)) +
      geom_point(aes(color=paste(class1, class2, sep="/"))) +
      scale_color_discrete("class")

对于粗体标签:

 #get unique labels
    legend_labs <- unique(paste(plot.data$class1, plot.data$class2, sep="/"))
    
    #order them in alphabetical order
    legend_labs <- legend_labs[order(legend_labs)]
    
    #Bold the correct letters
    legend_labs <- sub(pattern = class2.bold,
        replacement = paste0("bold(",class2.bold,")"),
        legend_labs)
    
    #make them expressions
    legend_labs <- lapply(parse(text = legend_labs), as.expression)
    
    #plot
    ggplot(plot.data, aes(x, y)) +
      geom_point(aes(color=paste(class1, class2, sep=" / "))) +
      scale_color_discrete("class",
                           label=legend_labs)

编辑:

如果你想加粗更多的字母，请改用第三步

#Bold the correct letters
for (i in 1:length(class2.bold)) {
legend_labs <- sub(pattern = class2.bold[i],
    replacement = paste0("bold(",class2.bold[i],")"),
    legend_labs)
}