Highlighting individual axis labels in bold using ggplot2

Question

I want to highlight individual axis labels in bold. I am aware of this answer by @MrFlick but I can\'t figure out how to do this a) for more than one item,

Answer 1

I'm not sure if you can map label characteristics by name, but it's definitely possible to do it by position with a call to theme:

ggplot(xx, aes(x=CLONE, y=VALUE, fill=YEAR)) + 
  geom_bar(stat="identity", position="dodge") +
  facet_wrap(~TREAT) +
  theme(axis.text.x = element_text(face = c('bold', 'bold', 'plain', 'plain', 'bold')))

Note that the listed font faces for axis.text.x are the same length as the labels of your x-axis (five elements). This produces:

Answer 2

Here's a generic method to create the emboldening vector:

colorado <- function(src, boulder) {
  if (!is.factor(src)) src <- factor(src)                   # make sure it's a factor
  src_levels <- levels(src)                                 # retrieve the levels in their order
  brave <- boulder %in% src_levels                          # make sure everything we want to make bold is actually in the factor levels
  if (all(brave)) {                                         # if so
    b_pos <- purrr::map_int(boulder, ~which(.==src_levels)) # then find out where they are
    b_vec <- rep("plain", length(src_levels))               # make'm all plain first
    b_vec[b_pos] <- "bold"                                  # make our targets bold
    b_vec                                                   # return the new vector
  } else {
    stop("All elements of 'boulder' must be in src")
  }
}

ggplot(xx, aes(x=CLONE, y=VALUE, fill=YEAR)) + 
  geom_bar(stat="identity", position="dodge") +
  facet_wrap(~TREAT) +
  theme(axis.text.x=element_text(face=colorado(xx$CLONE, c("A", "B", "E"))))

Answer 3

You can create a named vector of expressions (that turn text to bold) in scale_x_discrete and use parse=TRUE to evaluate the expressions:

ggplot(xx, aes(x=CLONE, y=VALUE, fill=YEAR)) + 
    geom_bar(stat="identity", position="dodge") +
    facet_wrap(~TREAT) +
    scale_x_discrete(labels=c("A"=expression(bold(A)), "C"=expression(bold(C)),
                              "E"=expression(bold(E)), parse=TRUE))

You can probably create the vector of expressions programmatically, rather than typing it out, but the way to do that is escaping me right now.