Color axis text by variable

Question

I'd like to vary the color of my heatmap axis text depending on another variable in the dataset. This is what I've tried so far:

#load data, scale numeric columns, add state abbreviation and region
state_data <- data.frame(state.x77)
state_data <- state_data[,1:8]
state_data <- rescaler(state_data, type='range')
state_data$State <- state.abb
state_data$Region <- state.region

#make heatmap
melted_state <- melt(state_data,id.vars=c('State', 'Region'))

p <- ggplot(melted_state, 
        aes(x=State, y=variable))
p <- p + geom_tile(aes(fill = value), colour = "white")
p <- p + theme(axis.text.x=element_text(colour="Region")) ## doesn't work!
p

I get this error: Error in grid.Call(L_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : invalid color name 'Region'

And if I remove the quotes around 'Region' I get this error:

Error in structure(list(family = family, face = face, colour = colour,  : 
  object 'Region' not found

How can I do this?

Answer 1

Settings accessed via theme cannot be mapped to the data like aesthetics unfortunately. You will need to construct an appropriate palette (read: list) of colors manually.

One way to do so would be something like this:

library(scales)
numColors <- length(levels(melted_state$Region)) # How many colors you need
getColors <- scales::brewer_pal('qual') # Create a function that takes a number and returns a qualitative palette of that length (from the scales package)
myPalette <- getColors(numColors)
names(myPalette) <- levels(state_data$Region) # Give every color an appropriate name
p <- p + theme(axis.text.x = element_text(colour=myPalette[state_data$Region])))