R Corrplot：更改列文本标签的颜色，但不更改行文本标签的颜色

Question

I'm using the corrplot function to create a visualisation of beta coefficients on 4 x variables and 12 y variables. Below is the code I have used so far:

library(corrplot)
corrplot(beta_m, is.corr=FALSE, method="square", 
     tl.srt=65, tl.cex=0.95, number.cex=0.8, tl.col='black', cl.length 
= 5, cl.align="l")

This is the data where it comes from:

beta_m<-structure(c(0, 0, -0.584090272222348, 0, 0, 0.11291756016558, 
-0.435893170100887, 0, 0, 0, -0.460227704153223, 0, 0, 
0.222130737988377, 
0, 0.0974747494000928, 0, 0.117470450053172, -0.723935606930726, 
0, -0.308146541461676, 0, 0, 0, 0, 0.153128826514138, 0, 0, 0, 
0, -0.360425453158442, 0, 0, 0, -0.508415520561608, 0, 
-0.303599419688516, 
0, 0, 0, 0, 0, -0.551430826126704, 0, 0, 0, 0, 0), .Dim = c(4L, 
12L), .Dimnames = list(c("Sex: Female", "Age", "IM: MVA", "LOC 
duration: < 5 min"
), c("PDGF", "$ paste(MIP-1,beta)", "$ paste(TNF,-alpha)", "IP-10", 
"Eotaxin", "FGF-basic", "IL-1ra", "IL-9", "MCP-1", "IL-17A", 
"IL-8", "$ paste(IFN,gamma)")))

And this is what it currently looks like:

What I need is for some of column labels (the 12 labels on the top) to be different colours, while the 4 row labels remain black. The tl.col argument changes both row and column label colours.

Edit: It could also work if some of the column labels are bold and others are not.

Answer 1

One solution is like this:

1. Plot the picture with all labels turned to tl.col="red" .
2. Remove the column names from the beta_m for those columns you want to remain red.
3. Add a second picture on top with tl.col="black" .

Here is a demonstration:

corrplot(beta_m, is.corr=FALSE, method="square", tl.srt=65, tl.cex=0.95, number.cex=0.8,
         tl.col='red', cl.length = 5, cl.align="l")

colnames(beta_m)[5:7] <- ""

corrplot(beta_m, is.corr=FALSE, method="square", tl.srt=65, tl.cex=0.95, number.cex=0.8,
         tl.col='black', cl.length = 5, cl.align="l", add=TRUE)

Here is the result: