Adjustable color palette for the Mk model plotting function, plot.Qmatrix

Yesterday I posted about how the default color palette of phytools::contMap should probably not be used – it is not colorblind friendly & renders poorly if converted to a grey color scale.

This is easy to adjust using the phytools function setMap (which also works for other "contMap" like graphs, such as the plotting method for multirateBM).

However, there are other plotting functions in phytools whose colors are not as easy to adjust. I just pushed a small update to one of these: plot.Qmatrix. This is the S3 plot method used for Q matrices (the object of class "Qmatrix") as well as fitted model objects from phytools::fitMk and geiger::fitDiscrete.

Let’s see how this works.

First, I’m going to load phytools and an example dataset used in my recent book with Luke Harmon.

## load phytools
library(phytools)
## check package version number
packageVersion("phytools")

## [1] '1.3.0'

## load tree & data
anole.tree<-read.tree(file="http://www.phytools.org/Rbook/1/Anolis.tre")
anole.ecomorphs<-read.csv(file="http://www.phytools.org/Rbook/1/ecomorph.csv",
	row.names=1)
ecomorph.tree<-drop.tip(anole.tree,
	geiger::name.check(anole.tree,anole.ecomorphs)$tree_not_data)
ecomorph<-setNames(anole.ecomorphs$ecomorph,row.names(anole.ecomorphs))

Now I’ll fit the “all-rates-different” ("ARD") model to these data. I’m not arguing this is a particularly good model; however, it is one that gives us lots of different rates to plot!

## fit ARD model
fitARD<-fitMk(ecomorph.tree,ecomorph,model="ARD",pi="fitzjohn",
	opt.method="optimParallel")
fitARD

## Object of class "fitMk".
## 
## Fitted (or set) value of Q:
##           CG       GB        TC        TG       Tr        Tw
## CG -0.000002 0.000000  0.000000  0.000000 0.000000  0.000001
## GB  0.000000 0.000000  0.000000  0.000000 0.000000  0.000000
## TC  0.379452 0.000000 -0.594337  0.000000 0.214884  0.000000
## TG  0.090593 0.334218  0.084092 -1.002463 0.072221  0.421338
## Tr  0.000000 0.000000  0.000000  0.000000 0.000000  0.000000
## Tw  0.000000 0.330015  0.768605  0.000000 0.000001 -1.098622
## 
## Fitted (or set) value of pi:
## CG GB TC TG Tr Tw 
##  0  0  0  1  0  0 
## due to treating the root prior as (a) nuisance.
## 
## Log-likelihood: -68.387541 
## 
## Optimization method used was "optimParallel"

Now, let’s plot it using the default color palette in plot.Qmatrix.

plot(fitARD,color=TRUE,lwd=3,tol=1e-2)

To simulate how this would look when rendered into grey scale, we need to use TeachingDemos::col2grey again, but this time with the newplot.Qmatrix argument palette as follows.

cols<-colorRampPalette(c("blue", "purple", "red"))(20)
plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols))

Obviously, that’s no good!

As before, let’s try the ‘viridis’ palette using viridisLite::viridis. First in color:

cols<-viridisLite::viridis(n=20)
plot(fitARD,color=TRUE,lwd=3,palette=cols,tol=1e-2)

Now, rendered into grey scale:

plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols))

Although this is obviously much better, something that we see right away is that our zero rates (which are shown using a grey color) are indistinguishable from our high rates. We can fix this either by using show.zeros=FALSE, e.g.:

plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols),
	show.zeros=FALSE)

Or, perhaps even better, by turning onwidth=TRUE.

plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols),
	width=TRUE)

Let’s see all our four plots together:

par(mfrow=c(2,2))
plot(fitARD,color=TRUE,lwd=3,tol=1e-2,width=TRUE,
	offset=0.025,mar=rep(1.1,4))
mtext("a) default palette",line=-1,adj=0.1)
cols<-colorRampPalette(c("blue", "purple", "red"))(20)
plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols),
	width=TRUE,offset=0.025,mar=rep(1.1,4))
mtext("b) default palette rendered in grey scale",line=-1,
	adj=0.1)
cols<-viridisLite::viridis(n=20)
plot(fitARD,color=TRUE,lwd=3,palette=cols,tol=1e-2,
	width=TRUE,offset=0.025,mar=rep(1.1,4))
mtext("c) viridis palette",line=-1,
	adj=0.1)
plot(fitARD,color=TRUE,lwd=3,tol=1e-2,
	palette=TeachingDemos::col2grey(cols),
	width=TRUE,offset=0.025,mar=rep(1.1,4))
mtext("d) viridis palette rendered in grey scale",line=-1,
	adj=0.1)