correlation plot (Corrplot heatmap) where showing significance level but row and column will be be different (selected)?

0
votes

I am trying to plot a correlation heatmap by using corrplot package. Details can be found here. But my question is if I want to limit the row and column name where we need the modification?

Row name

mpg cyl disp hp drat

Column name

wt qsec vs am gear carb

Data M = cor(mtcars)

corrplot(M, p.mat = testRes$p, method = 'color', diag = FALSE, type = 'upper', sig.level = c(0.001, 0.01, 0.05), pch.cex = 0.9, insig = 'label_sig', pch.col = 'grey20', order = 'AOE')

Figure should be look like below:

enter image description here

1
rlimitheatmapcorrelationr-corrplot

1 Answers

1
votes

Here is a partial solution.

library(corrplot)
library(Hmisc)    # Computes correlation and p-value matrix
Mp <- rcorr(as.matrix(mtcars))
corrplot(Mp$r[1:5, 6:11], p.mat=Mp$P[1:5, 6:11], method="color", addCoef.col="black")

produces a plot with the correlation coefficients.

corrplot(Mp$r[1:5, 6:11], p.mat=Mp$P[1:5, 6:11], method="color", sig.level = c(0.001, 0.01, 0.05), insig = 'label_sig')

produces a plot that labels the significance of the coefficients. I have not seen a way to combine both so that they do not overprint. Several of your arguments: method = 'color', diag = FALSE, type = 'upper', order = 'AOE' are not appropriate for non-symmetrical plots.

Corrplots