pheatmap discrete colormap placement of tick labels

Question

I have a matrix with thousands of cells, with values range from 0-5. I want to use a discrete color palette to indicate the value of the respective cell. This code already works pretty well, but the position of the colormap labels is off. I simply want, that each tick is in the center for the corresponding color..

library("pheatmap")
library("RColorBrewer")

matrix <- round(matrix(rexp(200, rate=.1), ncol=20)/10)
color <- brewer.pal(max(matrix)+1,"Blues")
pheatmap(matrix,color=color,cluster_rows = F,cluster_cols = F)

The example produces a heatmap like this:

I want to move the colorbar labels, so it looks more like this:

If anyone has an idea how to do this, Id be very thankful!

Answer 1

You can work on the text grob of legend labels (p$gtable$grobs[[2]]$children[[2]]).
The y-positions of labels are defined as follows:

library("pheatmap")
library("RColorBrewer")

set.seed(123)
matrix <- round(matrix(rexp(200, rate=.1), ncol=20)/10)
color <- brewer.pal(max(matrix)+1,"Blues")
p <- pheatmap(matrix, color=color,cluster_rows = F,
              cluster_cols = F, silent=T)

p$gtable$grobs[[2]]$children[[2]]$y

[1] sum(0*min(1npc, 150bigpts), 1npc, -1*min(1npc, 150bigpts))
[2] sum(0.142857142857143*min(1npc, 150bigpts), 1npc, -1*min(1npc, 150bigpts))
    ....
[8] sum(min(1npc, 150bigpts), 1npc, -1*min(1npc, 150bigpts))

Label positions can be modified by changing the multiplicative constants placed before the first min (0, 0.1428, ..., 1).

# Sometimes the number of colored areas in the legend is not 
# equal to the number of labels    
N_col_areas <- length(p$gtable$grobs[[2]]$children[[1]]$y)
N_labels <- length(p$gtable$grobs[[2]]$children[[2]]$label)
yseq1 <- seq(0, 1, length.out=N_col_areas)
center <- yseq1[2]/2
yseq2 <- seq(center, 1-center, length.out=N_labels)

for (k in 1:(N_labels)) {
  p$gtable$grobs[[2]]$children[[2]]$y[[k]] <- 
          sum(yseq2[k]*min(unit(1,"npc"), unit(150,"bigpts")), 
              unit(1,"npc"), 
              -1*min(unit(1,"npc"), unit(150,"bigpts")))
}
plot.new()
print(p)