Colour in heatmap labels

data <- read.table("third.txt", sep = "\t", header = TRUE, row.names = 1)
data_matrix <- as.matrix(data)
colnames(data_matrix) <- c("473", "477", "474", "480", "481", "475", "479", "478", "495", "503", "499", "502", "494", "497", "496", "500")
heatmap(data_matrix)
library("RColorBrewer")
colr <- colorRampPalette(brewer.pal(8, "BuPu"))(256)
heatmap(data_matrix, scale = "column", col = colr, labRow = NA, cexCol = 0.75, margins = c(5, 2), xlab = "Data Samples", ylab = "gene expressions", main = "Differentially expressed genes")

I have written the code above and it produces the following heatmap. I also want to colour the name of the data samples with two different colours or draw a line to seperate the two clusters. Is there any way I could do that? enter image description here

labels colour heatmap

• 1,676 views • link

updated 5.3 years ago by seidel

• written 5.3 years ago by kbaitsi •

1 answer

One easy way to illustrate the groupings in the heat map would be to simply use a different heat map library. pheatmap is fairly easy to use. The help for pheatmap has some nice examples showing how to have several layers of groupings if you need it.

library(pheatmap)
library("RColorBrewer")

# create a matrix of data
x <- matrix(rnorm(300), nrow=30, ncol=10)
colnames(x) <- paste0("c", 1:ncol(x))
rownames(x) <- paste0("g", 1:ncol(x))

# skew the data so 1/2 the columns cluster together
x[,6:10] <- x[,6:10] + 3

# create a table describing columns
annotation_col <- data.frame(SampeType=rep(c(1,2),each=5))
rownames(annotation_col) <- colnames(x)

# set your color palette
colr <- colorRampPalette(brewer.pal(8, "BuPu"))(256)

# draw the heatmap
pheatmap(x, annotation_col=annotation_col, color=colr)

example heat map

• 0 views • link

written 5.3 years ago by seidel