how to get the list of genes in each cluster of a heatmap in R

Question

I have used the following code to make this heatmap for my large dataset. Now I want to get the name of rows (name of the genes) for the second cluster (red on the left and green on the right). First I tried to follow the instructions on this page to separate clusters but due to the complexity of the dendrogram on the y-axis, the solutions were not practical for me. Then I tried to make clusters before making the heatmap hoping to get the same result. the codes on this page did not generate the same heatmap.

is there any possible way that I could get the list of genes for the desired regions of the heatmap?

library(RColorBrewer)
my_palette <- colorRampPalette(c("red", "black", "green"))(n = 100)

heatmap.2(as.matrix(sd_pol),
          col=my_palette, 
          density.info="none",
          trace="none",
          symm=F,symkey=F,symbreaks=F, scale="none",
          labRow=NA,
          main="test_heatmap",
          margins = c(8, 8))

Answer 1

As detailed in the docs, heatmap.2 uses the function hclust for clustering, so an approach for what you want to achieve would be to create the clustering object and then apply cutree as described here.

However, you will have to tune the parameters h or k to get the results you want:

require(graphics)
hc <- hclust(dist(USArrests))
cutree(hc,k = 2)

Output:

  Alabama         Alaska        Arizona       Arkansas     California       Colorado    Connecticut 
         1              1              1              2              1              2              2 
  Delaware        Florida        Georgia         Hawaii          Idaho       Illinois        Indiana 
         1              1              2              2              2              1              2 
      Iowa         Kansas       Kentucky      Louisiana          Maine       Maryland  Massachusetts 
         2              2              2              1              2              1              2 
...