When looking at a heatmap created by DESeq2, how can I find what genes are in specific regions of interest?

Question

[mgranada3]

Once I have generated my heat map and clustered the rows, how can I find out what genes are there? For example, in my image, I am interested in knowing what gene is shown as up-regulated (in red) which I squared off in my image above sample AS62.

Is there a way to make heat maps interactive? Or how can I view the gene names for certain areas?

[Mark]

Look at the dataframe/matrix that was used to generate the heatmap and apply a filter rule to find which genes have fc expression of greater than say 20.

If you want an interactive pheatmap look at this: https://jokergoo.github.io/InteractiveComplexHeatmap/articles/interactivate_indirect.html

Also have you searched biostars for prev questions similar to yours? Specific gene labeling in a heatmap

[ATpoint]

You should be plotting standardized values, not (log)expression value. Here is why: Scaling RNA-Seq data before clustering?

Answer 1

[fracarb8]

The clustering information is stored inside heatmap$tree_row.

tmp_heatmap <- pheatmap(...)
plot(tmp_heatmap$tree_row)

Based on the clustering cut and the expression values, you should be able to pinpoint your genes of interest.