Help with plotting heatmap with both row and column label annotations

Question

[mohammedtoufiq91]

Hi,

I am working with the transcriptomics FC data matrix (genes on rows and samples on columns) containing healthy controls and subjects treated with different compounds. I usually use ComplexHeatmap for plotting genes on y-axis (right) and samples/groups on x-axis. This time I am interested in plotting a similar heatmap but with additional annotation on y-axis (left) (I have provided an example dataset along with the example). Basically, this will help in identifying top condition and genes; that is different between subjects for comparison.

print(Input.df)
          E0 Stim 1    E0 Stim 2  E0 Stim 3   P0 Stim 1  P0 Stim 2  P0 Stim 3   R0 Stim 1
Gene_1   0.02864341  0.312575401 -1.0881433  0.04872680 -1.1258452  2.0826464 -0.32690037
Gene_2  -0.17059789  0.080230470 -0.2826396 -0.00328775 -0.8016243  2.7087269 -0.36272853
Gene_3  -0.13343139 -0.952555759  1.2838753  2.07844595  1.6697437  1.2567775  0.50958702
Gene_4   0.36260241 -0.087364689  0.4363719 -0.01461830  0.4152043 -0.4914261 -0.16521061
Gene_5   0.65949239  2.492343411  7.8758014  4.73509139  4.9869428  2.4593781 -0.28308204
Gene_6   0.12986250  0.594530215  3.8886445 -0.22032753  1.4960060  2.9811058  0.75367603
Gene_7   0.30802148  1.257403638  4.8511548  2.87228300  3.3363807  0.9741493  0.36067011
Gene_8   0.19141119 -0.005561728  2.1122420  0.77093127  1.1628178 -0.2225401  0.03739008
Gene_9  -0.40870696 -0.372470805  0.3546816  0.16857304  1.4915303 -1.4071508  0.10397316
Gene_10  0.14843173  0.019060103  2.2315277  1.23777701  1.4841049 -0.1470720  0.31101290
          R0 Stim 2  R0 Stim 3
Gene_1  -2.79322025 -0.4150097
Gene_2  -0.82672908  0.0895918
Gene_3   2.50904053  1.2459908
Gene_4  -0.34493807  1.5299697
Gene_5   2.86835551  1.6966721
Gene_6  -1.02554246  1.5783651
Gene_7   1.87564089  2.2184017
Gene_8   0.09636701  1.3912611
Gene_9   0.23346362 -1.7168698
Gene_10  0.64624116  1.8690974

print(Sample_metadata.df)
          Subject  Stims
E0 Stim 1      E0 Stim 1
E0 Stim 2      E0 Stim 2
E0 Stim 3      E0 Stim 3
P0 Stim 1      P0 Stim 1
P0 Stim 2      P0 Stim 2
P0 Stim 3      P0 Stim 3
R0 Stim 1      R0 Stim 1
R0 Stim 2      R0 Stim 2
R0 Stim 3      R0 Stim 3

Thank you,

Toufiq

[ATpoint]

Sorry to be blunt, but this is all described extensively in the ComplexHeatmap manual under the "annotations" tab:

https://jokergoo.github.io/ComplexHeatmap-reference/book/heatmap-annotations.html

[mohammedtoufiq91]

ATpoint

Thank you. I am familiar with adding multiple row and column annotations on either side. I wanted to add sample metadata/condition information to the corresponding each gene, this will help in identifying top condition and genes; that is different between subjects (x-axis) for comparison. Perhaps, I was thinking if the data matrix I am using should be reshaped before plotting.

Answer 1

[Trivas]

Are you comfortable with pheatmap? With pheatmap, you would plot genes on the y-axis and the values from your conditions (E0 Stim 1 etc.) on the x-axis. I would then add an annotation_col to color your replicates. I find this blog post pretty helpful