Dear Community,
I am drawing a heatmap with pheatmap package in R. I would like to show a certain property of two datasets and show two heatmaps next to each other. However, the order of the columns/rows in both heatmaps is slightly different. I understand that the order in which the columns/rows are organized has it's function in heatmap.
However, the two sample types I would like to change the order of, are next to another and this shouldn't disturb the whole idea of clustering. Is it possible to change the order of samples on a heatmap in pheatmap package ?
Please look at the example below:
I would like to have control in the bottom-right, and f4 in the middle. That would make this heatmap consistent with the other one I generated.
Thanks Devon! I am getting so much help from you on different platforms, I really appreciate it.
So it seems that it's not possible to tweak the order of the samples without losing (and manipulating) the dendrogram ? In my case rotating the left end of the dendrogram by 180 degrees forward (from our faces) wouldn't distort the meaning, but I don't expect it to be a one-liner option in the package...
Best regards ;)
It seems that that's not a built-in option. You do it by tweaking the source code of that function, but since you only have one affected figure, it's probably faster to just save it as a pdf and reorder things in Adobe Illustrator (this sort of thing would obviously constitute the maximum permissible changes you could make to a figure). Just make sure you mention having done that in the figure legend!