How to reorder dendrogram (R)?

Question

[khaeuk]

Hello everyone! As my title says, I am having difficulty reordering my dendrogram. What I would like to do with my dendrogram is to flip two nodes along the y axis. For example,

How would I do this? I looked at reorder.dendrogram. But I don't quite get what it means and how you're suppose to do it, for most sources basically say the same thing as the manual page without any other examples/explanation. I tried messing around with 'weights' for reorder, but it wouldn't change my dendrogram. I tried looking other pages like, https://www.mail-archive.com/[e-mail]/msg195536.html.

This is an example on the man page.

set.seed(123)

x <- rnorm(10)

hc <- hclust(dist(x))

dd <- as.dendrogram(hc)

dd.reorder <- reorder(dd, 10:1)

plot(dd, main = "random dendrogram 'dd'")

>

op <- par(mfcol = 1:2)

plot(dd.reorder, main = "reorder(dd, 10:1)")

plot(reorder(dd, 10:1, agglo.FUN = mean), main = "reorder(dd, 10:1, mean)")

par(op)

I don't know what that mfcol is but I tried changing the numbers, and it did swap some values, but I don't know how that works.

So next, I tried 'dendextend' package. I looked at the man page, seemed like there are some sort of ordering of the dendrogram, but more like ordering them in reverse or forward. So, my dendrogram would look like ...

Which is not what I'm looking for. Is there any other options? or is reorder.dendrogram is what i'm looking for, and i'm just not getting it? if so, can someone explain how it works?

Thank you!!

[e.rempel]

Hi,

have a look at my answer here:

heatmap.2 reorder branches

Answer 1

[Arnaud Ceol]

I've used a few time dendosort (https://f1000research.com/articles/3-177/v1 and https://cran.r-project.org/web/packages/dendsort/index.html), this is a nice library to sort your dendograms. You should have a look.

Answer 2

[Macherki M E]

I thinks that the dendrogram had the some order of user matrix input. In that way user will control the position of each vertice. If you wont separate a specific class statistically based on distance or graphically we have to reformulate data input.