Statistically defining deep branching clades in phylogenetic tree

Question

[Daniel]

I have a phylogenetic tree which visually has 3 major clades, one of which forks into two minor clades. I am discussing the groups in the text in regards to the metadata but I want to statistically/technically describe why the groups are seperate.

What is the best way to achieve this? The tree is UPGMA Jackknife generated, with around 150 samples. Note: These are not based on sequence similarity but community structure, so it has to be based on the tree rather than that underlying sequence or similar.

I have been looking at https://www.dendropy.org/ and the https://biopython.org/wiki/Phylo package but can't see a way to achieve this.

Thanks

[Joe]

I’ve used Simon Harris’ script (https://github.com/simonrharris/tree_gubbins) to ‘objectively define clades’ in the past to varying degrees of success. May be of interest?

[Daniel]

Thanks for the suggestion. I've been giving it a go but it's basically only identifying everything into one cluster. Having a look at ete but not solved it yet.