search

amkozlov / raxml-ng

RAxML Next Generation: faster, easier-to-use and more flexible
GNU Affero General Public License v3.0
374 stars 62 forks source link

tbe values when using a constraint tree #175

Closed padpadpadpad closed 1 month ago

padpadpadpad commented 5 months ago

Hi everyone

I am running raxml-ng with a constraint tree and it does everything I want for a single tree. I manually re-root the tree using figtree and all my monophyletic groups remain monophyletic. However I have some taxa that are allowed to be free across the tree.

However after I have bootstrapped the tree and calculated the tbe, I would have expected some nodes where my constraints were to have a value of 1, but this does not seem to be the case.

Is it because some of the nodes which can be free place differently so this split is not always containing exactly the same taxa.

Having read what the tbe means I think this is correct but would be good to sanity check myself.

Cheers Dan

amkozlov commented 5 months ago

Hi Dan,

so based on your description it sounds reasonable. As long as TBE values for your "monophyletic" clades are high / close to 1, TBE is still doing what is supposed to: it prevents support from dropping low due to a few rogue taxa,