search

jinyizju / U.PhyloMaker

This package aims to generate phylogenetic trees for a list of species of plants or animals, based on an existent backbone phylogeny.
6 stars 2 forks source link

Genus level trees #2

Open alrichardbollans opened 4 months ago

alrichardbollans commented 4 months ago

Is there any functions to generate genus-level trees from the given mega trees? Something like: rotl::tol_induced_subtree

Specifically, is it possible to extract a specific set of plant families from the GBOTB.extended.WCVP.tre and also reduce the tree to a genus rather than species level tree?

jinyizju commented 4 months ago

Is there any functions to generate genus-level trees from the given mega trees? Something like: rotl::tol_induced_subtree

Specifically, is it possible to extract a specific set of plant families from the GBOTB.extended.WCVP.tre and also reduce the tree to a genus rather than species level tree?

At the moment, there is no function that generates genus-level trees in U.PhyloMaker, however, it is possible to reduce the megatree to genus level tree of a specific set of families. Here is a way, first, you bind a marker(e.g. Genus_spmarker) of every genus that you intend to retain to the megatree (using the phylo.maker function in U.PhyloMaker), then you drop all the non-marker tips from the megatree (using the drop.tip function in ape package) and retain only the markers, you replace the markers with their genus names, and you will get the genus level tree. A reminder is that some genera are scattered on the megatree (i.e. polyphyletic), so you need to decide for yourself which nodes.type to use when using phylo.maker to bind the markers to the megatree.

alrichardbollans commented 4 months ago

Ok thanks! This is what I'd thought.

In this case, my intuition is to extend the remaining tips to make an ultrametric tree in order to reflect the current state of each genus. Does this make sense?

jinyizju commented 4 months ago

Ok thanks! This is what I'd thought.

In this case, my intuition is to extend the remaining tips to make an ultrametric tree in order to reflect the current state of each genus. Does this make sense?

The genus level tree is expected to be ultrametric, if not, it should be very close, and you can make it ultrametric, I think.