You might need to reindex each time, something like (pseudocode):
bad.nodes <- function_to_get_node_numbers_with_support_less_than_90(phy)
while(length(bad.nodes>0)) {
phy <- collapse.to.star(phy, bad.nodes[1])
bad.nodes <- function_to_get_node_numbers_with_support_less_than_90(phy)
}
But this will collapse the bad node and all its descendants. I don't see a
function that just collapses at a node and leaves descendants intact, but
this may be what you're looking for (the stuff that ape::collapse.singles
does is similar, and could perhaps be modified).
One pedantic point -- we often talk about node support, collapsing nodes,
etc., but it's really about the branch: the node doesn't have 83% bootstrap
support but the bipartition created by the edge subtending it does. For
example, if the tree is (A,(B,(C,(D,E)))), if the support for the CDE clade
is 83%, it's actually the proportion of trees that have a split between AB
| CDE, so the branch separating the CDE clade from everything else. The
node itself there has one branch coming in (that comes from the rest of the
tree containing A and B) and two branches coming out (one going to C, one
going to DE). We could summarize trees by actual node support: how many
times we have a node that has an edge going to AB, an edge going to C, and
an edge going to DE, but we typically don't. [For example, the branch could
separate AB | CDE, and the node is AB | C | DE, but one could also have a
node that is AB | CD | E for the same bipartition on the subtending edge].
I only bring it up here because if you end up having to write code to do
collapsing, it can be important to remember that what is actually being
collapsed is the edge subtending the labeled node because it is the the
edge with low support.
Best,
Brian
_______________________________________________________________________
Brian O'Meara
Professor, Dept. of Ecology & Evolutionary Biology, UT Knoxville
Associate Head, Dept. of Ecology & Evolutionary Biology, UT Knoxville
He/Him/His
On Fri, May 22, 2020 at 9:42 AM Karla Shikev <karlashi...@gmail.com> wrote:
> Dear all,
>
> I'm surprised I could not find a simple function to do this in any of the
> packages I know. I just want to take a tree with node labels (e.g.
> bootstrap values from a RAxML analysis) and collapse all nodes with support
> values below, say, 90%. I tried to do this recursively using
> collapse.to.star in phytools, but node numbers change after the first
> collapsed node and my indexing gets completely off.
>
> with best regards,
>
> Karla
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-phylo mailing list - R-sig-phylo@r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
> Searchable archive at
> http://www.mail-archive.com/r-sig-phylo@r-project.org/
>
[[alternative HTML version deleted]]
_______________________________________________
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/
