Higher-level phylogeny in BEAST

[samuel.ge...@gmail.com]

Hi Everyone,

My doubt is rather theoretical, but I would welcome some practice advice. It's been a few years since I've last worked with BEAST and I've always done strictly species trees in it, mostly phylogeography and species delimitation.

Now, I'm trying to make a family tree, in which each tip is a species representing a family inside a superfamily. I have 4 loci (5 markers, 2 of which are mitochondrial), but they don't all have the same terminals and their topologies conflict a lot. Thus, I preferred to leave trees unlinked to keep that uncertainty in my tree. I'm using the Standard layout because my data has no support for the population priors in *BEAST (I actually tried to use *BEAST, but as predicted, the popmean prior doesn't converge and it pulls a lot of other priors down with it).

However, Standard BEAST doesn't give you a 'species tree' like *BEAST does. And I'm not sure how to get my 'final' tree from that data. How should I proceed? Should I somehow combine the separate gene trees or is linking all the tress the only way? Should I use *BEAST?

Cheers,

Sam