Non-ultrametric tree in Beast (Help!)

494 views
Skip to first unread message

Laura Cespedes

unread,
Jul 25, 2017, 1:31:55 PM7/25/17
to beast-users
Hello everyone,

I am trying to construct a non-ultrametric tree using Beast and I am not sure how to do it. I am using version 1.8.4 on Beauti and Beast. I know I should change parameters related to 'clock rate' but I have tried several things and failed (Always obtain an ultrametric tree). If someone know how to do this I would really appreciate any help you can give me!

Thank you all in advance.

Best,

Laura

Santiago Sánchez

unread,
Jul 25, 2017, 8:14:22 PM7/25/17
to beast-users
Hi Laura,

I don't think it is possible to estimate non-ultrametric trees with BEAST. There is no way of ignoring clock models (e.g strict or relaxed clocks). Also, the tree prior models (e.g speciation or coalescent) all require the trees to be ultrametric.

Perhaps you are referring to obtaining branch lengths in units of substitutions per site?

Cheers,
Santiago
--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To post to this group, send email to beast...@googlegroups.com.
Visit this group at https://groups.google.com/group/beast-users.
For more options, visit https://groups.google.com/d/optout.
--
==========================
Santiago Sanchez-Ramirez, PhD
Postdoctoral Associate
Ecology and Evolutionary Biology
University of Toronto
==========================

Andrew Rambaut

unread,
Jul 26, 2017, 6:49:32 AM7/26/17
to beast...@googlegroups.com
Hi Laura,

What do you mean by ‘non-ultrametric’? This term is rather vaguely defined I think in phylogenetics.  

Do you mean non-molecular-clock models? Relaxed molecular clock? BEAST can only do molecular clock trees - with branches in units of time (either absolute time if you have calibrating information or relative time).  

If you need non-molecular-clock trees (in a Bayesian framework) then MrBayes may be what you want.

For molecular clocks, we are starting to freshen up the documentation about the models in BEAST 1.X (over the next few days):


Andrew

Reply all
Reply to author
Forward
0 new messages