Different topology from starting tree

[Jose Ramirezs]

Dear Beast,

I am currently running an analysis with a fixed tree topology (newick). However, the resulting tree topology is different: some branches are swaped and the tips are no longer equidistant from the root. I revised the runs in Tracer and the parameters that relate to the tree (Tree-height, Tree.treeLength) have values above 200 ESS. I don’t know where the problem might be and I hope that someone could help me.

Your help would be greatly appreciated.

kind regards

José

[HS]

Hi  José,

Did you turn off tree operators besides uploading a newick tree?

Best,

Hovhannes

[Jose Ramirezs]

Hi Hovhanne, 

Thank you for your answer,

Yes, the operators (Wide-exchange, Narrow-exchange, Wilson-Balding, Subtree-slide) that are related with the search of the tree are disabled or set to zero. But, I don`t know if I should turn off the Tree-scaler, Tree-root-scaler, Uniform, Up-down-operators, because I have "ultrametrized" the tree (topology resulting of ML analysis) by calibrating some nodes with the Chronos function of the APE package of R.

Kind regards,

José

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/87ddb6c7-dc80-499e-b1a8-dc0cab72af14n%40googlegroups.com.

--

E-mail: are...@gmail.com; jsala...@mendoza-conicet.gob.ar

Tel: +54 261 2520334

Lab site: https://sites.google.com/site/entomologiampuj/integrantes/investigadores-asociados

[Shanta Hejmadi]

This is the same issue I am having (BEAST2.6.4).  I added an ultrametric Newick starting tree, I commented out the  Wide-exchange, Narrow-exchange, Wilson-Balding, and Subtree-slide operators. The analysis runs (suspiciously slowly for having a fixed topology), and yet my starting tree is not the same topology as the last tree in my tree.trees file.  What was the resolution to this issue, because I also need it? 

[mgbran...@gmail.com]

I never had this problem in the past, but I just upgraded to BEAST 2.7.3, and am encountering it now. I input a newick starting tree and changed the tree search operators to 0. For some reason, the clade sister to all others changed places during the analysis run. I just modified the analysis to include a monophyletic constraint around all the sister clade species and it seems that this has fixed the problem. I started the run and immediately examined the tree file and now all trees appears to be identical except for the branch lengths. Problem solved?