treeannotator

[Christopher Blair]

Hi everyone,

I am running TreeAnnotator for my *BEAST analysis (5 independent runs) and keep getting this error message: "Warning: a node has a disjoint 80.0% HPD region. This may be an artifact! Try
decreasing the enclosed mass or increasing the number of samples." There doesn't seem to be much info regarding this online. I have tried running additional runs and still get this message. Could someone explain what this means? I notice on the species tree in FigTree that some of the node bars are actually completely off the node. Is this what the error is referring to and if so, what can be done? Note that this error is only present in the species tree and not the individual gene trees. Thanks.

Chris

--
Christopher Blair, Ph.D. Candidate
Department of Ecology and Evolutionary Biology
University of Toronto and
Department of Natural History
Royal Ontario Museum
100 Queen's Park
Toronto, ON M5S 2C6
Canada
(416) 586-8094 (office)
http://individual.utoronto.ca/chrisblair/index.html

[Steen]

Dear Chris,

I have just experienced the exact same problem. I get the exact same warning message when I try Treeanotator on my star-BEAST run (v. 1.7.5)
I have run more than 100 parallel runs of the same star-BEAST xml-file and I keep getting this message.
I have tried to discard 50% of the first trees (burnin), but still get this message.
The individual gene-trees can be combined in Treeannotator without this message appearing.

The resulting combined tree is wrong. The species-tree have negative branch lengths, I can  'straighten the branches' by keeping the height of the branches, but the the nodes in my basal branches of fish are more than 1000 mio. years old, and these old (incorrect nodes) are present in both gene-trees and the species-tree.

Where you able to solve this problem?

Thanks in advance for your time

[Lorcan Carnegie]

Hi All,

I have a similar problem, did anyone find a solution?

Lorcan