Tree topology changing even when setting weights of relevant operators to zero

[­Wonseop Lim]

Hi, I'm new to BEAST, and actually whole field of phylogenetics, so this is my first time generating time calibrated tree.

I'm currently running BEAST analysis with 4 loci of 150 species with fossil calibration. However, even when setting weights of Subtree slide, Narrow/Wide exchange and Wilson Balding operators to zero resulting tree topology differs slightly from input starting tree. What could be causing the problem? 

[higg...@gmail.com]

Perhaps one of the operators in the models you are using is changing the topology. 

For example, the optimised relaxed clock (ORC) uses operators that affect the topology, and the ORCInternalnodesOperator and ORCAdaptableOperatorSampler_NER operator needs to be removed in order for the topology not to be changed.

If you can give more detail about the model that would help finding out which operator can cause this behaviour.

Cheers, Remco

[Omar Idris]

Does anybody know why treesnnotator is sending errors? Ignoring translations and then near the end of the nexus block near line 1452 there is Java element exception. I run 100 million mcmc and saved 10000 trees. Tracer shows excellent ESS in 1000’s for all operators. So this should be a good file not to be missed. I checked the syntax it looks like the taxa labels have spaces and that may be the reason why they are all under quotation in the output trees file.

I checked all over the internet over parsing issues it seems the latest version of treasnotator has a snug?

I would appreciate the help, I would send the tree file error and the tree file itself if need be.

Thanks for the help!

ONI

--
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/0073d1a1-61d0-4542-8112-5432503b25a6n%40googlegroups.com.

[higg...@gmail.com]

Would you be able to make the tree file available?

[Omar Idris]

I think it has to do with translate block where I had spaces on the taxon labels. I processed it using python code on pycharm with underscores now it seems to work. I will let you know if nothing does come up.

I do appreciate your concern.

ONI

To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/a26f53b6-c27b-4afa-9975-711adff89ae3n%40googlegroups.com.

[Omar Idris]

It is just annotating! Awesomeness.

ONI

On Mon, Jan 1, 2024 at 1:10 PM higg...@gmail.com <higg...@gmail.com> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/a26f53b6-c27b-4afa-9975-711adff89ae3n%40googlegroups.com.

[higg...@gmail.com]

Glad to hear it works now! Another way around this problem is keeping spaces but putting double quotes around the taxon labels.