Problem with fixing a Newick starting tree

[Tiago Simoes]

Dear all,

I am using BEAST 2 (v. 2.4.1) and implementing total evidence FBD dating. I am also contraining my ingroup as monophyletic. The analysis, however, generates a very poor ESS for the posterior, even after running for 10 000 000 generations with 100 initial attempts. Therefore, i would like to use a fully bifurcating tree obtained from Mr Bayes (converted into Newick format) as the initial tree for this analysis. I have followed the suggestions on how to include the Newick tree in the xml file in the BEAST website tutorial section, but BEAST crashed when I try to run it. COuld you give me any help on this?

I am attaching here the original file (that runs fine, called Analysis_original), and the new file (Analysis_Newicktree) with the <init element replaced by the new TreeParser element with my Newick tree. Is there any changes you suggest I try to modify this element, or others in the xml file?

Any help is terribly appreciated

Best regards!

[Remco Bouckaert]

Hi Tiago,

When you run the file, the error message shown says that there are duplicate IDs for elements with ID=Tree.t:Acro_tree, so you have to change one (preferably the ID of the newick tree), so make it say id=“startingTree”.

The element name should not be stateNode, but init, since it is an initialiser, and not part of the state.

The parser should refer to the tree it is supposed to initialise, so set initial='@Tree.t:Acro_tree’.

Did you follow the instruction from http://beast2.org/2014/07/28/all-about-starting-trees/?

I assumed that the taxon name Bikem_aleadonta should be Jeddaherdan_aleadonta? There is no sequence named Bikem_aleadonta, so without changing this, the Newick string cannot be parsed.

In short, if you replace the tree parser element with this fragment, it runs.

<init id="startingTree" spec='beast.util.TreeParser' initial='@Tree.t:Acro_tree' IsLabelledNewick='true' adjustTipHeights='false'

 taxa='@Acro_MORPHO_nex2'

 newick="(Basiliscus_vittatus:0.1966125,Iguana_iguana:0.2092816,(((((Brookesia_sp:0.3331211,(Trioceros_hoehnellii:0.1806134,Chamaeleo_dilepis:0.2311545):0.09052158):0.3817216,((Trapelus_sp:0.3533364,(Agama_agama:0.8578723,Phrynocephalus_mystaceus:0.3125925,Laudakia_stellio:0.3060172):0.05699074):0.2379796,(Draco_timoriensis:0.4123655,(Calotes_versicolor:0.2963657,Bronchocela_cristatella:0.3916853,(Pseudocalotes_sp:0.2171786,Japalura_splendida:0.2001971):0.1217244,Tikiguania_estesi:0.07525839):0.1811805):0.181529):0.05913425):0.04184527,(Physignathus_cocincinus:0.3010555,(((Physignathus_lesueurii:0.1550491,(((((Pogona_barbata:0.08315177,Diporiphora_albilabris:0.1686838):0.02331075,Tympanocryptis_lineata:0.1028939):0.02050809,Rankinia_diemensis:0.1093417):0.03793772,Amphibolurus_gilberti:0.2740804):0.02553191,Ctenophorus_decresii:0.2075914):0.05000412):0.03430809,Moloch_horridus:0.1645485):0.03787057,Hypsilurus_dilophus:0.2312193):0.06658397):0.3276727):0.03930245,Hydrosaurus_pustulatus:0.4187122):0.05802363,(Leiolepis_belliana:0.4055104,(Uromastyx_aegyptius:0.2424928,(Gueragama_sulamericana:0.2664751,Jeddaherdan_aleadonta:0.1181934):0.2476835):0.2532973):0.05021431):0.6701399);"/>

Cheers,

Remco

--
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.
<Analysis_original.xml><Analysis_newicktree.xml>

[Tiago Simoes]

Dea Remco, another message to clarify your question. I followed the instruction from: http://beast2.org/fix-starting-tree/ at first...then I tried to follow from the other tutorial webpage (All About Starting Trees) you mentioned. They seem to provide alternative ways of providing a starting tree. In the second attempt, I produced a script element for starting trees similar to the ones you provided, but I did not notice the ID name was similar to another ID in the same file.

Cheers,

Tiago