Fatal exception: Could not find a proper state to initialise. Perhaps try another seed.

[Astrid Schuster]

Hello Beast 2.4.3 users
I generated a xml file and tried to run it with Beast 2.4.3.
However I am getting the following error message.
Can anyone help me with this? I assume, I have some mistakes in my priors, as it does not find a proper state to initalise, I changed them several times and tried to fix it, however it does not work. Aslo I thought that my calibration scale might be a problem? I am using 100.0 for 100 Ma. I also tried to scale it with 1.0. Just in case. But it doesn't work.
Many thanks in advanced,
A.




Start likelihood: -Infinity after 10 initialisation attempts
prot_fossils) = NaN (was NaN)  **
java.lang.RuntimeException: Could not find a proper state to initialise. Perhaps try another seed.
    at beast.core.MCMC.run(Unknown Source)
    at beast.app.BeastMCMC.run(Unknown Source)
    at beast.app.beastapp.BeastMain.<init>(Unknown Source)
    at beast.app.beastapp.BeastMain.main(Unknown Source)
    at beast.app.beastapp.BeastLauncher.main(Unknown Source)
Fatal exception: Could not find a proper state to initialise. Perhaps try another seed.
Fatal exception: Could not find a proper state to initialise. Perhaps try another seed.

BEAST has terminated with an error. Please select QUIT from the menu.

[Remco Bouckaert]

Hi Astrid,

The problem seem to be caused by the time units — you were using millions of years, and ages were in the hundreds, which causes numerical problems in the tree prior. 

If you change the units to hundreds of millions of years, so all ages are below 10, the analysis starts without problems (as in the attached).

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.
<DemoDivtime_FBD-1.xml>

[Astrid Schuster]

Hi Remco,

many thanks for your help! Yes it seems that this was causing the problem, it is running now. However, I  recocnized that the priors I put on the defined taxasets were removed. I thought I could add those node calibrations as well using the FBD Model. Was there a certain reason, why it was removed? some of the nodes are deeper in the past than the fossils included in the taxasets.

Again, many thanks for your help!
Astrid

[Remco Bouckaert]

Hi Astrid,

The XML file you posted did not have any node calibrations, though there were a few monophyletic constraints (see diff of the file you posted and the one I did below — only time units were changed). Perhaps you added the node distributions and saved them in another file? 

I am not sure whether mixing node distributions with the FBD model works as expected — it is well known that the Yule model and node distributions interact in unexpected ways, so the calibrated Yule is preferred over the Yule. Alternatively, if the node distributions are based on the fossil record, perhaps you can consider adding information of these fossils to the analysis.

Cheers,

Remco

< Phrissospongia_glandiformis_75.0=75.0,

< Leiocarenas_planus_155.5=155.5,

< Pachinion_scriptum_84.9=84.9,

< Procorallistes_polymorphus_83.6=83.6,

< Geodia_hopetouni_37.8=37.8,

< Crellastrina_sp_37.8=37.8,

< Tethya_sp_37.8=37.8,

< Paleospongilla_chubutensis_100.0=100.0,

< Mastosia_wetzleri_155.5=155.5                <taxa id="TaxonSet.b_Demo_prot_fossils" spec="TaxonSet">

---

> Phrissospongia_glandiformis_75.0=0.750,

> Leiocarenas_planus_155.5=1.555,

> Pachinion_scriptum_84.9=0.849,

> Procorallistes_polymorphus_83.6=0.836,

> Geodia_hopetouni_37.8=0.378,

> Crellastrina_sp_37.8=0.378,

> Tethya_sp_37.8=0.378,

> Paleospongilla_chubutensis_100.0=1.000,

> Mastosia_wetzleri_155.5=1.555                <taxa id="TaxonSet.b_Demo_prot_fossils" spec="TaxonSet">

132c132

<         <parameter id="originFBD.t:demoTree" lower="252.0" name="stateNode" upper="201.0">252.0</parameter>

---

>         <parameter id="originFBD.t:demoTree" upper="2.520" name="stateNode" lower="2.010">2.520</parameter>

139c139

<             <parameter id="randomPopSize.t:demoTree" name="popSize">1.0</parameter>

---

>             <parameter id="randomPopSize.t:demoTree" name="popSize">0.100</parameter>

155c155

<                 <LogNormal id="LogNormalDistributionModel.0" M="@parameter.hyperLogNormalDistributionModel-M-originPriorFBD.t:demoTree" S="@parameter.hyperLogNormalDistributionModel-S-originPriorFBD.t:demoTree" meanInRealSpace="true" name="distr" offset="252.0"/>

---

>                 <LogNormal id="LogNormalDistributionModel.0" M="@parameter.hyperLogNormalDistributionModel-M-originPriorFBD.t:demoTree" S="@parameter.hyperLogNormalDistributionModel-S-originPriorFBD.t:demoTree" meanInRealSpace="true" name="distr" offset="2.010"/>