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

[Olga Morozova]

Merry Christmas, Everybody!

I am a BEAST 2.4.7. user. I generated an XML file with 208 sequences and tried to run it, however, I got an error below.

The error appears when I add a discrete trait.  Also, It's OK with small data sets.

Start likelihood: -Infinity after 10 initialisation attempts

P(posterior) = -Infinity (was -Infinity)

            P(prior) = -Infinity (was -Infinity)

                        P(BayesianSkyline.t:NN_VP8_P8_ALLSEQ) = -1230.1344228825812 (was -1230.1344228825812)

                        P(MarkovChainedPopSizes.t:NN_VP8_P8_ALLSEQ) = -33.700856263602155 (was -33.700856263602155)

                        P(GammaShapePrior.s:NN_VP8_P8_ALLSEQ) = -1.0 (was -1.0)

                        P(KappaPrior.s:NN_VP8_P8_ALLSEQ) = -1.8653600339742873 (was -1.8653600339742873)

                        P(nonZeroRatePrior.s:location) = -Infinity (was -Infinity)

                        P(relativeGeoRatesPrior.s:location) = -299.99999999999994 (was -299.99999999999994)

                        P(geoclockPrior.c:location) = -6.915086640662835 (was -6.915086640662835)

                        P(MeanRatePrior.c:NN_VP8_P8_ALLSEQ) = 0.0 (was 0.0)

                        P(ucldStdevPrior.c:NN_VP8_P8_ALLSEQ) = 0.819364145120802 (was 0.819364145120802)

            P(likelihood) = NaN (was NaN)  **

                        P(treeLikelihood.NN_VP8_P8_ALLSEQ) = NaN (was NaN)  **

                        P(traitedtreeLikelihood.location) = NaN (was NaN)  **

Fatal exception: Could not find a propjava.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.

Many thanks for any help,

Olga

[Alexei Drummond]

Dear Olga,

Merry Christmas!

You have a prior on the number of non-zero location transition rates which is Poisson with a mean of 0.693 and an offset of 24. This is specified here in the code:

            <prior id="nonZeroRatePrior.s:location" name="distribution">

                <x id="nonZeroRates.s:location" spec="util.Sum">

                    <arg idref="rateIndicator.s:location"/>

                </x>

                <distr id="Poisson.0" spec="beast.math.distributions.Poisson" offset="24.0">

                    <parameter id="RealParameter.10" estimate="false" name="lambda">0.693</parameter>

                </distr>

            </prior>

But you initialise the 300 dimension rateIndicator.s:location parameter with all the indicators to true:

        <stateNode id="rateIndicator.s:location" spec="parameter.BooleanParameter" dimension="300">true</stateNode>

The poisson probability of a value of 276 (300-24) with a lambda parameter of 0.693 is too small to be calculated efficiently numerically, so it gets numerically approximated as zero. You can’t start an MCMC run with a state that has probability zero. So you have two options:

(1) Initialise the rateIndicator.s:location with an explicit initialisation string that contains about 30-40 trues and 260-270 falses.

(2) Change the poisson prior lambda and/or offset parameters.

Personally I think that your prior is a bit too strong. So you could probably increase lambda quite a lot. But you may also want to change the initial state to start with a number of indicators closer to what your prior is suggesting.

Cheers

Alexei

--
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.
<NN_VP8_P8_ALLSEQ.xml>

[Olga Morozova]

Dear Prof Drummond

 

It is run now!

Thanks a lot for your recommendation!

 

Olga

[Taehyung Kwon]

Dear Dr.Drummond

I was curious why I kept failing on running BEAST for discrete phylogeography.

I have a same problem as Olga had, after I added discrete trait.

Since I did not understand some of those priors, I don't know what to adjust in my xml file.

Currently I'm trying to adjust the poisson parameters for "nonZeroRates.s:Location", but it doesn't go well as I thought (I changed the lambda value from 0.693 to 15.0, as the previous post).

I'll keep trying, and i would be really thankful for any kinds of advices.

Here, I attached my error and xml file.

Have a great day!

Best regards,

Taehyung

Start likelihood: -Infinity after 10 initialisation attempts

P(posterior) = -Infinity (was -Infinity)

        P(prior) = -Infinity (was -Infinity)

                P(CoalescentConstant.t:tree) = -872.6037354912571 (was -872.6037354912571)

                P(ClockPrior.c:clock) = -5.919035481085749 (was -5.919035481085749)

                P(GammaShapePrior.s:cds) = -1.0 (was -1.0)

                P(GammaShapePrior.s:cr) = -1.0 (was -1.0)

                P(MutationRatePrior.s:cds) = 0.5412848312506993 (was 0.5412848312506993)

                P(MutationRatePrior.s:cr) = -2.090999490513565 (was -2.090999490513565)

                P(nonZeroRatePrior.s:Location) = -Infinity (was -Infinity)

                P(HyperPrior.hyperPoisson-lambda-nonZeroRatePrior.s:Location) = -2.70805020110221 (was -2.70805020110221)

                P(PopSizePrior.t:tree) = 1.2039728043259361 (was 1.2039728043259361)

                P(PropInvariantPrior.s:cds) = 0.0 (was 0.0)

                P(PropInvariantPrior.s:cr) = 0.0 (was 0.0)

                P(RateACPrior.s:cds) = -3.184008455701433 (was -3.184008455701433)

                P(RateACPrior.s:cr) = -3.184008455701433 (was -3.184008455701433)

                P(RateAGPrior.s:cds) = -3.1686658147294304 (was -3.1686658147294304)

                P(RateAGPrior.s:cr) = -3.1686658147294304 (was -3.1686658147294304)

                P(RateATPrior.s:cds) = -3.184008455701433 (was -3.184008455701433)

                P(RateATPrior.s:cr) = -3.184008455701433 (was -3.184008455701433)

                P(RateCGPrior.s:cds) = -3.184008455701433 (was -3.184008455701433)

                P(RateCGPrior.s:cr) = -3.184008455701433 (was -3.184008455701433)

                P(RateCTPrior.s:cds) = -3.1686658147294304 (was -3.1686658147294304)

                P(RateCTPrior.s:cr) = -3.1686658147294304 (was -3.1686658147294304)

                P(relativeGeoRatesPrior.s:Location) = -629.9999999999999 (was -629.9999999999999)

                P(geoclockPrior.c:Location) = -6.915086640662835 (was -6.915086640662835)

                P(PQTI.prior) = NaN (was NaN)  **

        P(likelihood) = NaN (was NaN)  **

                P(treeLikelihood.cds) = NaN (was NaN)  **

                P(treeLikelihood.cr) = NaN (was NaN)  **

                P(traitedtreeLikelihood.Location) = NaN (was NaN)  **

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

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)

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

java.lang.RuntimeException: An error was encounted. Terminating BEAST

        at beast.app.util.ErrorLogHandler.publish(Unknown Source)

        at java.util.logging.Logger.log(Logger.java:738)

        at java.util.logging.Logger.doLog(Logger.java:765)

        at java.util.logging.Logger.log(Logger.java:788)

        at java.util.logging.Logger.severe(Logger.java:1463)

[Remco Bouckaert]

Hi Taehyung,

Perhaps this helps: http://www.beast2.org/2018/07/04/fatal-errors.html under "Ancestral ancestor analysis with too many states”

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.

<phg.zip>

[Taehyung Kwon]

Thank you for taking your time!

Considering number of states in my xml (around 400), I adjusted lambda to 200 and offset to 0.

It works, but ESS seems too low for now. Maybe I need to simplify my discrete traits.

That helped me a lot. Thanks again. 

Best regards,

Taehyung Kwon

Ph.D. candidate 

Laboratory of Bioinformatics and Population genetics

Seoul National University

Seoul, South Korea 

ted...@gmail.com | (+82-2-880-4822) | (+82-10-9903-3465) | http://biopop.snu.ac.kr

2018년 8월 17일 (금) 오전 3:28, Remco Bouckaert <higg...@gmail.com>님이 작성:

[marybe...@gmail.com]

Hello Dr. Bouckaert,

I'm transitioning over to BEAST2 from BEAST1 and aiming to run a dating analysis for 95 taxa using ~20 genes and eight calibrations under a relaxed uncorrelated lognormal clock. I am trying to use a starting tree (though not fixing the topology) for this relatively large analysis as a way to speed up convergence but cannot get it running.

I made sure that the starting tree is compatible with the calibrations by checking for monophyly issues and used bladj to adjust tree node heights. Based on BEAST documentation, I see that the error message I get (please see below) says there's a problem with P(1_AngiospermCrown.prior) = -Infinity (was -Infinity), and was able to run an analysis so when I removed this calibration. However, this is a key calibration for the root of the tree and so I would really like to include it. I would really appreciate guidance for troubleshooting. Apologies in advance if it's something obvious that I missed!

Start likelihood: -Infinity after 10 initialisation attempts

P(posterior) = -Infinity (was -Infinity)

P(prior) = -Infinity (was -Infinity)

P(BirthDeath.t:MonocotTree) = -780.1238207723977 (was -780.1238207723977)

P(BirthRatePrior.t:MonocotTree) = 0.0 (was 0.0)

P(DeathRatePrior.t:MonocotTree) = 0.0 (was 0.0)

P(GammaShapePrior.s:pos1) = -1.0 (was -1.0)

P(GammaShapePrior.s:pos2) = -1.0 (was -1.0)

P(GammaShapePrior.s:pos3) = -1.0 (was -1.0)

P(RateACPrior.s:pos1) = -0.6137056388801093 (was -0.6137056388801093)

P(RateACPrior.s:pos2) = -0.6137056388801093 (was -0.6137056388801093)

P(RateACPrior.s:pos3) = -0.6137056388801093 (was -0.6137056388801093)

P(RateAGPrior.s:pos1) = -0.6137056388801093 (was -0.6137056388801093)

P(RateAGPrior.s:pos2) = -0.6137056388801093 (was -0.6137056388801093)

P(RateAGPrior.s:pos3) = -0.6137056388801093 (was -0.6137056388801093)

P(RateATPrior.s:pos1) = -0.6137056388801093 (was -0.6137056388801093)

P(RateATPrior.s:pos2) = -0.6137056388801093 (was -0.6137056388801093)

P(RateATPrior.s:pos3) = -0.6137056388801093 (was -0.6137056388801093)

P(RateCGPrior.s:pos1) = -0.6137056388801093 (was -0.6137056388801093)

P(RateCGPrior.s:pos2) = -0.6137056388801093 (was -0.6137056388801093)

P(RateCGPrior.s:pos3) = -0.6137056388801093 (was -0.6137056388801093)

P(RateGTPrior.s:pos1) = -0.6137056388801093 (was -0.6137056388801093)

P(RateGTPrior.s:pos2) = -0.6137056388801093 (was -0.6137056388801093)

P(RateGTPrior.s:pos3) = -0.6137056388801093 (was -0.6137056388801093)

P(MeanRatePrior.c:pos1) = -2.402585092994046 (was -2.402585092994046)

P(MeanRatePrior.c:pos2) = -2.402585092994046 (was -2.402585092994046)

P(MeanRatePrior.c:pos3) = -2.402585092994046 (was -2.402585092994046)

P(ucldStdevPrior.c:pos1) = 0.819364145120802 (was 0.819364145120802)

P(ucldStdevPrior.c:pos2) = 0.819364145120802 (was 0.819364145120802)

P(ucldStdevPrior.c:pos3) = 0.819364145120802 (was 0.819364145120802)

P(1_AngiospermCrown.prior) = -Infinity (was -Infinity)

P(2_MonocotsStem.prior) = 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.

NaN (was NaN)  **

P(3_LemnoideaeStem.prior) = NaN (was NaN)  **

P(4_DioscoreaCrown.prior) = NaN (was NaN)  **

P(5_TriuridaceaeStem.prior) = NaN (was NaN)  **

P(6_CyclanthusStem.prior) = NaN (was NaN)  **

P(7_CoryphoideaeStem.prior) = NaN (was NaN)  **

P(8_TyphaStem.prior) = NaN (was NaN)  **

P(likelihood) = NaN (was NaN)  **

P(treeLikelihood.pos1) = NaN (was NaN)  **

P(treeLikelihood.pos2) = NaN (was NaN)  **

P(treeLikelihood.pos3) = NaN (was NaN)  **

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]

Unfortunately, it appears that the starting tree is not compatible with the first calibration (and maybe others). Probably the easiest to solve this problem is using the RandomTree as starting tree. You can do this in BEAUti by using the “View/Show Starting Tree panel” menu and go to the starting tree panel, or by editing the XML in a text editor and replace this line:

    <init id="NewickTree.t:MonocotTree" spec="beast.util.TreeParser" IsLabelledNewick="true"...

with these two lines:

 <init id="RandomTree.t:tree" spec="beast.evolution.tree.RandomTree" estimate="false" initial="@Tree.t:MonocotTree”> 

 <populationModel spec='ConstantPopulation' popSize="1.0"/>

In general, it is good practice to run several MCMC runs with different random starting trees just to make sure your analysis does not depend on the starting state, and this should be feasible with just the 95 taxa and 4 partitions in the analysis you posted.

To speed up the analysis, you can try to run it faster (using threads, BEAGLE, more here: http://www.beast2.org/performance-suggestions/index.html) or by simplifying the model (replace the birth death model by a Yule model, and/or the GTR with HKY models) to see whether there is sufficient signal in the data before trying more complex models.

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.

<UCLN_startingtree_all.xml>

[marybel soto gomez]

Hi Dr. Bouckaert,

Thanks very much for your reply and your helpful suggestions. I wanted to let you know that I got the analysis to run by pruning several outgroups that were on very long branches and using a different calibration for the new root of the tree.

When you say that it's good practice to do multiple MCMC runs with different starting trees to make sure the analysis does not depend on the starting tree, is it possible that using a starting tree could affect estimated ages? Or the final tree topology? Or both? My reasoning for providing a starting tree was to speed up convergence, but also because I'm confident about the tree topology from likelihood and parsimony analyses with the same data. I figured that fixing the tree topology may be a bit riskier, so did not do that.

I will try to install BEAGLE on my computer and using CIPRES to speed up analyses.

Again, thank you very much for your help.

Marybel

You received this message because you are subscribed to a topic in the Google Groups "beast-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/beast-users/B1dB757O4l8/unsubscribe.
To unsubscribe from this group and all its topics, 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.

--

Marybel Soto Gomez

Ph.D. Student, Graham Lab

Botany Department

University of British Columbia

[Claudia Vaga]

Hi everyone,

I am a BEAST2 user and I am running in a similar error.

I have already tried to troubleshoot removing a prior regarding the root but it is still not working.

The error I am getting is the following:

Start likelihood: -Infinity after 10 initialisation attempts
P(posterior) = -Infinity (was -Infinity)
    P(prior) = -Infinity (was -Infinity)

        P(BirthDeath.t:treem) = -29380.15811341078 (was -29380.15811341078)
        P(BirthRatePrior.t:treem) = -6.907755278982137 (was -6.907755278982137)
        P(DeathRatePrior.t:treem) = 0.0 (was 0.0)
        P(RateACPrior.s:uce-112122) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-112189) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-112322) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-112515) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-112569) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-112674) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-113162) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-11319) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-114242) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-115112) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-115578) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-11985) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-2150) = -3.184008455701433 (was -3.184008455701433)
        P(RateACPrior.s:uce-255) = -3.184008455701433 (was -3.184008455701433)

.......

.......

.......

(MeanRatePrior.c:clockm) = -21.609518090779318 (was -21.609518090779318)
        P(ucldStdevPrior.c:clockm) = 0.0 (was 0.0)
        P(acropora.prior) = -1.7227666048839765 (was -1.7227666048839765)
        P(caryophyllia.prior) = -2.772588724739783 (was -2.772588724739783)
        P(dendrophylliidae.prior) = -2.5416019966141543 (was -2.5416019966141543)
        P(madrepora.prior) = -1.9600947882726252 (was -1.9600947882726252)
        P(oculina.prior) = -Infinity (was -Infinity)

    P(likelihood) = NaN (was NaN)  **

        P(treeLikelihood.uce-278) = NaN (was NaN)  **
        P(treeLikelihood.uce-524) = NaN (was NaN)  **
        P(treeLikelihood.uce-970) = NaN (was NaN)  **
        P(treeLikelihood.uce-112189) = NaN (was NaN)  **
        P(treeLikelihood.uce-2150) = NaN (was NaN)  **
        P(treeLikelihood.uce-99029913) = NaN (was NaN)  **
        P(treeLikelihood.uce-9901241) = NaN (was NaN)  **
        P(treeLikelihood.uce-112322) = NaN (was NaN)  **
        P(treeLikelihood.uce-114242) = NaN (was NaN)  **
        P(treeLikelihood.uce-990292913) = NaN (was NaN)  **
        P(treeLikelihood.uce-990440) = NaN (was NaN)  **
        P(treeLikelihood.uce-115578) = NaN (was NaN)  **
        P(treeLikelihood.uce-112569) = NaN (was NaN)  **
        P(treeLikelihood.uce-112122) = NaN (was NaN)  **
        P(treeLikelihood.uce-781) = NaN (was NaN)  **
        P(treeLikelihood.uce-990293354) = NaN (was NaN)  **
        P(treeLikelihood.uce-9901718) = NaN (was NaN)  **
        P(treeLikelihood.uce-990291573) = NaN (was NaN)  **
        P(treeLikelihood.uce-11985) = NaN (was NaN)  **
        P(treeLikelihood.uce-255) = NaN (was NaN)  **
        P(treeLikelihood.uce-11319) = NaN (was NaN)  **
        P(treeLikelihood.uce-115112) = NaN (was NaN)  **
        P(treeLikelihood.uce-990907) = NaN (was NaN)  **
        P(treeLikelihood.uce-99029792) = NaN (was NaN)  **
        P(treeLikelihood.uce-990505) = NaN (was NaN)  **
        P(treeLikelihood.uce-3911726) = NaN (was NaN)  **
        P(treeLikelihood.uce-112515) = NaN (was NaN)  **
        P(treeLikelihood.uce-9901308) = NaN (was NaN)  **
        P(treeLikelihood.uce-113162) = NaN (was NaN)  **
        P(treeLikelihood.uce-112674) = NaN (was NaN)  **

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

See http://www.beast2.org/2018/07/04/fatal-errors.html for other possible solutions.

java.lang.RuntimeException: Could not find a proper state to initialise. Perhaps try another seed.

See http://www.beast2.org/2018/07/04/fatal-errors.html for other possible solutions.

    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)

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

See http://www.beast2.org/2018/07/04/fatal-errors.html for other possible solutions.

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

Could someone help me troubleshoot this? And sorry in advance if it is something obvious!

[Tarani Jindal]

Hi Dr. Bouckaert,
I am a BEAST 2.4.7. user. I generated an XML file with 129 sequences and tried to run it, however, I got an error below.
The error appears when I add a newick tree to the MultiMonophyleticConstraint.

Start likelihood: -Infinity after 10 initialisation attempts
P(posterior) = -Infinity (was -Infinity)
P(prior) = -Infinity (was -Infinity)

P(FBD.t:part1) = -263.6778938478095 (was -263.6778938478095)
P(MultiMonophyleticConstraint.t:part1) = -Infinity (was -Infinity)
P(ORCRatePriorDistribution.c:part1) = -1095.793286120492 (was -1095.793286120492)
P(ORCsigmaPrior.c:part1) = 1.3628558876856076 (was 1.3628558876856076)
P(ORCucldMeanPrior.c:part1) = -2.7232296703330143 (was -2.7232296703330143)
P(diversificationRatePriorFBD.t:part1) = 0.0 (was 0.0)
P(GammaShapePrior.s:part1) = -1.0 (was -1.0)
P(GammaShapePrior.s:part2) = -1.0 (was -1.0)
P(GammaShapePrior.s:part3) = -1.0 (was -1.0)
P(KappaPrior.s:part2) = -1.8653600339742873 (was -1.8653600339742873)
P(originPriorFBD.t:part1) = 0.0 (was 0.0)
P(RateACPrior.s:part1) = -3.184008455701433 (was -3.184008455701433)
P(RateACPrior.s:part3) = -3.184008455701433 (was -3.184008455701433)
P(RateAGPrior.s:part1) = -3.1686658147294304 (was -3.1686658147294304)
P(RateAGPrior.s:part3) = -3.1686658147294304 (was -3.1686658147294304)
P(RateATPrior.s:part1) = -3.184008455701433 (was -3.184008455701433)
P(RateATPrior.s:part3) = -3.184008455701433 (was -3.184008455701433)
P(RateCGPrior.s:part1) = -3.184008455701433 (was -3.184008455701433)
P(RateCGPrior.s:part3) = -3.184008455701433 (was -3.184008455701433)
P(RateGTPrior.s:part1) = -3.184008455701433 (was -3.184008455701433)
P(RateGTPrior.s:part3) = -3.184008455701433 (was -3.184008455701433)
P(samplingProportionPriorFBD.t:part1) = 0.0 (was 0.0)
P(turnoverPriorFBD.t:part1) = 0.0 (was 0.0)
P(DM1.prior) = NaN (was NaN)  **
P(VP1.prior) = NaN (was NaN)  **

P(likelihood) = NaN (was NaN)  **

P(treeLikelihood.part1) = NaN (was NaN)  **
P(treeLikelihood.part2) = NaN (was NaN)  **
P(treeLikelihood.part3) = NaN (was NaN)  **

java.lang.RuntimeException: Could not find a proper state to initialise. Perhaps try another seed.