Multiple Monophyletic Constraint doesn't work

[Humberto Nappo]

Hi,

I'm trying to run BEAST (v2.7.2) with a multiple monophyletic constraint, but I keep receiving the error message below. I tried changing other parameters, but the same error occurs every time (and only when) I try to impose this constraint, regardless of other changes. I tried to use AlmostMultiMRCAPrior from AlmostDistributions package (with BEAST v2.6.7), so the run could start. It did, but even with penalties, the trees I got were contradictory with the constraint. Defining a starting tree (congruent with the constraint), didn't work as well. I also tried to run it in BEAST 2.6.7. without using AlmostMultiMRCAPrior, but I got the same error message. Using MRCA priors works, though, but it would be prohibitive to define the amount of MRCA priors needed to cover the information on the tree I was trying to use as constraint. Someone has a clue of what may be happening?

Start likelihood: -Infinity after 10 initialisation attempts
P(posterior) = -Infinity (was -Infinity)
    P(prior) = -Infinity (was -Infinity)
        P(YuleModel.t:Concatenated_Alignment) = -10.738813436382559 (was -10.738813436382559)
        P(BMT_PropInvariablePrior.s:Concatenated_Alignment) = 1.0702128141464131 (was 1.0702128141464131)
        P(BMT_RatesPrior.s:Concatenated_Alignment) = -2.462082084518882 (was -2.462082084518882)
        P(BMT_GammaShapePrior.s:Concatenated_Alignment) = -1.0 (was -1.0)
        P(MultiMonophyleticConstraint.t:Concatenated_Alignment) = -Infinity (was -Infinity)
        P(YuleBirthRatePrior.t:Concatenated_Alignment) = 0.0 (was 0.0)
    P(likelihood) = NaN (was NaN)  **
        P(treeLikelihood.Concatenated_Alignment) = NaN (was NaN)  **
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.base.inference.MCMC.run(Unknown Source)
    at beastfx.app.beast.BeastMCMC.run(Unknown Source)
    at beastfx.app.beast.Controller$2.run(Unknown Source)

[lilne...@gmail.com]

Did ever got this fixed? I am having the same issue and can't seem to get it to run and encounter the same problem where i get -Infinity

[Kenzie Grover]

I also ran into this issue. I believe it has to do with the topology constraint barring the analysis from getting started, and you just need to define a starting tree. The starting tree tab is now in v2.7.5. 

[Humberto Nappo]

Yeah, this indeed solved the problem. If you have a constraint tree as a multi monophyletic constraint (as it was my case), the option Simple Constrained Random Tree seems to work fine.

--
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/6gibcw_8B-M/unsubscribe.
To unsubscribe from this group and all its topics, send an email to beast-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/e6989f62-43cb-4368-9971-2515b9ed9c47n%40googlegroups.com.

--

Humberto C. Nappo

Graduated in Biological Sciences and master's student in Ecology by Universidade de Brasília

Graduado em Ciências Biológicas e mestrando em Ecologia pela Universidade de Brasília

ORCID ID: https://orcid.org/0000-0001-7810-1635

Lattes CV: http://lattes.cnpq.br/8150109942426326