beast-dev / beast-mcmc

Bayesian Evolutionary Analysis Sampling Trees
http://beast.community
GNU Lesser General Public License v2.1
182 stars 71 forks source link

Beast 1.10.4 program error message Total = - Infinity #1144

Closed luguohui1999 closed 1 year ago

luguohui1999 commented 1 year ago

When I run Beast1.10.4, the program stops and the following information is displayed: Could you help me to solve the problem? Using BEAGLE resource 0: CPU with instance flags: PRECISION_DOUBLE COMPUTATION_SYNCH EIGEN_REAL SCALING_MANUAL SCALERS_RAW VECTOR_SSE THREADING_CPP PROCESSOR_CPU FRAMEWORK_CPU Ignoring ambiguities in tree likelihood. With 62 unique site patterns. Using rescaling scheme : dynamic (rescaling every 100 evaluations, delay rescaling until first overflow)

Using TreeDataLikelihood Branch rate model used: discretizedBranchRates Creating swap operator for parameter branchRates.categories (weight=10.0) Parameter weights for delta exchange are: 657 925 559 751 551 740 1129 953 752 942 1456
Optimization Schedule: default

Likelihood computation is using an auto sizing thread pool.

Creating the MCMC chain: chain length = 1000000 operator adaption = true adaptation delayed for 10000 steps WARNING: Likelihood component, null, created but not used in the MCMC WARNING: Likelihood component, treeLikelihood, created but not used in the MCMC Dec 20, 2022 5:08:52 PM dr.app.beast.BeastMain SEVERE: Error running file: Cco_2022_yule.xml Dec 20, 2022 5:08:52 PM dr.app.beast.BeastMain SEVERE: The initial model is invalid because state has a zero probability.

If the log likelihood of the tree is -Inf, this may be because the initial, random tree is so large that it has an extremely bad likelihood which is being rounded to zero.

Alternatively, it may be that the product of starting mutation rate and tree height is extremely small or extremely large.

Finally, it may be that the initial state is incompatible with one or more 'hard' constraints (on monophyly or bounds on parameter values. This will result in Priors with zero probability.

The individual components of the posterior are as follows: The initial posterior is zero: CompoundLikelihood(compoundModel)=( BooleanLikelihood=0.0, Truncated(tmrca(Cco))=-Inf, LogNormal(accD_psaI.kappa)=-1.8654, LogNormal(petA_psbE.kappa)=-1.8654, Exponential(petA_psbE.alpha)=-0.3069, LogNormal(petB_petD.kappa)=-1.8654, Exponential(petB_petD.alpha)=-0.3069, Gamma(psbA_TrnH_GUG.gtr.AC)=-3.184, Gamma(psbA_TrnH_GUG.gtr.AG)=-3.1687, Gamma(psbA_TrnH_GUG.gtr.AT)=-3.184, Gamma(psbA_TrnH_GUG.gtr.CG)=-3.184, Gamma(psbA_TrnH_GUG.gtr.GT)=-3.184, Gamma(psbB_psbH.gtr.AC)=-3.184, Gamma(psbB_psbH.gtr.AG)=-3.1687, Gamma(psbB_psbH.gtr.AT)=-3.184, Gamma(psbB_psbH.gtr.CG)=-3.184, Gamma(psbB_psbH.gtr.GT)=-3.184, Exponential(psbB_psbH.alpha)=-0.3069, LogNormal(rpoC1_rpoB.kappa)=-1.8654, Exponential(rpoC1_rpoB.alpha)=-0.3069, Gamma(rps18_clpP.gtr.AC)=-3.184, Gamma(rps18_clpP.gtr.AG)=-3.1687, Gamma(rps18_clpP.gtr.AT)=-3.184, Gamma(rps18_clpP.gtr.CG)=-3.184, Gamma(rps18_clpP.gtr.GT)=-3.184, Exponential(rps18_clpP.alpha)=-0.3069, LogNormal(rps8_rpl16.kappa)=-1.8654, Exponential(rps8_rpl16.alpha)=-0.3069, LogNormal(trnK_UUU_rps16.kappa)=-1.8654, Exponential(trnK_UUU_rps16.alpha)=-0.3069, Gamma(trnL_UAA_trnF_GAA.gtr.AC)=-3.184, Gamma(trnL_UAA_trnF_GAA.gtr.AG)=-3.1687, Gamma(trnL_UAA_trnF_GAA.gtr.AT)=-3.184, Gamma(trnL_UAA_trnF_GAA.gtr.CG)=-3.184, Gamma(trnL_UAA_trnF_GAA.gtr.GT)=-3.184, LogNormal(trnS_GCT_trnR_TCT.kappa)=-1.8654, Exponential(trnS_GCT_trnR_TCT.alpha)=-0.3069, Uniform(allMus)=-2532.8436, Uniform(ucld.mean)=-230.2585, Uniform(ucld.stdev)=-230.2585, LogNormal(yule.birthRate)=-2.0385, BirthDeathGernhard08Model/yule(treeModel)=-5777.2021, DiscretizedBranchRates(discretizedBranchRates[branchRates])=-350.5621 Total = -Infinity ), CompoundLikelihood(compoundModel)=( TreeDataLikelihood(TreeDataLikelihood)=-15535.2854, TreeDataLikelihood(TreeDataLikelihood)=-19994.3218, TreeDataLikelihood(TreeDataLikelihood)=-12602.2418, TreeDataLikelihood(TreeDataLikelihood)=-31677.2163, TreeDataLikelihood(TreeDataLikelihood)=-12022.8554, TreeDataLikelihood(TreeDataLikelihood)=-16168.4752, TreeDataLikelihood(TreeDataLikelihood)=-25509.857, TreeDataLikelihood(TreeDataLikelihood)=-20643.8731, TreeDataLikelihood(TreeDataLikelihood)=-15803.3828, TreeDataLikelihood(TreeDataLikelihood)=-46280.7862, TreeDataLikelihood(TreeDataLikelihood)=-29191.6099 Total = -245429.9050874092 ) Total = -Infinity For more information go to http://beast.community.

rambaut commented 1 year ago

The initial tree is not compatible with the conditions of the prior for the TMRCA of Cco. Not a BEAST issue.