tess.analysis.efbd: tess.analysis.efbd

In hoehna/TESS: Diversification Rate Estimation and Fast Simulation of Reconstructed Phylogenetic Trees under Tree-Wide Time-Heterogeneous Birth-Death Processes Including Mass-Extinction Events

Description

Running an analysis on a given tree and estimating the diversification rates including rate-shifts and mass-extinction events.

Usage

tess.analysis.efbd(
  tree,
  initialSpeciationRate,
  initialExtinctionRate,
  initialFossilizationRate,
  empiricalHyperPriors = TRUE,
  empiricalHyperPriorInflation = 10,
  empiricalHyperPriorForm = c("lognormal", "normal", "gamma"),
  speciationRatePriorMean = 0,
  speciationRatePriorStDev = 1,
  extinctionRatePriorMean = 0,
  extinctionRatePriorStDev = 1,
  fossilizationRatePriorMean = 0,
  fossilizationRatePriorStDev = 1,
  initialSpeciationRateChangeTime = c(),
  initialExtinctionRateChangeTime = c(),
  initialFossilizationRateChangeTime = c(),
  estimateNumberSpeciationRateChanges = TRUE,
  estimateNumberExtinctionRateChanges = TRUE,
  estimateNumberFossilizationRateChanges = TRUE,
  numExpectedRateChanges = 2,
  samplingProbability = 1,
  samplingStrategy = "uniform",
  missingSpecies = c(),
  timesMissingSpecies = c(),
  tInitialMassExtinction = c(),
  pInitialMassExtinction = c(),
  pMassExtinctionPriorShape1 = 5,
  pMassExtinctionPriorShape2 = 95,
  estimateMassExtinctionTimes = TRUE,
  numExpectedMassExtinctions = 2,
  estimateNumberMassExtinctions = TRUE,
  MRCA = TRUE,
  CONDITION = "survival",
  BURNIN = 10000,
  MAX_ITERATIONS = 2e+05,
  THINNING = 100,
  OPTIMIZATION_FREQUENCY = 500,
  CONVERGENCE_FREQUENCY = 1000,
  MAX_TIME = Inf,
  MIN_ESS = 500,
  ADAPTIVE = TRUE,
  dir = "",
  priorOnly = FALSE,
  verbose = TRUE
)

Arguments

tree	the phylogeny
initialSpeciationRate	The initial value of the speciation rate when the MCMC is started.
initialExtinctionRate	The initial value of the extinction rate when the MCMC is started.
initialFossilizationRate	The initial value of the fossilization rate when the MCMC is started.
empiricalHyperPriors	.
empiricalHyperPriorInflation	.
empiricalHyperPriorForm	.
speciationRatePriorMean	mean parameter for the lognormal prior on lambda
speciationRatePriorStDev	standard deviation for the lognormal prior on lambda
extinctionRatePriorMean	mean parameter for the lognormal prior on mu
extinctionRatePriorStDev	standard deviation for the lognormal prior on mu
fossilizationRatePriorMean	mean parameter for the lognormal prior on phi
fossilizationRatePriorStDev	standard deviation for the lognormal prior on phi
initialSpeciationRateChangeTime	The initial value of the time points when speciation rate-shifts occur.
initialExtinctionRateChangeTime	The initial value of the time points when extinction rate-shifts occur.
initialFossilizationRateChangeTime	The initial value of the time points when fossilization rate-shifts occur.
estimateNumberSpeciationRateChanges	Do we estimate the number of rate shifts?
estimateNumberExtinctionRateChanges	Do we estimate the number of rate shifts?
estimateNumberFossilizationRateChanges	Do we estimate the number of rate shifts?
numExpectedRateChanges	Expected number of rate changes which follow a Poisson process.
samplingProbability	probability of uniform sampling at present
samplingStrategy	Which strategy was used to obtain the samples (taxa). Options are: uniform|diversified|age
missingSpecies	The number of species missed which originated in a given time interval (empirical taxon sampling)
timesMissingSpecies	The times intervals of the missing species (empirical taxon sampling)
tInitialMassExtinction	The initial value of the vector of times of the mass-extinction events.
pInitialMassExtinction	The initial value of the vector of survival probabilities of the mass-extinction events.
pMassExtinctionPriorShape1	The alpha (first shape) parameter of the Beta prior distribution for the survival probability of a mass-extinction event.
pMassExtinctionPriorShape2	The beta (second shape) parameter of the Beta prior distribution for the survival probability of a mass-extinction event.
estimateMassExtinctionTimes	Do we estimate the times of mass-extinction events?
numExpectedMassExtinctions	Expected number of mass-extinction events which follow a Poisson process.
estimateNumberMassExtinctions	Do we estimate the number of mass-extinction events?
MRCA	does the tree start at the mrca?
CONDITION	do we condition the process on nothing|survival|taxa?
BURNIN	number of starting iterations to be dropped
MAX_ITERATIONS	The maximum number of iteration of the MCMC.
THINNING	The frequency how often samples are recorded during the MCMC.
OPTIMIZATION_FREQUENCY	The frequency how often the MCMC moves are optimized
CONVERGENCE_FREQUENCY	The frequency how often we check for convergence?
MAX_TIME	The maximum time the MCMC is allowed to run in seconds.
MIN_ESS	The minimum number of effective samples (ESS) to assume convergence.
ADAPTIVE	Do we use auto-tuning of the MCMC moves?
dir	The subdirectory in which the output will be stored.
priorOnly	Do we sample from the prior only?
verbose	Do we want to print the progress of the MCMC?

Value

nothing. outputs are written to the file system

Examples

# we load the conifers as the test data set
data(conifers)

# for the conifers we know what the total number of species is
total <- 630
# thus, we can compute what the sampling fraction is
rho <- (conifers$Nnode+1)/total


# next, we specify the prior mean and standard deviation
# for the speciation and extinction rate
mu_lambda = 0.15
std_lambda = 0.02
mu_mu = 0.09
std_mu = 0.02
mu_phi = 0.10
std_phi = 0.02



# now we can run the entire analysis.
# note that a full analyses should be run much longer
tess.analysis.efbd( tree=conifers,
                    initialSpeciationRate = exp(mu_lambda),
                    initialExtinctionRate = exp(mu_mu),
                    initialFossilizationRate = exp(mu_phi),
                    empiricalHyperPriors = FALSE,
                    speciationRatePriorMean = mu_lambda,
                    speciationRatePriorStDev = std_lambda,
                    extinctionRatePriorMean = mu_mu,
                    extinctionRatePriorStDev = std_mu,
                    fossilizationRatePriorMean = mu_phi,
                    fossilizationRatePriorStDev = std_phi,
                    numExpectedRateChanges = 2,
                    samplingProbability = rho,
                    numExpectedMassExtinctions = 2,
                    BURNIN = 100,
                    MAX_ITERATIONS = 200,
                    THINNING = 10,
                    dir = "analysis_conifer")

hoehna/TESS documentation built on Feb. 3, 2022, 5:59 a.m.