DAISIE_MW_ML: Maximization of the loglikelihood under the DAISIE model with...

In xieshu95/Trait_dependent_TraiSIE: Dynamical Assembly of Islands by Speciation, Immigration and Extinction

Description

This function computes the maximum likelihood estimates of the parameters of the relationships between parameters of the DAISIE model with clade-specific diversity-dependence and island area and distance of the island to the mainlandor nearest landmass, for data from lineages colonizing several islands/archipelagos. It also outputs the corresponding loglikelihood that can be used in model comparisons.

A note on the sigmoidal functions used in distance_dep: For anagenesis and cladogenesis, the functional relationship is k * (d/d0)^x/(1 + (d/d0)^x); for colonization the relationship is: k - k * (d/d0)^x/(1 + (d/d0)^x). The d0 parameter is the 11th parameter entered. In the of 'sigmoidal_col_ana', the 11th parameter is the d0 for colonization and the 12th is the d0 for anagenesis.

Usage

DAISIE_MW_ML(datalist, initparsopt, idparsopt, parsfix, idparsfix,
  res = 100, ddmodel = 11, cond = 0, island_ontogeny = NA,
  tol = c(1e-04, 1e-05, 1e-07), maxiter = 1000 *
  round((1.25)^length(idparsopt)), methode = "lsodes",
  optimmethod = "subplex", CS_version = 1, verbose = 0,
  tolint = c(1e-16, 1e-10), distance_type = "continent",
  distance_dep = "power", parallel = "local", cpus = 3)

Arguments

datalist	Data object containing information on colonisation and branching times. This object can be generated using the DAISIE_dataprep function, which converts a user-specified data table into a data object, but the object can of course also be entered directly. It is an R list object with the following elements. The first element of the list has two three components: $island_age - the island age Then, depending on whether a distinction between types is made, we have: $not_present - the number of mainland lineages that are not present on the island The remaining elements of the list each contains information on a single colonist lineage on the island and has 5 components: $colonist_name - the name of the species or clade that colonized the island $branching_times - island age and stem age of the population/species in the case of Non-endemic, Non-endemic_MaxAge and Endemic anagenetic species. For cladogenetic species these should be island age and branching times of the radiation including the stem age of the radiation. $stac - the status of the colonist * Non_endemic_MaxAge: 1 * Endemic: 2 * Endemic&Non_Endemic: 3 * Non_endemic: 4 * Endemic_MaxAge: 5 $missing_species - number of island species that were not sampled for particular clade (only applicable for endemic clades)
initparsopt	The initial values of the parameters that must be optimized; they are all positive
idparsopt	The ids of the parameters that must be optimized. The ids are defined as follows: id = 1 corresponds to lambda^c0 (cladogenesis rate for unit area) id = 2 corresponds to y (exponent of area for cladogenesis rate) id = 3 corresponds to mu0 (extinction rate for unit area) id = 4 corresponds to x (exponent of 1/area for extinction rate) id = 5 corresponds to K0 (clade-level carrying capacity for unit area) id = 6 corresponds to z (exponent of area for clade-level carrying capacity) id = 7 corresponds to gamma0 (immigration rate for unit distance) id = 8 corresponds to alpha (exponent of 1/distance for immigration rate) id = 9 corresponds to lambda^a0 (anagenesis rate for unit distance) id = 10 corresponds to beta (exponent of 1/distance for anagenesis rate)
parsfix	The values of the parameters that should not be optimized
idparsfix	The ids of the parameters that should not be optimized, e.g. c(1,3) if lambda^c and K should not be optimized.
res	Sets the maximum number of species for which a probability must be computed, must be larger than the size of the largest clade
ddmodel	Sets the model of diversity-dependence: ddmodel = 0 : no diversity dependence ddmodel = 1 : linear dependence in speciation rate ddmodel = 11: linear dependence in speciation rate and in immigration rate ddmodel = 2 : exponential dependence in speciation rate ddmodel = 21: exponential dependence in speciation rate and in immigration rate
cond	cond = 0 : conditioning on island age cond = 1 : conditioning on island age and non-extinction of the island biota
island_ontogeny	type of island ontonogeny. If NA, then constant ontogeny is assumed
tol	Sets the tolerances in the optimization. Consists of: reltolx = relative tolerance of parameter values in optimization reltolf = relative tolerance of function value in optimization abstolx = absolute tolerance of parameter values in optimization
maxiter	Sets the maximum number of iterations in the optimization
methode	Method of the ODE-solver. See package deSolve for details. Default is "lsodes"
optimmethod	Method used in likelihood optimization. Default is "subplex" (see subplex package). Alternative is 'simplex' which was the method in previous versions.
CS_version	For internal testing purposes only. Default is 1, the original DAISIE code.
verbose	sets whether parameters and likelihood should be printed (1) or not (0)
tolint	Vector of two elements containing the absolute and relative tolerance of the integration
distance_type	Use 'continent' if the distance to the continent should be used, use 'nearest_big' if the distance to the nearest big landmass should be used, and use 'biologically_realistic' if the distance should take into account some biologically realism, e.g. an average of the previous two if both are thought to contribute.
distance_dep	Sets what type of distance dependence should be used. Default is a power law, denoted as 'power'. Alternatives are an exponantial relationship denoted by 'exp' or sigmoids, either 'sigmoidal_col' for a sigmoid in the colonization, 'sigmoidal_ana' for sigmoidal anagenesis, 'sigmoidal_clado' for sigmoidal cladogenesis, and 'sigmoidal_col_ana' for signoids in both colonization and anagenesis.
parallel	Sets whether parallel computation should be used. Use 'no' if no parallel computing should be used, 'cluster' for parallel computing on a unix/linux cluster, and 'local' for parallel computation on a local machine.
cpus	Number of cpus used in parallel computing. Default is 3. Will not have an effect if parallel = 'no'.

Value

The output is a dataframe containing estimated parameters and maximum loglikelihood.

lambda_c0	gives the maximum likelihood estimate of lambda^c, the rate of cladogenesis for unit area
y	gives the maximum likelihood estimate of y, the exponent of area for the rate of cladogenesis
mu0	gives the maximum likelihood estimate of mu0, the extinction rate
x	gives the maximum likelihood estimate of x, the exponent of 1/area for the extinction rate
K0	gives the maximum likelihood estimate of K0, the carrying-capacity for unit area
z	gives the maximum likelihood estimate of z, the exponent of area for the carrying capacity
gamma0	gives the maximum likelihood estimate of gamma0, the immigration rate for unit distance
y	gives the maximum likelihood estimate of alpha, the exponent of 1/distance for the rate of colonization
lambda_a0	gives the maximum likelihood estimate of lambda^a0, the rate of anagenesis for unit distance
beta	gives the maximum likelihood estimate of beta, the exponent of 1/distance for the rate of anagenesis
loglik	gives the maximum loglikelihood
df	gives the number of estimated parameters, i.e. degrees of feedom
conv	gives a message on convergence of optimization; conv = 0 means convergence

Author(s)

Rampal S. Etienne

References

Valente, L.M., A.B. Phillimore and R.S. Etienne (2015). Equilibrium and non-equilibrium dynamics simultaneously operate in the Galapagos islands. Ecology Letters 18: 844-852. <DOI:10.1111/ele.12461>.

See Also

DAISIE_ML_CS,

Examples

cat('No examples')

xieshu95/Trait_dependent_TraiSIE documentation built on Nov. 22, 2019, 7:51 a.m.