pez (version 1.2-4)

sim.phy: Simulate phylogenies

Description

Simulate phylogenies under pure birth/death or as a function of trait evolution

Usage

sim.bd.phy(speciate = 0.1, extinction = 0.025, time.steps = 20)
sim.bd.tr.phy(
  speciate = 0.1,
  extinction = 0.025,
  time.steps = 20,
  tr.range = c(0, 1),
  sp.tr = 2,
  ext.tr = 1,
  tr.walk = 0.2,
  tr.wrap = TRUE
)
edge2phylo(edge, s, e = numeric(0), el = NA, t = NULL)

Value

phylo object with random tip.labels, and trait values if using sim.br.tr.tree.

Arguments

speciate

probability each species will speciate in each time-step (0-1)

extinction

probability each species will go extinct in each time-step (0-1)

time.steps

number of time-steps for simulation

tr.range

vector of length two specifying boundaries for trait values (see notes); initial two species will be at the 25th and 75th percentiles of this space. See also tr.wrap

sp.tr

speciation rate's interaction with the minimum distance between a species and the species most similar to it (see details)

ext.tr

extinction rate's interaction with the minimum distance between a species and the species most similar to it (see details)

tr.walk

at each time-step a species not undergoing speciation or extinction has its trait value drawn from a distribution centered at its current value and with a standard deviation set by this value. I.e., this is the rate of the Brownian motion trait evolution.

tr.wrap

whether to force species' trait values to stay within the boundary defined by tr.range; default TRUE.

edge

a two-column matrix where the first column is the start node, the second the destination, as in phylo$edge

s

which of the rows in the edge matrix represent extant species

e

which of the tips in the edge matrix are extinct (DEFAULT: empty vector, i.e., none)

el

a vector to be used to give edge.length to the phylogeny (default NA, i.e., none)

t

if given (default NA), a vector to be used for traits ($traits slot) in the phylogeny

Author

Will Pearse

Will Pearse

Will Pearse

Details

sim.bd.tree simulates a pure birth/death speciation model. There are two important things to note: (1) speciation is randomised before extinction, and only one thing can happen to a lineage per timestep. (2) This code works well for my purposes, but absurd parameter values can cause the function to crash.

sim.bd.tr.tree is an extension of sim.bd.tree, and all its caveats apply to it. It additionally simulated the evolution of a trait under Brownain motion (tr.walk). Species' speciation/extinction rates change depending on whether they have a trait value similar to other species (sp.tr, ext.tr). When a speciation event happens, the two daughters split evenly about the ancestor's trait value, taking values half-way to whatever the nearest species' value is. To be precise: \(p(speciate)_i = speciate_i + sp.tr \times min(trait distance)\), \(p(extinct)_i = exinction_i + ext.tr \times min(trait distance)\), where \(i\) denotes each species.

edge2phylo is an internal function for the sim.phy and sim.meta function families, which may be of use to you. Check those functions' code for examples of use.

These functions are closely related to sim.meta; the latter are extensions that simulate meta-community structure at the same time.

See Also

sim.meta scape

Examples

tree <- sim.bd.phy(0.1, 0, 10)
plot(tree)