ape (version 4.0)

rTraitMult: Multivariate Character Simulation

Description

This function simulates the evolution of a multivariate set of traits along a phylogeny. The calculation is done recursively from the root.

Usage

rTraitMult(phy, model, p = 1, root.value = rep(0, p), ancestor = FALSE, asFactor = NULL, trait.labels = paste("x", 1:p, sep = ""), ...)

Arguments

phy

an object of class "phylo".

model

a function specifying the model (see details).

p

an integer giving the number of traits.

root.value

a numeric vector giving the values at the root.

ancestor

a logical value specifying whether to return the values at the nodes as well (by default, only the values at the tips are returned).

asFactor

the indices of the traits that are returned as factors (discrete traits).

trait.labels

a vector of mode character giving the names of the traits.

...

further arguments passed to model if it is a function.

Value

A data frame with p columns whose names are given by trait.labels and row names taken from the labels of the tree.

Details

The model is specified with an R function of the form foo(x, l) where x is a vector of the traits of the ancestor and l is the branch length. Other arguments may be added. The function must return a vector of length p.

See Also

rTraitCont, rTraitDisc, ace

Examples

## correlated evolution of 2 continuous traits:
mod <- function(x, l) {
    y1 <- rnorm(1, x[1] + 0.5*x[2], 0.1)
    y2 <- rnorm(1, 0.5*x[1] + x[2], 0.1)
    c(y1, y2)
}
set.seed(11)
tr <- makeNodeLabel(rcoal(20))
x <- rTraitMult(tr, mod, 2, ancestor = TRUE)
op <- par(mfcol = c(2, 1))
plot(x, type = "n")
text(x, labels = rownames(x), cex = 0.7)
oq <- par(mar = c(0, 1, 0, 1), xpd = TRUE)
plot(tr, font = 1, cex = 0.7)
nodelabels(tr$node.label, cex = 0.7, adj = 1)
par(c(op, oq))