bsd.dist: bsd.dist

In duchene/ClockstaRX: ClockstaRX: testing molecular clock hypotheses in genomic data

bsd.dist

Description

This function calculates the distance between two trees in terms of their branch lengths, assuming identical topologies.

Usage

bsd.dist(tree1, tree2, mean.brlen = 0.05)

Arguments

tree1	A phylogenetic tree of class 'phylo'.
tree2	A phylogenetic tree of class 'phylo'.
mean.brlen	Value used for scaling the trees to make them comparable, which can be arbitrary.

Details

The function uses the branch score distance (Kuhner & Felsenstein 1994) between two trees, but first rescaling them to be as similar to each other as possible.

Value

A numeric value that is the BSDmin distance.

Author(s)

David Duchene

References

Kuhner, M. K. and Felsenstein, J. (1994) A simulation comparison of phylogeny algorithms under equal and unequal evolutionary rates, Molecular Biology and Evolution, *11(3)*, 459-468

See Also

quart.scores

Examples

set.seed(12345)
myTree1 <- rtree(20)
myTree2 <- myTree1
myTree2$edge.length <- runif(Nedge(myTree2), min = 0, max = 10)
branchlengthdist <- bsd.dist(myTree1, myTree2)
branchlengthdist



## The function is currently defined as
function (tree1, tree2, mean.brlen = 0.05)
{
    if (sum(tree1$edge.length) < sum(tree2$edge.length)) {
        tree3 <- tree1
        tree1 <- tree2
        tree2 <- tree3
    }
    tree.dist.opt <- function(x) {
        tree3 <- tree2
        tree3$edge.length <- tree2$edge.length * x
        return(KF.dist(tree1, tree3))
    }
    opt.dist <- optim(0, fn = tree.dist.opt, method = "Brent", lower = 0, upper = 50)
    min.bdi <- opt.dist$value
    scaling <- opt.dist$par
    tree2.scaled <- tree2
    tree2.scaled$edge.length <- tree2$edge.length * scaling
    root.scaling <- mean.brlen/mean(c(tree1$edge.length[tree1$edge.length > 1e-05], tree2.scaled$edge.length[tree2.scaled$edge.length > 1e-05]))
    tree1.root.scaled <- tree1
    tree2.root.scaled <- tree2.scaled
    tree1.root.scaled$edge.length <- tree1$edge.length * root.scaling
    tree2.root.scaled$edge.length <- tree2.scaled$edge.length * root.scaling
    min.bdi.root.scaled <- KF.dist(tree1.root.scaled, tree2.root.scaled)
    return(min.bdi.root.scaled)
}

duchene/ClockstaRX documentation built on Oct. 22, 2023, 10:51 a.m.