AncCond: Calculate the mean of a continuous character at transitions...

In coleoguy/evobir: Evolutionary Biology in R

Calculate the mean of a continuous character at transitions in a binary charachter

Description

This function uses ancestral state estimations for a discrete character based on stochastic mapping under an Mk2 model and ancestral state estimates for a continuous trait under a Brownian motion model to determine if transitions in the binary trait coincide with extreme values of a continuous trait.

Usage

AncCond(tree, 
        data, 
        mc = 1000, 
        drop.state=NULL, 
        mat=c(0,2,1,0), 
        pi="equal", 
        n.tails = 1, 
        message = TRUE)

Arguments

tree	tree of class phylo
data	a dataframe with 3 columns. The first should match the taxa names in the tree, the second should have the continuous trait values and the third the states for the binary character. The binary trait should be coded as 1 and 2 if one is ancestral then it must be coded as 1.
mc	the number of iterations to use in building the null distribution.
drop.state	NULL or a numeric value of 1 or 2. If 1 or 2 are given then continuous data from species in the specified state will be dropped from the reconstruction of the continuous character.
mat	a vector describing the possible transition for the discrete trait. The default is equivelant to APE's "ARD" model c(0,2,1,0), other options are c(0,0,1,0) which would allow only for transition from state 1 to state 2 or c(0,1,1,0) which would allow for transition in either direction but at equal rates.
pi	The probabilities for the binary trait at the root of the tree. The values possible are "equal", "estimated", or a numeric vector of length 2 with probabilities for each state
n.tails	numeric value of 1 or 2 to indicate whether a 1 or 2 tailed p-value should be calculated
message	Logical value if TRUE then status messages will be printed to the console

Details

This function uses ancestral state estimates to determine if the transitions in the binary trait are associated with extreme values of the continuous trait. This test can incorporate the possibility that the derived state of a binary character may lead to correlated selection in the continuous trait. If this is desired then the drop.state argument should be used to specify the derived state of the binary character that should not be used in the ancestral state estimation of the continuous trait.

Value

Returns a list of length 4:

OriginatingVals	the mean value(s) for the continuous trait at the transition points of the binary character
NTrans	the number of transitions in the binary character
NullDist	the null distribution(s) produced by simulation
pval	pvalue

Author(s)

Nathan Anderson, Jeffery P. Demuth, Richard H. Adams, and Heath Blackmon

References

http://coleoguy.github.io/

Examples

## Not run: 
data(mite.trait)
data(trees.mite)
AncCond(trees[[1]], mite.trait) 

## End(Not run)

coleoguy/evobir documentation built on May 8, 2024, 6:57 p.m.