fit_FPK_multiple_clades_different_V_different_sig2: Fit independent models in several clades.

In BBMV: Models for Continuous Traits Evolving in Macroevolutionary Landscapes of any Shape

Description

This function is a wrapper for functions that fit the FPK or BBMV model to a single clade that does it repetitively over several clades.

Usage

fit_FPK_multiple_clades_different_V_different_sig2(trees, traits, 
  a = NULL, b = NULL, c = NULL, Npts = 50, method = "Nelder-Mead", init.optim = NULL)
fit_BBMV_multiple_clades_different_V_different_sig2(trees, traits, bounds, 
  a = NULL, b = NULL, c = NULL, Npts = 50, method = "Nelder-Mead", init.optim = NULL)

Arguments

trees	A list of phylogenetic trees in 'phylo' format, one per clade.
traits	A list of trait vectors for species in each clade. Should be in the same order as trees.
bounds	The two bounds that constrain trait values when fitting the BBMV model. Specified by a numeric vector containing the minimum and maximum bound of the trait interval as the first and second element, respectively.
a	The value of the x^4 term in the evolutionary potential. If set to NULL (the default), this parameter will be estimated. If a numeric value is provided, this parameter will be fixed to the value specified.
b	The value of the quadratic term in the evolutionary potential. If set to NULL (the default), this parameter will be estimated. If a numeric value is provided, this parameter will be fixed to the value specified.
c	The value of the linear term in the evolutionary potential. If set to NULL (the default), this parameter will be estimated. If a numeric value is provided, this parameter will be fixed to the value specified.
Npts	The number of points used in the discretization procedure.
method	The optimization routine to be used: can be either "Nelder-Mead" (the default) or "L-BFGS-B". See the documentation of the optim function for more details. From our experience, "Nelder-Mead" seems to produce better results.
init.optim	A vector of initial values for model parameters to start the optimization algorithm. If left NULL (as is by default), the function chooses a reasonable starting point, but you might want to play around with it.

Author(s)

F. C. Boucher

Examples

## Not run: 
# We first create a potential that we will use to simulate trait evolution
# It has two peaks of very unequal heights
x=seq(from=-1.5,to=1.5,length.out=100)
bounds=c(min(x),max(x)) # the bounds we use for simulating
a=8 ; b=-4 ; c=1
V6=a*x^4+b*(x^2)+c*x
step_size=(max(bounds)-min(bounds))/(100-1)
V6_norm=exp(-V6)/sum(exp(-V6)*step_size) # the step size on the grid
par(mfrow=c(1,1))
plot(V6_norm,type='l')

# Now we simulate a tree and a continuous trait for 3 independent clades.
tree=sim.bdtree(stop='taxa',n=25) 
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=1,bounds=bounds)
tree1=tree ; TRAIT1=TRAIT

tree=sim.bdtree(stop='taxa',n=25) 
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=0.5,bounds=bounds) 
tree2=tree ; TRAIT2=TRAIT

tree=sim.bdtree(stop='taxa',n=25) 
tree$edge.length=100*tree$edge.length/max(branching.times(tree))
TRAIT= Sim_FPK(tree,x0=0.5,V=V6,sigma=0.1,bounds=bounds) 
tree3=tree ; TRAIT3=TRAIT
rm(tree) ; rm(TRAIT)

TREES=list(tree1,tree2,tree3)
TRAITS=list(TRAIT1,TRAIT2,TRAIT3)

# Fit the FPK model using ML
fitmFPK4=fit_FPK_multiple_clades_different_V_different_sig2(trees=TREES,
  traits=TRAITS,a=NULL,b=NULL,c=NULL,Npts=50)

  
## End(Not run)

BBMV documentation built on May 1, 2019, 10:26 p.m.