lnl_FPK_multiclades_same_V_different_sig2: Likelihood functions for multiple clades

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

Description

These functions create likelihood functions of the FPK or BBMV model over multiple, independent, clades.

Usage

lnl_FPK_multiclades_same_V_different_sig2(trees, traits, 
  a = NULL, b = NULL, c = NULL, Npts = 50)
lnl_FPK_multiclades_same_V_same_sig2(trees, traits, 
  a = NULL, b = NULL, c = NULL, Npts = 50)
lnl_BBMV_multiclades_same_V_different_sig2(trees, traits,bounds, 
  a = NULL, b = NULL, c = NULL, Npts = 50)
lnl_BBMV_multiclades_same_V_same_sig2(trees, traits,bounds, 
  a = NULL, b = NULL, c = NULL, Npts = 50)

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.

Author(s)

F. C. Boucher

See Also

find.mle_FPK_multiple_clades_same_V_different_sig2 find.mle_FPK_multiple_clades_same_V_same_sig2

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
testbFPK4=lnl_FPK_multiclades_same_V_different_sig2(trees=TREES,
  traits=TRAITS,a=NULL,b=NULL,c=NULL,Npts=50)
fitbFPK4=find.mle_FPK_multiple_clades_same_V_different_sig2(model=testbFPK4,
  method='Nelder-Mead',init.optim=NULL)
  
## End(Not run)

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