Description Usage Arguments Value Note Author(s) References Examples
View source: R/bdsky.stt.optim.R
bdsky.stt.optim estimates the maximum likelihood birth and death rates together with the rate shift times t=(t[1],t[2] .., t[m]) for a given phylogenetic tree with sequentially sampled tips. At the times t, the rates are allowed to change.
1 | bdsky.stt.optim(x,ttype=0,rho=0,sampprob=c(0),constdeath=0,root=0)
|
x |
Vector of branching and sampling times in the phylogeny. Time is measured increasing going into the past with the present being time 0. x can be obtained from a phylogenetic tree using getx(TREE,sersampling=TRUE). |
ttype |
Vector of same length as x. If ttype[i]=0, then x[i] denotes a branching event. If ttype[i]=1, then x[i] denotes a sampling event. |
rho |
Probability of sampling individuals at present. rho=0 is default. |
sampprob |
Vector of length k where k is the number of different birth rates to be estimated. |
constdeath |
If constdeath=0 (default) then k death rates are estimated. If constdeath=1, then 1 death rate is estimated. |
root |
root=0 indicates that there is an edge above the root (mrca) in the tree. root=1 indicates that there is no edge above the root. |
out[[1]] |
Entry [[j]] are the maximum likelihood parameter estimates for j-1 shifts, j=1...k. The first entry of out[[1]][[j]] is the -log likelihood value, followed by the maximum likelihood parameter estimates. The parameters are stated in the following order: first the j turnover estimates (from recent to ancient), then the j diversification (speciation-extinction) rate estimates (from recent to ancient), then the j-1 shift times. |
out[[2]] |
Matrix where in each row, the first entry denotes the type of convergence problem, the second entry denotes the number of shifts in the problematic caluclation, and the third entry denotes in which interval it happened. |
bdsky.stt.optim extends the function bd.shifts.optim to trees with sequentially sampled tips.
Tanja Stadler
T. Stadler, D. Kuehnert, S. Bonhoeffer, A. Drummond. Birth-death skyline plot reveals temporal changes of epidemic spread in HIV and hepatitis C virus (HCV). Proc. Nat. Acad. Sci., 110(1): 228-233, 2013.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | set.seed(1)
# simulation of a tree with one rate shift at 0.5:
lambda<-c(3,4)
mu<-c(1,1)
sampprob<-c(0.5,0.5)
time<-c(0,0.5)
n<-10
tree<- sim.bdsky.stt(n,lambda,mu,time,sampprob)
tree2<-addroot(tree[[1]],tree[[1]]$root.edge)
summary<-getx(tree2,sersampling=TRUE)
times<-summary[,1]
ttype<-summary[,2]
# Maximum likelihood parameter estimation:
out <- bdsky.stt.optim(x=times,ttype=ttype,sampprob=sampprob,root=0)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.