R: Simulating General Model Trees on a fixed number of extant...
sim.taxa
R Documentation
Simulating General Model Trees on a fixed number of extant Taxa
Description
sim.taxa simulates trees on n species under a general model (i.e. any given distribution for speciation and extinction). The method allows to stop the simulation right after reaching a given number of tips/taxa or to simulate a bigger tree (m-tips) and then uniformly sample one of the n-tips trees that existed in the past until m tips. For this the gsa code from the R package TreeSim is used.
The method allows for symmetric and asymmetric binary split speciation mode. To keep all extinct tips use complete=TRUE, if complere=FALSE the tree is pruned, resulting on a so called reconstructed tree.
is the Number of tips in sampled trees (Number of extant sampled leaves).
m
is the number of standing taxa that will exist on the first generated trees, to then be sampled for n number of tips. Case gsa=FALSE, m is equal to n.
distributionspname
is the name of the desired probability function that will be used for the speciation process (e.g. distributionspname="rexp"). Note that the name should contain an 'r' before it, since it refers to the randon number of the desired function (e.g. "rweibull", "runif").
distributionspparameters
are the parameters for the specific function desired for speciation.
# IMPORTANT: this vector of function parameters must *start by the second one, since the first parameter will always be one for all the function and is added already by this function*. HINT: see the help of the desired function for more details (e.g. ?rexp) Example of parameter for a exponential distribution with lambda of one is: (distributionspparameters = c(1)). Entry in the distributionparameters can be "#", # or c(#,#) in case of more characters.
distributionextname
is the same as the distributionspname but for the probability of extinction (e.g. distributionextname = "rexp").
distributionextparameters
is the same as the distributionspparameters but for the extinction probability function. By default extinction is set to ZERO, i.e. no extinction (i.e. distributionextparameters = c(0)). Entries can be "#", # or c(#,#) in case of more characters.
symmetric
tells which macroevolutionary model should be used. If symmetric=TRUE the symmetric model will be used; if FALSE, the asymmetric model will be used. By default symmetric=TRUE.
complete
if complete = TRUE, the tree with the extinct and non-sampled lineages is returned. If complete = FALSE, the extinct and non-sampled lineages are suppressed. complete=FALSE by default.
labellivingsp
is the label that will be drawn on each tip surviving until the present. An automatic sequential number will be added to the chosen name. By default labellivingsp="sp."
labelextinctsp
is the label that will be drawn on each extinct tip. By default labelextinctsp="ext."
sampling
stochastic sampling of tips, see TreeSim for more details
gsa
gsa=TRUE indicates that the sim.gsa.taxa will be used and the n parameter will dictate the final number of tips on the tree. Note that m needs to be always bigger then n.
If gsa = FALSE, there is no need of specifying n, once the final trees will be of size m.
Value
treearray
Array of numbsim trees with the time since origin / most recent common ancestor being age. If tree goes extinct (only possible when mrca = FALSE), returns 0. If only one extant and no extinct tips, returns 1.
Author(s)
Oskar Hagen, Tanja Stadler
References
O. Hagen, K. Hartmann, M. Steel and T. Stadler (2014). Age-dependent speciation explains empirical tree shape distribution. Manuscript.
T. Stadler: Simulating trees on a fixed number of extant species. Syst. Biol. (2011) 60: 676-684.
See Also
sim.age, sim.gsa.taxa
Examples
## example 1 ##
# without gsa, extinction and using weibull probability for speciation
my3s <- sim.taxa(1, n=10, distributionspname="rweibull", distributionspparameters=c(1.5,1),
symmetric=TRUE, complete=TRUE, labellivingsp="tip")
plot(my3s[[1]])
## example 2 ##
# using gsa=TRUE with extinction and exponential distribution for speciation and extinction
# note that this function may take a long timce since gsa=TRUE
mytree <- sim.taxa(numbsim=10, n=10, m=15, distributionspname="rexp",
distributionspparameters=c(1.5), distributionextname="rexp",
distributionextparameters="0.1",symmetric = TRUE, complete=TRUE,
labellivingsp="sp.", labelextinctsp="ext.", sampling=2, gsa=TRUE)
providing 
.