run_sampler_phy: Aggregated and overdispersed sampling for phylo objects
In BrunoVilela/sampler: Aggregated and overdispersed sampling

Description Usage Arguments Details Value Author(s) See Also Examples

Generate aggregated or overdispersed sampling designs for any given phylogenetic tree (class phylo). Results can be used to design experiments/samples, for resample proposes and data bias removal.

1 2	run_sampler_phy(x, n, alpha, dist.func = cophenetic, n_start = 1, return_start = FALSE, starting = NULL)

`x`	`phylo` object.
`n`	A positive integer number indicating the sample size.
`alpha`	Number indicating the strength of aggregation (if negative) or overdispersion (if positive). When alpha = 0 sample is random.
`dist.func`	Function to calculate the phylogenetic distance. The function should be able to receive a phylo object and return a `matrix` indicating the pairwise distance between tips. Row and column names should be given. Default uses `cophenetic` from `ape` package. Other examples include
`n_start`	Number of initial selected tips. Default is one starting tip.
`return_start`	if `TRUE` the starting tip is returned.
`starting`	Character vector indicating the starting tips. If not provided, random starting value(s) is(are) selected.

The function uses the algorithm in run_sampler, but here it accepts a phylo object as input.

The function returns a pruned phylogenetic tree.

Bruno Vilela

run_sampler

Nee_May_1997

# Generate a random tree
require(ape)
set.seed(100)
tree <- rcoal(10)
set.seed(2)
# Highly overdispersed 50% resample design (alpha = 100)
overdispersed <- run_sampler_phy(tree, 5, alpha = 100, starting = "t10")
# Highly aggregated 50% resample design (alpha = -100)
aggregated <- run_sampler_phy(tree, 5, alpha = -100, starting = "t10")
# Random 50% resample design (alpha = 0)
random <- run_sampler_phy(tree, 5, alpha = 0, starting = "t10")
# Plot to compare
par(mfrow = c(2, 2))
plot(tree, main = "Full tree", cex = 1)
axis(1)
plot(overdispersed, main = "Overdispersed 50% sampling", cex = 1)
axis(1)
plot(aggregated, main = "Aggregated 50% sampling", cex = 1)
axis(1)
plot(random, main = "Random 50% sampling", cex = 1)
axis(1)