Phylogenetic community structure hypothesis test

This function computes the maximum likelihood estimates of colonisation and local extinction rate for a given phylogeny and presence-absence data under the DAMOCLES model. These rate estimates are used to simulate null communities under the DAMOCLES model. Standardized effect size of mean nearest taxon distance (mntd), mean phylogentic distance (mpd) and loglikelihood are calculated For comparison, standardised effect sizes are also calculated relative to a "Random-Draw" null model i.e. presence absence randomised across tips

Usage

DAMOCLES_bootstrap(
  phy = ape::rcoal(10),
  pa = matrix(c(phy$tip.label, sample(c(0, 1), ape::Ntip(phy), replace = T)), nrow =
    ape::Ntip(phy), ncol = 2),
  initparsopt = c(0.1, 0.1),
  idparsopt = 1:length(initparsopt),
  parsfix = NULL,
  idparsfix = NULL,
  pars2 = c(0.001, 1e-04, 1e-05, 1000),
  pchoice = 0,
  runs = 999,
  estimate_pars = FALSE,
  conf.int = 0.95
)

Arguments

phy: phylogeny in phylo format
pa: presence-absence table.
The first column contains the labels of the species (corresponding to the tip labels in the phylogeny.
The second column contains the presence (1) or absence (0) of species in the local community.
initparsopt: The initial values of the parameters that must be optimized
idparsopt: The ids of the parameters that must be optimized, e.g. 1:2 for extinction rate, and offset of immigration rate The ids are defined as follows:
id == 1 corresponds to mu (extinction rate)
id == 2 corresponds to gamma_0 (offset of immigration rate)
parsfix: The values of the parameters that should not be optimized. See idparsfix.
idparsfix: The ids of the parameters that should not be optimized, e.g. c(1) if mu should not be optimized, but only gamma_0. In that case idparsopt must be c(2). The default is to fix the parameters not specified in idparsopt.
pars2: Vector of settings:
pars2[1] sets the relative tolerance in the parameters

pars2[2] sets the relative tolerance in the function

pars2[3] sets the absolute tolerance in the parameters

pars2[4] sets the maximum number of iterations
pchoice: sets which p-value to optimize and with which root state to simulate (default pchoice = 0)
pchoice == 0 correspond to optimizing sum of p_0f + p_1f, and simulating with an equal number of root states being 0 or 1
pchoice == 1 correspond to optimizing p_0f, and simulating with root state being 0
pchoice == 2 correspond to optimizing p_1f, and simulating with root state being 1
runs: the number null communities to generate.
estimate_pars: Whether to estimate parameters on the simulated datasets (default = FALSE).
conf.int: The width of the conifdence intervals calculated on bootstrapped parameter estimates

Value

summary_table: mu gives the maximum likelihood estimate of mu and confidence intervals in brackets if estimate_pars = TRUE gamma_0 gives the maximum likelihood estimate of gamma_0 and confidence intervals in brackets if bootstrap=TRUE loglik gives the maximum loglikelihood df gives the number of estimated parameters, i.e. degrees of feedom conv gives a message on convergence of optimization; conv = 0 means convergence n.obs gives the number of species locally present in the observed community mntd.obs gives the MNTD of the observed community mpd.obs gives the MPD of the observed community runs gives the number of null communities simulated mntd.mean.RD mean of MNTD from null communities generated by a "Random Draw" model mntd.sd.RD standard deviation of MNTD from null communities generated by a "Random Draw" model mntd.obs.z.RD standardized effect size of MNTD compared to null communities generated by a "Random Draw" model (= -1*(mntd.obs - mntd.mean.RD)/ mntd.sd.RD) mntd.obs.rank.RD rank of observed MNTD compared to null communities generated by a "Random Draw" model mntd.obs.q.RD quantile of observed MNTD vs. null communities (= mntd.obs.rank.RD /runs + 1) mpd.mean.RD mean of MPD from null communities generated by a "Random Draw" model mpd.sd.RD standard deviation of MPD from null communities generated by a "Random Draw" model mpd.obs.z.RD standardized effect size of MPD compared to null communities generated by a "Random Draw" model (= -1*(mpd.obs - mpd.mean.RD)/ mpd.sd.RD) mpd.obs.rank.RD rank of observed MPD compared to null communities generated by a "Random Draw" model mpd.obs.q.RD quantile of observed MPD vs. null communities (= mpd.obs.rank.RD /runs + 1) n.mean.DAMOCLES mean number of species locally present in the null communities generated by DAMOCLES mntd.mean.DAMOCLES mean of MNTD from null communities generated by DAMOCLES mntd.sd.DAMOCLES standard deviation of MNTD from null communities generated by DAMOCLES mntd.obs.z.DAMOCLES standardized effect size of MNTD compared to null communities generated by DAMOCLES (= -1*(mntd.obs - mntd.mean.DAMOCLES)/ mntd.sd.DAMOCLES) mntd.obs.rank.DAMOCLES rank of observed MNTD compared to null communities generated by DAMOCLES mntd.obs.q.DAMOCLES quantile of observed MNTD vs. null communities (= mntd.obs.rank.DAMOCLES /runs + 1) mpd.mean.DAMOCLES mean of MPD from null communities generated by DAMOCLES mpd.sd.DAMOCLES standard deviation of MPD from null communities generated by DAMOCLES mpd.obs.z.DAMOCLES standardized effect size of MPD compared to null communities generated by DAMOCLES (= -1*(mpd.obs - mpd.mean.DAMOCLES)/ mpd.sd.DAMOCLES) mpd.obs.rank.DAMOCLES rank of observed MPD compared to null communities generated by DAMOCLES mpd.obs.q.DAMOCLES quantile of observed MPD vs. null communities (= mpd.obs.rank.DAMOCLES /runs + 1) loglik.mean.DAMOCLES mean of loglikelihoods from null communities generated by DAMOCLES loglik.sd.DAMOCLES standard deviation of loglikelihoods from null communities generated by DAMOCLES loglik.obs.z.DAMOCLES standardized effect size of loglikelihood compared to null communities generated by DAMOCLES (= -1*(loglik.obs - loglik.mean.DAMOCLES)/ loglik.sd.DAMOCLES) loglik.obs.rank.DAMOCLES rank of observed loglikelihood compared to null communities generated by DAMOCLES loglik.obs.q.DAMOCLES quantile of observed loglikelihoods vs. null communities (= loglik.obs.rank.DAMOCLES /runs + 1)
null_community_data: run gives the simulation run root.state.print gives the state of the ancestral species in the local community assumed in the simulation, i.e. present (1) or absent (0) n gives the number of species locally present in the observed community n.RD gives the number of species locally present in the null community generated by a "Random Draw" model mntd.RD gives the MNTD of the null community generated by a "Random Draw" model mpd.RD gives the MPD of the null community generated by a "Random Draw" model n.DAMOCLES gives the number of species locally present in the null community generated by DAMOCLES mntd.DAMOCLES gives the MNTD of the null community generated by DAMOCLES mpd.DAMOCLES gives the MPD of the null community generated by DAMOCLES loglik.DAMOCLES gives the maximum loglikelihood for the null community generated by DAMOCLES mu.DAMOCLES gives the maximum likelihood estimate of mu for the null community generated by DAMOCLES gamma_0.DAMOCLES gives the maximum likelihood estimate of gamma_0 for the null community generated by DAMOCLES

Details

The output is a list of two dataframes. The first dataframe, summary_table, contains the summary results. The second dataframe, null_community_data, contains decsriptive statistics for each null community.

References

Pigot, A.L. & R.S. Etienne (2015). A new dynamic null model for phylogenetic community structure. Ecology Letters 18: 153-163.

Author

Rampal S. Etienne