Keeps the community and what describes its species and sites together, with names checked
once. The analyses accept an mc_data object as their first argument and take from it what
the call leaves out, so pcps_test(d) uses d$tree and d$env. Simulations convert with
as_mc_data().
Usage
mc_data(
comm,
tree = NULL,
traits = NULL,
env = NULL,
coords = NULL,
areas = NULL,
series = NULL
)
as_mc_data(x, ...)
# Default S3 method
as_mc_data(x, ...)
# S3 method for class 'mc_data'
as_mc_data(x, ...)
# S3 method for class 'mc_sim'
as_mc_data(x, ...)
# S3 method for class 'mc_data'
print(x, ...)
# S3 method for class 'mc_data'
x[i, j, drop = FALSE]
# S3 method for class 'mc_data'
summary(object, ...)
# S3 method for class 'mc_data'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)Arguments
- comm
Matrix or data frame with one row per site and one column per species, named by species. Sites without names are called
site1,site2and so on.- tree
Optional
phyloobject.- traits
Optional data frame, matrix or named vector with one row per species, named by species.
- env
Optional data frame, matrix or vector with one row per site.
- coords
Optional matrix or data frame of coordinates with one row per site.
- areas
Optional vector with the area or biogeographic event of each site, for
biogeography_matrix().- series
Optional array of sites x species x steps, such as the
historyof a simulation.- x, object
An
mc_dataobject, or foras_mc_data()an object to convert.- ...
Passed to
mc_data()byas_mc_data(); not used elsewhere.- i, j
Sites and species to keep, by position, name or logical vector.
- drop, row.names, optional
Not used.
Value
mc_data() and as_mc_data() give a list of class mc_data with comm, tree,
traits (data frame), env (data frame), coords (matrix), areas and series, each
NULL when absent. [ gives the same for a subset, summary() gives summary_stats()
and as.data.frame() a long table with site, species, step (NA without series)
and abundance, without absences.
Details
Species are matched by name between comm, tree and traits, and those missing from any
of them are dropped with a warning, as in the analyses. Species without individuals stay,
since they belong to the regional pool. Sites are matched by name when env, coords,
areas or series have names, and by position otherwise.
Examples
set.seed(1)
tree <- treeshapesim::sim_tree_shape(12)
traits <- species_traits(tree, lambda = 1, optima_range = c(0, 100), rmax_range = c(0, 30))
env <- seq(0, 100, length.out = 20)
comm <- simulate_static(traits, env, incidence = TRUE)$abundance
d <- mc_data(comm, tree, traits["optimum"], env = data.frame(env = env))
d
#> <mc_data> 20 sites x 12 species
#> tree: 12 tips
#> traits: optimum
#> env: env
d[1:10, ]
#> <mc_data> 10 sites x 12 species
#> tree: 12 tips
#> traits: optimum
#> env: env
pcps_test(d[rowSums(d$comm) > 0, ], n_perm = 9)
#> <mc_pcps> principal coordinates of phylogenetic structure
#> permutations: 9 free of sites, 9 of species; smallest possible p: 0.1
#> test share r2 statistic p_site p_species p_max
#> whole 1.000 0.235 5.53 0.1 0.5 0.5
#> pcps1 0.277 0.735 49.95 0.1 0.2 0.2
#> pcps2 0.237 0.093 1.84 0.1 0.6 0.6
