Permutation test for double principal coordinate analysis (DPCoA)

Share:

Description

randtest.dpcoa calculates the ratio of beta to gamma diversity associated with DPCoA and compares the observed value to values obtained by permuting data.

Usage

1
2
3
## S3 method for class 'dpcoa'
randtest(xtest, model = c("1p","1s"), nrep = 99,
alter = c("greater", "less", "two-sided"), ...)

Arguments

xtest

an object of class dpcoa

model

either "1p", "1s", or the name of a function, (see details)

nrep

the number of permutations to perform, the default is 99

alter

a character string specifying the alternative hypothesis, must be one of "greater" (default), "less" or "two-sided"

...

further arguments passed to or from other methods

Details

Model 1p permutes the names of the columns of the abundance matrix. Model 1s permutes the abundances of the categories (columns of the abundance matrix, usually species) within collections (rows of the abundance matrix, usually communities). Only the categories with positive abundances are permuted. The null models were introduced in Hardy (2008).

Other null model can be used by entering the name of a function. For example, loading the picante package of R, if model=randomizeMatrix, then the permutations will follow function randomizeMatrix available in picante. Any function can be used provided it returns an abundance matrix of similar size as the observed abundance matrix. Parameters of the chosen function can be added to randtest.dpcoa. For example, using parameter null.model of randomizeMatrix, the following command can be used: randtest.dpcoa(xtest, model = randomizeMatrix, null.model = "trialswap")

Value

an object of class randtest

Author(s)

Sandrine Pavoine pavoine@mnhn.fr

References

Hardy, O. (2008) Testing the spatial phylogenetic structure of local communities: statistical performances of different null models and test statistics on a locally neutral community. Journal of Ecology, 96, 914–926

See Also

dpcoa

Examples

1
2
3
data(humDNAm)
dpcoahum <- dpcoa(data.frame(t(humDNAm$samples)), sqrt(humDNAm$distances), scan = FALSE, nf = 2)
randtest(dpcoahum)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.