randtest.dpcoa: Permutation test for double principal coordinate analysis...

View source: R/randtest.dpcoa.R

randtest.dpcoaR Documentation

Permutation test for double principal coordinate analysis (DPCoA)


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


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



an object of class dpcoa


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


the number of permutations to perform, the default is 99


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


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")


an object of class randtest


Sandrine Pavoine pavoine@mnhn.fr


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



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

ade4 documentation built on Feb. 16, 2023, 7:58 p.m.

Related to randtest.dpcoa in ade4...