ecos_nullmodel: Null models for ecostructure_fit clustering validation

In kkdey/ecostructure: Visualizing Structure in ecological site-species abundance data

Description

Generate randomized matrix of counts given the observed data matrix for a particular null model, run ecos_fit on these null matrices and compare the fit on null model data with that on the observed data. Used for validating the clustering.

Usage

ecos_nullmodel(
  counts,
  K,
  tol = 0.1,
  null.model = c("frequency", "richness", "independentswap", "trialswap"),
  iter_fill = 100,
  iter_randomized = 30,
  option,
  plot = TRUE
)

Arguments

counts	The counts matrix (N x G): N- the number of samples (sites), G- number of features (bird species)
K	The number of clusters to fit
tol	The tolerance limit of the ecos_fit.
null.model	The type of nullmodel used (similar to the picante::randomizeMatrix() function argument in picante package)
iter_fill	The number of swaps/fills in each randomized matrix build
iter_randomized	The number of randomization matrices generated
plot	If TRUE, plots density function of log Bayes factor over the randomized iterations

Value

Returns a list with

BF.obs	log Bayes Factor for the observed counts with K=2 against the null with no clusters
BF.rand	a vector of log BF for each randomized count matrix with K=2 against the null with no clusters
pval	the p-value of the observed log Bayes factor against the ones from randomized matrices

Examples

data("himalayan_birds")
species_abundance_counts <- t(exprs(himalayan_birds));
out <- ecos_nullmodel(species_abundance_counts, K=2, 
                iter_randomized=5, option = "BF")
out2 <- ecos_nullmodel(species_abundance_counts, K=2, 
                iter_randomized=5, ption = "BIC")              
  

kkdey/ecostructure documentation built on Jan. 26, 2021, 4:10 p.m.