dot-boottest: Bootstrap Procedure for Testing Statistical Significance of...
In stefpeschel/NetCoMi: Network Construction and Comparison for Microbiome Data

.boottest

R Documentation

Bootstrap Procedure for Testing Statistical Significance of Correlation Values

Description

Statistical significance of correlations between pairs of taxonomic units is tested using a bootstrap procedure as proposed by Friedman and Alm (2012).

Usage

.boottest(
  countMat,
  assoMat,
  nboot = 1000,
  measure,
  measurePar,
  cores = 4,
  logFile = NULL,
  verbose = TRUE,
  seed = NULL,
  assoBoot = NULL
)

Arguments

`countMat`	matrix containing microbiome data (read counts) for which the correlations are calculated (rows represent samples, columns represent taxa)
`assoMat`	matrix containing associations that have been estimated for `countMat`.
`nboot`	number of bootstrap samples.
`measure`	character specifying the method used for computing the associations between taxa.
`measurePar`	list with parameters passed to the function for computing associations/dissimilarities. See details for the respective functions.
`cores`	number of CPU cores used for parallelization.
`logFile`	character defining a log file, where the number of iteration is stored. If NULL, no log file is created. wherein the current iteration numbers are stored.
`verbose`	logical; if `TRUE`, the iteration numbers are printed to the R console
`seed`	an optional seed for reproducibility of the results.
`assoBoot`	list with bootstrap association matrices.