prepare_network_data: Prepare ASV count data for correlation network analysis

In mselensky/bngal: Biological Network Graphing and Learning

Prepare ASV count data for correlation network analysis

Description

This function prepares data for correlation network analysis at a specified level of taxonomic classification. Singleton ASVs are first removed from a provided binned data (from bin_taxonomy()) that are filtered either above or below a given abundance cutoff value based on user input. Corresponding metadata columns (such as a geochemical parameter for each sample) may be included in the preprocessed network data by specifying the corr.cols option. Microbial abundance and environmental metadata (if applicable) are normalized on a scale of 0-1 for downstream correlation matrix construction.

Usage

prepare_network_data(binned.tax, meta.data, corr.cols, sub.comms)

Arguments

binned.tax	Output from bin_taxonomy(). Must be an absolute abundance ASV table.
meta.data	See bin_taxonomy()
corr.cols	Metadata variables (such as sample environmental data) to include in network correlations. Variables must be supplied as a concatenated string of column names from meta.data (e.g., c('var1','var2')). Default = NULL
sub.comms	Optional Metadata column by which to separate networks.

mselensky/bngal documentation built on June 3, 2024, 6:27 a.m.