dataprep.mcl: dataprep.mcl

In aplantin/MCL: Fits multilevel compositional lasso

Description

Prepares input matrices for MCL function based on otu count matrix. Rows of the otu table should be subjects.

Usage

dataprep.mcl(otu.table, groups, rare.count = 1, rare.prop = 0,
  pseudocount = 0.5)

Arguments

otu.table	Matrix of otu counts; rows are subjects, columns are OTUs.
groups	Vector indicating group membership
rare.count	OTU count considered "rare"; OTUs with total read count <= rare.count will be excluded. Default is 1 (exclude singletons). If <= 0, no filter is applied for rare OTUs based on counts.
rare.prop	Alternative to rare.count; minimum proportion of samples in which the OTU should appear. Default is 0. If <= 0, no filter is applied for rare OTUs based on proportions.
pseudocount	Count to replace zeros; default is 0.5.

Value

Returns a list with elements Z, W, and groups for input into MCL function, and a vector with the names of groups with only one element (excluded from W matrix).

aplantin/MCL documentation built on May 25, 2019, 2:26 p.m.