fitZINB {benchdamic}R Documentation

fitZINB

Description

Fit a Zero-Inflated Negative Binomial (ZINB) distribution for each taxon of the countdata. The ZINB estimation procedure is performed by zinbwave zinbFit function with commondispersion = FALSE, regularization parameter epsilon = 1e10, and not assuming the presence of any group in the samples (design matrix equal to a column of ones.)

Usage

fitZINB(counts, verbose = TRUE)

Arguments

counts

a phyloseq object or a matrix of counts with features (OTUs, ASVs, genes) by row and samples by column.

verbose

an optional logical value. If TRUE information on the steps of the algorithm is printed. Default verbose = TRUE.

Value

A data frame containing the continuity corrected logarithms of the average fitted values for each row of the matrix of counts in the Y column, and the estimated probability to observe a zero in the Y0 column.

Examples

# Generate some random counts
counts = matrix(rnbinom(n = 60, size = 3, prob = 0.5), nrow = 10, ncol = 6)

# Fit model on the counts matrix
ZINB <- fitZINB(counts)
head(ZINB)

[Package benchdamic version 1.0.0 Index]