mean_efficiency: Compute sample mean efficiencies
In mikemc/metacal: Bias estimation and calibration for metagenomics experiments

Description Usage Arguments Details See Also

Given a matrix of 'observed' or 'actual' (e.g. calibrated) abundance profiles and a vector of relative efficiencies, compute the estimated mean efficiency for each sample.

mean_efficiency(x, ...)

## S3 method for class 'matrix'
mean_efficiency(x, bias, margin, type)

## S3 method for class 'otu_table'
mean_efficiency(x, bias, type)

## S3 method for class 'phyloseq'
mean_efficiency(x, bias, type)

## S3 method for class 'mc_bias_fit'
mean_efficiency(x, newdata = NULL, margin = NULL, type = NULL)

`x`	A matrix, a phyloseq object with an otu_table, or an object of class 'mc_bias_fit'
`bias`	A (possibly named) numeric vector of relative efficiencies
`margin`	The margin containing the samples (observations)
`type`	'actual' or 'observed', indicating the type of abundance profiles in the matrix `x` or `newdata`
`newdata`	NULL or an abundance matrix

The mean efficiency of a sample equals weighted.mean(bias, y), where bias is the vector of taxa efficiencies and y is the vector of taxa proportions.

The type %in% c('actual', 'observed') specifies whether the data is 'actual' (i.e. nominally known or calibrated abundances) or 'observed' (i.e. uncalibrated) data. 'Observed' data is calibrated to determine y prior to computing the mean efficiency.

If x is an mc_bias_fit object, then the efficiencies will be extracted with coef(x). An abundance matrix can optionally be given with newdata; otherwise, the mean efficiency will be computed for the control samples in x.