addNMF: Non-negative Matrix Factorization

In FelixErnst/mia: Microbiome analysis

Non-negative Matrix Factorization

Description

These functions perform Non-negative Matrix Factorization on data stored in a TreeSummarizedExperiment object.

Usage

getNMF(x, ...)

addNMF(x, ...)

## S4 method for signature 'SummarizedExperiment'
getNMF(x, k = 2, assay.type = "counts", eval.metric = "evar", ...)

## S4 method for signature 'SummarizedExperiment'
addNMF(
  x,
  k = 2,
  assay.type = "counts",
  eval.metric = "evar",
  name = "NMF",
  ...
)

Arguments

x	a TreeSummarizedExperiment object.
...	optional arguments passed to nmf::NMF.
k	numeric vector. A number of latent vectors/topics. (Default: 2)
assay.type	Character scalar. Specifies which assay to use for NMF ordination. (Default: "counts")
eval.metric	Character scalar. Specifies the evaluation metric that will be used to select the model with the best fit. Must be one of the following options: "evar" (explained variance; maximized), "sparseness.basis" (degree of sparsity in the basis matrix; maximized), "sparseness.coef" (degree of sparsity in the coefficient matrix; maximized), "rss" (residual sum of squares; minimized), "silhouette.coef" (quality of clustering based on the coefficient matrix; maximized), "silhouette.basis" (quality of clustering based on the basis matrix; maximized), "cophenetic" (correlation between cophenetic distances and original distances; maximized), "dispersion" (spread of data points within clusters; minimized). (Default: "evar")
name	Character scalar. The name to be used to store the result in the reducedDims of the output. (Default: "NMF")

Details

The functions getNMF and addNMF internally use nmf::NMF compute the ordination matrix and feature loadings.

If k is a vector of integers, NMF output is calculated for all the rank values contained in k, and the best fit is selected based on eval.metric value.

Value

For getNMF, the ordination matrix with feature loadings matrix as attribute "loadings".

For addNMF, a TreeSummarizedExperiment object is returned containing the ordination matrix in reducedDims(x, name) with the following attributes:

• "loadings" which is a matrix containing the feature loadings

• "NMF_output" which is the output of function nmf::NMF

• "best_fit" which is the result of the best fit if k is a vector of integers

Examples

data(GlobalPatterns)
tse <- GlobalPatterns

# Reduce the number of features
tse <- agglomerateByPrevalence(tse, rank = "Phylum")

# Run NMF and add the result to reducedDim(tse, "NMF").
tse <- addNMF(tse, k = 2, name = "NMF")

# Extract feature loadings
loadings_NMF <- attr(reducedDim(tse, "NMF"), "loadings")
head(loadings_NMF)

# Estimate models with number of topics from 2 to 4. Perform 2 runs.
tse <- addNMF(tse, k = c(2, 3, 4), name = "NMF_4", nrun = 2)

# Extract feature loadings
loadings_NMF_4 <- attr(reducedDim(tse, "NMF_4"), "loadings")
head(loadings_NMF_4)

FelixErnst/mia documentation built on Nov. 18, 2024, 5:02 a.m.