nmf: NMF Iteration

In mskcc/tempoSig: Maximum Likelihood Fitting of Mutational Catalogs and De Novo Inference of Signatures

NMF Iteration

Description

Non-negative matrix factorization with either fixed or variable signature matrix.

Usage

nmf(
  catalog,
  sig = NULL,
  denovo = TRUE,
  K = NULL,
  nrun = 10,
  verbose = TRUE,
  progress.bar = FALSE,
  Itmax = 1e+05,
  Tol = 1e-04,
  a = 10,
  nprint = 100,
  useC = FALSE
)

Arguments

catalog	Catalog matrix
denovo	Full iteration of both signature and exposure; if FALSE (hnmf), only the exposure is fit (requires signat input).
K	Number of signatures.
nrun	Number of independent runs to generate
verbose	Verbosity level
progress.bar	Display progress bar.
Itmax	Maximum no. of iteration.
Tol	Tolerance for checking convergence.
signat	Signature matrix

Details

If denovo = TRUE, the full NMF is performed with both W and H matrix determined de novo. If denovo = FALSE, W is fixed as input and only H is determined.

Value

Object of class tempoSig.

mskcc/tempoSig documentation built on Feb. 3, 2023, 8:35 a.m.