fitSnpNmf: Non-negative matrix factorization (NMF) of a matrix...
In HenrikBengtsson/ACNE: Affymetrix SNP Probe-Summarization using Non-Negative Matrix Factorization

fitSnpNmf

R Documentation

Non-negative matrix factorization (NMF) of a matrix containing SNP probe signals

Non-negative matrix factorization (NMF) of a matrix containing SNP probe signals.

fitSnpNmf(V, acc=0.02, maxIter=10, maxIterRlm=20, refs=NULL)

`V`	An KxI `matrix` where I is the number of arrays and K is the number of probe where K should be even (K=2L).
`acc`	A positive `double` specifying the converence threshold. For more details on convergence, see below.
`maxIter`	A positive `integer` specifying the maximum number of iterations used to calculate the decomposition.
`maxIterRlm`	A positive `integer` specifying the maximum number of iterations used in rlm.
`refs`	An index `vector` (`integer` or `logical`) specifying the reference samples. If `NULL`, all samples are used as a reference.

The algorithm is considered to have converged when the maximum update of any allele-specific copy number of any array (H) is greater than acc.

Returns a list:

`W`	The Kx2 `matrix` containing allele-specific affinity estimates.
`H`	A 2xI `matrix` containing allele-specific copy number estimates.
`hasConverged`	`TRUE` if the algorithm converged, otherwise `FALSE`. If not applicable, it is `NA`.
`nbrOfIterations`	The number of iteration ran before stopping. If not applicable, it is `NA`.