NMF (version 0.5.06)

NMFfitX-class: Virtual Class to Handle Results from Multiple Runs of a NMF Algorithms

Description

This class defines a common interface to handle the results from multiple runs of a single NMF algorithm, performed with the nmf method. Currently, this interface is implemented by two classes, NMFfitX1 and NMFfitXn, which respectively handle the case where only the best fit is kept, and the case where the list of all the fits is returned.

See nmf-multiple for more details on the method arguments.

References

Metagenes and molecular pattern discovery using matrix factorization Brunet, J.~P., Tamayo, P., Golub, T.~R., and Mesirov, J.~P. (2004) Proc Natl Acad Sci U S A 101(12), 4164--4169.

Sparse non-negative matrix factorizations via alternating non-negativity-constrained least squares for microarray data analysis Kim, H. & Park, H. (2007) Bioinformatics. http://dx.doi.org/10.1093/bioinformatics/btm134.

See Also

nmf-methods, nmf-multiple, NMFfitX1, NMFfitXn

Examples

# generate a synthetic dataset with known classes
n <- 50; counts <- c(5, 5, 8);
V <- syntheticNMF(n, counts, noise=TRUE)

# perform multiple runs of one algorithm (default is to keep only best fit)
res <- nmf(V, 3, nrun=5)
str(res)

# perform multiple runs of one algorithm (keep all the fits)
res <- nmf(V, 3, nrun=5, .options='k')
str(res)