irMF: inferential, robust Matrix Factorization

Supports robust and regular singular value decomposition

Supports three algorithms for non-negative matrix factorization

Provides visualization, heatmaps, of the factorization

Provides fitted and residual matrices

Effects a clustering of the rows and columns of the matrix

Gives a novel method of determining which columns can be used to predict class labels that is more powerful than traditional methods

Fogel, P., Young, S.S., Hawkins, D.M., Ledirac N. Inferential, robust non-negative matrix factorization analysis of microarray data. Bioinformatics. (PDF file). Supplemental material mentioned in this article (ZIP file)

Liu, L., Hawkins, D.M., Ghosh, S., Young, S.S. (2003). Robust singular value decomposition analysis of microarray data. PNAS 100, 13167-13172.
(Free download from http://www.pnas.org/cgi/content/full/100/23/13167)

Young, S.S., Fogel, P., Hawkins, D. M. Clustering Scotch Whiskies using Non-Negative Matrix Factorization (PDF file)
(Scotch Whiskies dataset)

BASIC VERSION. Download Now! The install includes a JMP script, install instructions, a ReadMe file, and a test data set.

Version 3.5 released! 4/2/2008

There are no matrix size limitations in the basic version from version 3.2.0 on.

Non-negative Matrix Factorization Workshop