Nonnegative Matrix Factorization: an Analytical and Interpretive Tool in Computational Biology

Overview

Journal PLoS Comput Biol

Specialty Biology

Date 2008 Jul 26

PMID 18654623

Citations 131

Authors

Karthik Devarajan

Affiliations

Soon will be listed here.

Abstract

In the last decade, advances in high-throughput technologies such as DNA microarrays have made it possible to simultaneously measure the expression levels of tens of thousands of genes and proteins. This has resulted in large amounts of biological data requiring analysis and interpretation. Nonnegative matrix factorization (NMF) was introduced as an unsupervised, parts-based learning paradigm involving the decomposition of a nonnegative matrix V into two nonnegative matrices, W and H, via a multiplicative updates algorithm. In the context of a pxn gene expression matrix V consisting of observations on p genes from n samples, each column of W defines a metagene, and each column of H represents the metagene expression pattern of the corresponding sample. NMF has been primarily applied in an unsupervised setting in image and natural language processing. More recently, it has been successfully utilized in a variety of applications in computational biology. Examples include molecular pattern discovery, class comparison and prediction, cross-platform and cross-species analysis, functional characterization of genes and biomedical informatics. In this paper, we review this method as a data analytical and interpretive tool in computational biology with an emphasis on these applications.

Citing Articles

Population-level analyses identify host and environmental variables influencing the vaginal microbiome.

Qin L, Sun T, Li X, Zhao S, Liu Z, Zhang C Signal Transduct Target Ther. 2025; 10(1):64.

PMID: 39966341 PMC: 11836416. DOI: 10.1038/s41392-025-02152-8.

Simplicity within biological complexity.

Przulj N, Malod-Dognin N Bioinform Adv. 2025; 5(1):vbae164.

PMID: 39927291 PMC: 11805345. DOI: 10.1093/bioadv/vbae164.

CSI-GEP: A GPU-based unsupervised machine learning approach for recovering gene expression programs in atlas-scale single-cell RNA-seq data.

Liu X, Chapple R, Bennett D, Wright W, Sanjali A, Culp E Cell Genom. 2025; 5(1):100739.

PMID: 39788105 PMC: 11770216. DOI: 10.1016/j.xgen.2024.100739.

Decomposition of the pangenome matrix reveals a structure in gene distribution in the species.

Chauhan S, Ardalani O, Hyun J, Monk J, Phaneuf P, Palsson B mSphere. 2025; 10(1):e0053224.

PMID: 39745367 PMC: 11774025. DOI: 10.1128/msphere.00532-24.

Nonnegative matrix factorization for analyzing state dependent neuronal network dynamics in calcium recordings.

Carbonero D, Noueihed J, Kramer M, White J Sci Rep. 2024; 14(1):27899.

PMID: 39537711 PMC: 11560946. DOI: 10.1038/s41598-024-78448-6.

References

Costanzo M, Crawford M, Hirschman J, Kranz J, Olsen P, ROBERTSON L . YPD, PombePD and WormPD: model organism volumes of the BioKnowledge library, an integrated resource for protein information. Nucleic Acids Res. 2000; 29(1):75-9. PMC: 29810. DOI: 10.1093/nar/29.1.75. View

Kelm B, Menze B, Zechmann C, Baudendistel K, Hamprecht F . Automated estimation of tumor probability in prostate magnetic resonance spectroscopic imaging: pattern recognition vs quantification. Magn Reson Med. 2006; 57(1):150-9. DOI: 10.1002/mrm.21112. View

Pascual-Montano A, Carmona-Saez P, Chagoyen M, Tirado F, Carazo J, Pascual-Marqui R . bioNMF: a versatile tool for non-negative matrix factorization in biology. BMC Bioinformatics. 2006; 7:366. PMC: 1550731. DOI: 10.1186/1471-2105-7-366. View

Kim P, Tidor B . Subsystem identification through dimensionality reduction of large-scale gene expression data. Genome Res. 2003; 13(7):1706-18. PMC: 403744. DOI: 10.1101/gr.903503. View

Gao Y, Church G . Improving molecular cancer class discovery through sparse non-negative matrix factorization. Bioinformatics. 2005; 21(21):3970-5. DOI: 10.1093/bioinformatics/bti653. View

Buchsbaum G, Bloch O . Color categories revealed by non-negative matrix factorization of Munsell color spectra. Vision Res. 2002; 42(5):559-63. DOI: 10.1016/s0042-6989(01)00303-0. View

Ashburner M, Ball C, Blake J, Botstein D, Butler H, Cherry J . Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000; 25(1):25-9. PMC: 3037419. DOI: 10.1038/75556. View

Heger A, Holm L . Sensitive pattern discovery with 'fuzzy' alignments of distantly related proteins. Bioinformatics. 2003; 19 Suppl 1:i130-7. DOI: 10.1093/bioinformatics/btg1017. View

Fogel P, Young S, Hawkins D, Ledirac N . Inferential, robust non-negative matrix factorization analysis of microarray data. Bioinformatics. 2006; 23(1):44-9. DOI: 10.1093/bioinformatics/btl550. View

10.

Wang G, Kossenkov A, Ochs M . LS-NMF: a modified non-negative matrix factorization algorithm utilizing uncertainty estimates. BMC Bioinformatics. 2006; 7:175. PMC: 1450309. DOI: 10.1186/1471-2105-7-175. View

11.

Chagoyen M, Carmona-Saez P, Shatkay H, Carazo J, Pascual-Montano A . Discovering semantic features in the literature: a foundation for building functional associations. BMC Bioinformatics. 2006; 7:41. PMC: 1386711. DOI: 10.1186/1471-2105-7-41. View

12.

Tresch M, Cheung V, dAvella A . Matrix factorization algorithms for the identification of muscle synergies: evaluation on simulated and experimental data sets. J Neurophysiol. 2006; 95(4):2199-212. DOI: 10.1152/jn.00222.2005. View

13.

Jung I, Lee J, Lee S, Kim D . Application of nonnegative matrix factorization to improve profile-profile alignment features for fold recognition and remote homolog detection. BMC Bioinformatics. 2008; 9:298. PMC: 2459191. DOI: 10.1186/1471-2105-9-298. View

14.

Isakoff M, Sansam C, Tamayo P, Subramanian A, Evans J, Fillmore C . Inactivation of the Snf5 tumor suppressor stimulates cell cycle progression and cooperates with p53 loss in oncogenic transformation. Proc Natl Acad Sci U S A. 2005; 102(49):17745-50. PMC: 1308926. DOI: 10.1073/pnas.0509014102. View

15.

Pascual-Montano A, Carazo J, Kochi K, Lehmann D, Pascual-Marqui R . Nonsmooth nonnegative matrix factorization (nsNMF). IEEE Trans Pattern Anal Mach Intell. 2006; 28(3):403-15. DOI: 10.1109/TPAMI.2006.60. View

16.

Plumbley M . Algorithms for nonnegative independent component analysis. IEEE Trans Neural Netw. 2008; 14(3):534-43. DOI: 10.1109/TNN.2003.810616. View

17.

Nandigam R, Kim S, Singh J, Chuaqui C . Position specific interaction dependent scoring technique for virtual screening based on weighted protein--ligand interaction fingerprint profiles. J Chem Inf Model. 2009; 49(5):1185-92. DOI: 10.1021/ci800466n. View

18.

Carrasco D, Tonon G, Huang Y, Zhang Y, Sinha R, Feng B . High-resolution genomic profiles define distinct clinico-pathogenetic subgroups of multiple myeloma patients. Cancer Cell. 2006; 9(4):313-25. DOI: 10.1016/j.ccr.2006.03.019. View

19.

Oja E, Plumbley M . Blind separation of positive sources by globally convergent gradient search. Neural Comput. 2004; 16(9):1811-25. DOI: 10.1162/0899766041336413. View

20.

Brunet J, Tamayo P, Golub T, Mesirov J . Metagenes and molecular pattern discovery using matrix factorization. Proc Natl Acad Sci U S A. 2004; 101(12):4164-9. PMC: 384712. DOI: 10.1073/pnas.0308531101. View