Jove
Visualize
Contact Us

Related Experiment Videos

Estimating gene function with least squares nonnegative matrix factorization.

Guoli Wang1, Michael F Ochs

  • 1Fox Chase Cancer Center, Philadelphia, PA, USA.

Methods in Molecular Biology (Clifton, N.J.)
|March 5, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effects of the traditional nine steaming and nine sun-drying on the bioactivity of traditional Chinese medicine.

Journal of ethnopharmacology·2026
Same author

Thiol-Isocyanate Photoprinting of Liquid Crystal Elastomers Toward High Toughness and Reprintability.

Angewandte Chemie (International ed. in English)·2026
Same author

Effects of oil types on the stability of oil-in-water emulsions prepared using Dioscorea opposita mucilage and gums.

Food chemistry·2026
Same author

Dynamic assessment of gastric emptying patterns after endoscopic bowel preparation using ultrasonography: A prospective observational study.

Journal of clinical anesthesia·2026
Same author

Impact of gut microbiome, plasma metabolites, peripheral immune cells, and circulating inflammatory protein on chronic spontaneous urticaria: Bidirectional 2-sample Mendelian randomization study and mediation analysis.

The journal of allergy and clinical immunology. Global·2026
Same author

Comparison of the Effects of Glycopyrrolate and Atropine on Postoperative Delirium in Older Adult Patients Undergoing Laparoscopic Colorectal Surgery: A Randomized Controlled Trial.

Drug design, development and therapy·2026
Same journal

Isolation of Mesenchymal Stem Cell-Derived Extracellular Vesicles.

Methods in molecular biology (Clifton, N.J.)·2026
Same journal

Modeling Melanoma Immune Surveillance by CAR-T Cells in Human Skin Organoids.

Methods in molecular biology (Clifton, N.J.)·2026
Same journal

Stepwise Optimization of a Matrigel-Based In Vitro Angiogenesis Assay for Reproducible and Quantifiable 2D-Tube Formation Using HUVECs.

Methods in molecular biology (Clifton, N.J.)·2026
Same journal

Quantifying Mechanical Properties of Fresh Ovarian Tissue with Optical Brillouin Microscopy.

Methods in molecular biology (Clifton, N.J.)·2026
Same journal

3D Chromatin Architecture During Early Development: New Methods and New Findings.

Methods in molecular biology (Clifton, N.J.)·2026
Same journal

Metabolic Plasticity in Embryogenesis Throughout the Lens of NAD<sup></sup>.

Methods in molecular biology (Clifton, N.J.)·2026
See all related articles
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Least squares nonnegative matrix factorization (LS-NMF) improves gene function prediction from microarray data by accounting for transcription level variance. This method enhances accuracy in analyzing gene expression patterns.

Area of Science:

  • Bioinformatics
  • Machine Learning
  • Genomics

Background:

  • Nonnegative matrix factorization (NMF) is widely used for data analysis in fields like genomics.
  • A key limitation of NMF in microarray analysis is high variance in transcription levels, hindering accurate gene function prediction.
  • Existing NMF methods often minimize Euclidean distance or divergence, which may not optimally handle biological data variability.

Purpose of the Study:

  • To apply a modified NMF approach, specifically least squares NMF (LS-NMF), to microarray data.
  • To address the challenge of high variance in gene transcription levels for improved gene function estimation.
  • To demonstrate the utility of LS-NMF in predicting gene function from gene expression data.

Main Methods:

  • Implementation of least squares nonnegative matrix factorization (LS-NMF).

Related Experiment Videos

  • Incorporation of uncertainties in mRNA levels for each gene and condition into the LS-NMF algorithm.
  • Utilizing normalized chi2 as the objective function to minimize, guiding the algorithm towards a local minimum.

    • Main Results:

    • LS-NMF was applied to microarray data to predict gene function.
    • The method demonstrated an ability to handle variations in transcription levels more effectively than standard NMF.
    • The application showed promise in improving the accuracy of gene function prediction.

    Conclusions:

    • Least squares nonnegative matrix factorization offers a robust approach for analyzing gene expression data.
    • LS-NMF effectively mitigates the impact of high variance in transcription levels.
    • This enhanced NMF technique shows significant potential for advancing gene function prediction in bioinformatics.