Jove
Visualize
Contact Us
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

Related Experiment Videos

Subsystem identification through dimensionality reduction of large-scale gene expression data.

Philip M Kim1, Bruce Tidor

  • 1Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.

Genome Research
|July 4, 2003
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

Making invisible excited-state structures of pro-interleukin-18 visible by combining NMR and machine learning.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Language models for protein design.

Current opinion in structural biology·2025
Same author

FlowPacker: protein side-chain packing with torsional flow matching.

Bioinformatics (Oxford, England)·2025
Same author

A High-Throughput Method for Screening Peptide Activators of G-Protein-Coupled Receptors.

ACS omega·2024
Same author

EuDockScore: Euclidean graph neural networks for scoring protein-protein interfaces.

Bioinformatics (Oxford, England)·2024
Same author

Antibody-SGM, a Score-Based Generative Model for Antibody Heavy-Chain Design.

Journal of chemical information and modeling·2024

Non-negative matrix factorization (NMF) enhances gene expression analysis by identifying localized patterns, improving functional predictions in biological data. This machine learning approach offers up to twice the accuracy of traditional methods for understanding cellular subsystems.

Area of Science:

  • Systems Biology
  • Computational Biology
  • Genomics

Background:

  • High-throughput biological experiments generate vast datasets of cellular observables.
  • Analyzing complex gene expression data requires advanced algorithms for multi-level resolution.
  • Current methods often rely on large-scale expression pattern similarity.

Purpose of the Study:

  • To apply non-negative matrix factorization (NMF) for analyzing gene array experiments.
  • To develop algorithms capable of recognizing similarity within subportions of data.
  • To improve the accuracy of predicting functional cellular relationships.

Main Methods:

  • Utilized non-negative matrix factorization (NMF), a machine learning technique.
  • Applied NMF to a dataset of 300 genome-wide expression measurements in yeast.

Related Experiment Videos

  • Compared NMF's predictions with those from standard data analysis approaches.

    • Main Results:

    • NMF successfully identified localized features in gene expression data.
    • Detected local features mapped effectively to known functional cellular subsystems.
    • NMF-based predictions showed up to twofold improvement in accuracy compared to conventional methods.

    Conclusions:

    • Non-negative matrix factorization is a powerful tool for analyzing gene expression data.
    • NMF's ability to detect local features enhances the understanding of cellular subsystems.
    • This approach offers a more accurate method for predicting functional relationships in biological systems.