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

Recovering Hidden Diagonal Structures via Non-Negative Matrix Factorization with Multiple Constraints.

Xi Yang, Guoqiang Han, Hongmin Cai

    IEEE/ACM Transactions on Computational Biology and Bioinformatics
    |April 4, 2017
    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

    MGCL-ST: multi-view graph contrastive learning for spatial transcriptomics imputation.

    Genome medicine·2026
    Same author

    Decoding Spatial Heterogeneity and Multi-Omics Regulation with Hierarchical Graph Learning.

    Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
    Same author

    PIEZO1 is Required for Acute Myeloid Leukemia Progression and Leukemia Stem Cell Maintenance via HIF1A-SLC7A11 Axis-Mediated Ferroptosis Defense.

    Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
    Same author

    FKDNuSeg: Flawless knowledge distillation for lightweight and fast nuclei instance segmentation and classification.

    Medical image analysis·2026
    Same author

    Cross-Modal Denoising and Integration of Spatial Multi-Omics Data with CANDIES.

    Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
    Same author

    Overall survival across registry-defined biopsy and registry-defined STR categories in glioblastoma: a population-based matched and weighted SEER study.

    Scientific reports·2026

    This study introduces a new constrained non-negative matrix factorization (NMF) model to effectively reveal diagonal block structures in correlated data. The method accurately identifies hidden patterns in both simulated and biological datasets.

    Area of Science:

    • Data analysis
    • Machine learning
    • Bioinformatics

    Background:

    • Analyzing highly correlated variables often requires revealing underlying block structures.
    • Non-negative matrix factorization (NMF) is a technique used for data decomposition, but standard methods do not directly exploit diagonal block structures.
    • Existing NMF variants have limitations in estimating data sampled from multiple independent subspaces.

    Purpose of the Study:

    • To propose a novel non-negative matrix factorization with multiple constraints (NMF-MC) model.
    • To enable the direct estimation and recovery of intrinsic diagonal block structures in data.
    • To address the limitations of standard NMF in analyzing data with multiple independent subspaces.

    Main Methods:

    • Developed a non-negative matrix factorization with multiple constraints (NMF-MC) model.

    Related Experiment Videos

  • Incorporated sparsity norm on the feature matrix and total variational norm on the loading matrix columns.
  • Employed an alternating direction method of multipliers (ADMM) for efficient numerical optimization.

    • Main Results:

    • The proposed NMF-MC model successfully recovers hidden diagonal block structures in observed data.
    • The method demonstrates robust and effective performance on both simulated and real biological datasets.
    • Achieved superior results compared to several benchmark NMF models in identifying block structures.

    Conclusions:

    • The NMF-MC model offers an effective approach for analyzing data with diagonal block structures.
    • The proposed method advances the application of NMF in fields requiring the analysis of correlated variables, such as bioinformatics.
    • The efficient ADMM-based algorithm facilitates the practical application of this technique.