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

Fast protein classification with multiple networks.

Koji Tsuda1, HyunJung Shin, Bernhard Schölkopf

  • 1Max Planck Institute for Biological Cybernetics, Tübingen, Germany.

Bioinformatics (Oxford, England)
|October 6, 2005
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

Dual-Layer Grain-Boundary In Situ Polymerization Modulates Elastic Modulus for Mechanically Stable Flexible All-Perovskite Tandem Solar Cells.

ACS applied materials & interfaces·2026
Same author

A critical perspective on finite sample conformal prediction theory in medical applications.

Artificial intelligence in medicine·2026
Same author

Synergistic bone regeneration through sequential dual-drug delivery.

Biomedical engineering letters·2026
Same author

The role of the tryptophan-rich allosteric network and sodium egress in GPCR activation.

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

Molecular Design with Artificial Intelligence: Progress and Perspectives for Small Molecules.

Chemical reviews·2026
Same author

Imagining and building wise machines: the centrality of AI metacognition.

Trends in cognitive sciences·2026
Same journal

3DICE: Interpretable 3D Cross-Modal Learning for Drug-Target Interaction Prediction and Large-Scale Drug Discovery.

Bioinformatics (Oxford, England)·2026
Same journal

KASSPer: Kinase Active Site Structure Prediction using Protein and Ligand Language Models and Its Application to Virtual Screening.

Bioinformatics (Oxford, England)·2026
Same journal

IDR searcher: a search engine solution for public image resources.

Bioinformatics (Oxford, England)·2026
Same journal

KCFtools: Rapid alignment-free method for introgression screening and GWAS using k-mer profiles.

Bioinformatics (Oxford, England)·2026
Same journal

Meta2DB: Curated shotgun metagenomic feature sets and metadata for health state prediction.

Bioinformatics (Oxford, England)·2026
Same journal

conMItion: an R package adjusting confounding factors for associations in multi-omics.

Bioinformatics (Oxford, England)·2026
See all related articles

We developed an efficient protein classification method using multiple protein networks. This approach significantly reduces computation time and maintains accuracy compared to existing methods, aiding in functional prediction.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Systems Biology

Background:

  • Support Vector Machines (SVMs) are effective for protein functional classification.
  • Semidefinite programming (SDP) based SVMs integrate multiple data sources but face computational challenges with large datasets and protein networks.
  • Existing methods like the diffusion kernel for protein networks have high time complexity (O(n^3)) and produce dense matrices.

Purpose of the Study:

  • To propose an efficient method for protein classification that directly incorporates multiple protein networks.
  • To overcome the computational limitations of existing methods, particularly for large-scale network data.
  • To enable the identification and potential discarding of noisy or irrelevant data sources.

Main Methods:

Related Experiment Videos

  • Directly incorporating multiple protein networks (e.g., physical interaction, metabolic) into a classification model.
  • Converting vectorial data into a network format using neighbor point connection for integration.
  • Utilizing convex optimization to determine combination weights for different data sources, similar to SDP/SVM.
  • Leveraging the sparsity of network edges for computational efficiency.

    • Main Results:

    • Achieved nearly linear computation time with respect to the number of network edges.
    • Demonstrated that combination weights can help identify and discard noisy or irrelevant networks.
    • Obtained comparable accuracy to the SDP/SVM method in yeast protein function prediction experiments.
    • Significantly reduced computation time for classifying 3588 yeast proteins.

    Conclusions:

    • The proposed method offers a computationally efficient alternative for protein classification using multiple networks.
    • The approach maintains high accuracy while drastically improving speed, making it suitable for large biological datasets.
    • The method provides insights into data source relevance, enhancing the reliability of functional predictions.