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

Diffusion kernel-based logistic regression models for protein function prediction.

Hyunju Lee1, Zhidong Tu, Minghua Deng

  • 1Department of Computer Science, University of Southern California, Los Angeles, 90089, USA.

Omics : a Journal of Integrative Biology
|April 6, 2006
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

Infective Endocarditis in a Previously Healthy Adolescent Caused by <i>Streptococcus sanguinis</i>: Diagnostic Challenges and Clinical Implications.

Infection & chemotherapy·2026
Same author

Efficacy and safety of Doxycycline versus Macrolides for <i>Mycoplasma pneumoniae</i> INfectiOn in Children (DOMINO): a protocol for a multicentre, randomised, open-label, superiority trial.

BMJ open·2026
Same author

Quantitative Analysis of the Impact of Region of Interest Information on Deep Learning Algorithms for Thyroid Ultrasound Imaging.

IEEE open journal of engineering in medicine and biology·2026
Same author

High-Grade Astrocytoma With Piloid Features: An Aggressive Clinicogenomic Entity Distinct From Pilocytic Astrocytoma.

Journal of Korean medical science·2026
Same author

Effectiveness of toothbrushing as a replacement for chlorhexidine in oral care, combined with head elevation for the prevention of ventilator-associated pneumonia in intensive care units: interrupted time-series analysis.

Critical care (London, England)·2026
Same author

Metagenomic Hi‑C Protocols for Viral Genome Binning, Taxonomic Annotation, and Interaction Network Visualization.

Current protocols·2026

Predicting protein functions is crucial. A new kernel logistic regression (KLR) model improves accuracy by considering all protein neighbors and multiple functions, outperforming previous methods.

Area of Science:

  • Proteomics
  • Bioinformatics
  • Computational Biology

Background:

  • Assigning functions to unknown proteins is a key challenge in proteomics.
  • Previous methods, like Markov Random Field (MRF), primarily used direct protein-protein interactions and individual functions.
  • Limitations included considering only direct interactions and analyzing functions separately.

Purpose of the Study:

  • To develop an improved computational model for protein function prediction.
  • To leverage information from all neighboring proteins in interaction networks.
  • To extend function prediction from single to multiple functions.

Main Methods:

  • Developed a novel kernel logistic regression (KLR) method using diffusion kernels for protein interaction networks.

Related Experiment Videos

  • Incorporated correlated functions identified via chi-square test.
  • Extended the model by integrating multiple biological data sources (protein domains, complexes, gene expression) as networks.

    • Main Results:

    • The KLR model significantly improved protein function prediction accuracy compared to the MRF model.
    • Incorporating multiple data sets further enhanced prediction accuracy.
    • The KLR model's performance was comparable to a support vector machine (SVM) diffusion kernel approach.

    Conclusions:

    • The KLR approach effectively utilizes information from all protein neighbors for more accurate function prediction.
    • Integrating diverse biological data sources enhances prediction capabilities.
    • The KLR model offers a simple yet powerful method for understanding protein functions within interaction networks.