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

Multi-class support vector machines for protein secondary structure prediction.

Minh N Nguyen1, Jagath C Rajapakse

  • 1School of Computer Engineering, Nanyang Technological University, Singapore. minhnguyen@pmail.ntu.edu.sg

Genome Informatics. International Conference on Genome Informatics
|February 12, 2005
PubMed
Summary

Multi-class Support Vector Machines (SVM) are more effective for protein secondary structure (PSS) prediction. A two-stage SVM approach further enhances accuracy by capturing contextual information, achieving up to 79.5% accuracy.

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

D3I-COMCF: Dissimilarity-driven dual-interaction via co-occurrence motifs and complementary fragments for drug-drug interaction prediction.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Real-time quantification of nanoplastics deposition in nanofiltration using laser-induced breakdown detection (LIBD).

Water research·2026
Same author

Soil Microbial Diversity and Network Organization Respond to Land Use and Agricultural Inputs Worldwide.

Global change biology·2026
Same author

Integrating multimodal features with deep learning for protein solubility prediction.

Journal of cheminformatics·2026
Same author

LABMA: Latent-bottleneck attention-based multimodal architecture for integration of transcriptomics, proteomics, and MRI in neurodegeneration.

Computers in biology and medicine·2026
Same author

The role of hydration in the removal of glyphosate (GLY) and aminomethylphosphonic acid (AMPA) by nanofiltration membranes.

Nature communications·2026

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning

Background:

  • Support Vector Machines (SVM) are well-established for binary classification.
  • Multi-class classification is often handled by combining binary classifiers, but recent methods consider all classes simultaneously.
  • Existing multi-class methods may involve larger optimization problems and are limited to small datasets.

Purpose of the Study:

  • To evaluate various Support Vector Machine (SVM) approaches for protein secondary structure (PSS) prediction.
  • To compare single-stage multi-class SVMs against traditional binary SVM strategies (one-against-all, one-against-one, directed acyclic graph).
  • To investigate the efficacy of a two-stage SVM approach for improving PSS prediction accuracy.

Main Methods:

Related Experiment Videos

  • Implementation of three binary classification-based SVM methods: one-against-all (OAA), one-against-one (OAO), and directed acyclic graph (DAG).
  • Implementation of two single-optimization problem multi-class SVM approaches.
  • Development and testing of a two-stage multi-class SVM model incorporating contextual information.

    • Main Results:

    • Multi-class SVM methods demonstrated superior suitability for protein secondary structure (PSS) prediction compared to binary SVMs.
    • The single-step optimization of multi-class SVMs proved advantageous.
    • A two-stage SVM approach significantly outperformed single-stage methods, achieving a maximum prediction accuracy of 79.5% on two datasets.

    Conclusions:

    • Multi-class SVMs are more effective for protein secondary structure prediction than traditional binary SVM methods.
    • A two-stage SVM strategy, by incorporating contextual information, offers a viable method for enhancing PSS prediction accuracy.
    • The proposed two-stage SVM model achieves high accuracy, demonstrating its potential for advanced bioinformatics applications.