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

Predicting protein structural class by SVM with class-wise optimized features and decision probabilities.

Ashish Anand1, Ganesan Pugalenthi, P N Suganthan

  • 1School of Electrical and Electronic Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore.

Journal of Theoretical Biology
|April 22, 2008
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

CDK12/CDK13 inhibition disrupts transcriptional elongation and replication fork progression in glioblastoma.

EMBO molecular medicine·2026
Same author

Shallow and ensemble deep randomized neural network for anomaly detection.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

Evaluation of the Effect of Parental Smoking on Gingival Melanin Pigmentation in Children.

International journal of clinical pediatric dentistry·2024
Same author

A Comparative Evaluation of Nanosilver Fluoride, Chlorhexidine, and Sodium Fluoride When Used as a Varnish on <i>Streptococcus mutans</i> Levels in Children with Caries.

International journal of clinical pediatric dentistry·2024
Same author

Observation and Assessment of the Parameters of Facial Esthetics in 6-year-old Children with Healthy Dentition.

International journal of clinical pediatric dentistry·2023
Same author

Online dynamic ensemble deep random vector functional link neural network for forecasting.

Neural networks : the official journal of the International Neural Network Society·2023
Same journal

Evolution of quantitative traits: exploring the ecological, social and genetic bases of adaptive polymorphism.

Journal of theoretical biology·2026
Same journal

The male-biased sex ratio in humans and its role in the transition from promiscuity to pair bonding.

Journal of theoretical biology·2026
Same journal

Quantifying the counter-intuitive effects of vaccination by coupling the transmission dynamics of COVID-19 and the evolution of human behaviors.

Journal of theoretical biology·2026
Same journal

An integrative model of FGF2-induced signaling and muscle cell proliferation.

Journal of theoretical biology·2026
Same journal

A hybrid reaction-diffusion and mechanical stimulus model for mandibular bone remodeling under chewing and vibratory loading.

Journal of theoretical biology·2026
Same journal

Integrated tick management strategies in fragmented peridomestic environments.

Journal of theoretical biology·2026
See all related articles

Predicting protein structural class from sequence data is now more accurate using a novel support vector machine (SVM) approach. This method employs optimized features and probability estimates for enhanced protein classification.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Structural Bioinformatics

Background:

  • Determining protein structural class from sequence alone is a significant challenge in bioinformatics.
  • Previous methods have explored various computational approaches with limited success.

Purpose of the Study:

  • To develop and evaluate a novel Support Vector Machine (SVM) based approach for predicting protein structural class from sequence information.
  • To improve prediction accuracy by utilizing class-wise optimized feature sets and probability-based decision making.

Main Methods:

  • Implementation of a Support Vector Machine (SVM) algorithm incorporating probability-based decision making.
  • Optimization of feature sets specific to each protein structural class.
  • Testing the algorithm on three diverse datasets (498, 1092, and 5261 protein domains) using ten-fold external cross-validation.

Related Experiment Videos

Main Results:

  • Achieved a high accuracy of 92.89% on the 498-domain dataset.
  • Surpassed previous benchmarks with 54.67% accuracy on the 1092-domain dataset (previously 53.8%).
  • Obtained 59.43% prediction accuracy on the large 5261-domain dataset, demonstrating the efficacy of class-wise features.

Conclusions:

  • The proposed SVM approach with class-wise optimized features and probability estimates significantly enhances protein structural class prediction accuracy.
  • Class-wise feature optimization offers a clear advantage over conventional methods using a union of features.
  • The selected features exhibit biological relevance, contributing to the model's predictive power.