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

A robust hybrid between genetic algorithm and support vector machine for extracting an optimal feature gene subset.

Li Li1, Wei Jiang, Xia Li

  • 1Department of Bioinformatics, Harbin Medical University, Harbin 150086, People's Republic of China.

Genomics
|December 21, 2004
PubMed
Summary

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

Guiding ultrasound image capture with artificial intelligence.

Lancet (London, England)·2020
Same author

Welcoming new guidelines for AI clinical research.

Nature medicine·2020
Same author

Prevalence of Asymptomatic SARS-CoV-2 Infection : A Narrative Review.

Annals of internal medicine·2020
Same author

Computer vision's potential to improve health care.

Lancet (London, England)·2020
Same author

Medicine and meteorology: Cloud, connectivity, and care.

Lancet (London, England)·2020
Same author

Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies.

BMJ (Clinical research ed.)·2020

This study introduces a novel hybrid approach combining genetic algorithms and support vector machines for robust gene selection in microarray data. This method effectively identifies key genes for disease prediction, achieving 99% accuracy in diffuse large B-cell lymphoma classification.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Microarray data presents challenges like high dimensionality and noise, requiring efficient gene selection methods.
  • Existing dimensional reduction techniques may fail when a few highly influential genes dominate feature subsets.
  • Identifying key molecular signatures is crucial for understanding complex biological phenotypes.

Purpose of the Study:

  • To develop a robust and efficient gene selection approach for high-dimensional microarray data.
  • To address limitations in current methods, particularly when dealing with influential genes.
  • To identify key feature genes for predicting complex biological phenotypes.

Main Methods:

  • Formalized a robust gene selection approach using a hybrid of genetic algorithm (GA) and support vector machine (SVM).

Related Experiment Videos

  • Leveraged GA's robustness to solution space size and SVM's capability in handling high-dimensional features.
  • Applied the hybrid approach to diffuse large B-cell lymphoma (DLBCL) microarray data.

    • Main Results:

    • The hybrid GA-SVM approach successfully identified optimal gene subsets for classification.
    • Achieved a high prediction accuracy of 99% for independent microarray samples.
    • Outperformed marginal filters and a GA-K nearest neighbors hybrid in predictive accuracy.

    Conclusions:

    • The hybrid GA-SVM method offers a powerful tool for mining high-dimensional gene expression data.
    • This approach effectively identifies molecular signatures for accurate disease prediction.
    • Demonstrated superior performance in classifying DLBCL samples compared to existing methods.