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

Multiclass cancer classification and biomarker discovery using GA-based algorithms.

Jane Jijun Liu1, Gene Cutler, Wuxiong Li

  • 1Tularik Inc., South San Francisco, CA 94080, USA.

Bioinformatics (Oxford, England)
|April 9, 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

Atmospheric versus soil Cadmium exposure in rice (Oryza sativa L.): Divergent toxicological mechanisms and human health risks via integrated multi-omics.

Journal of hazardous materials·2026
Same author

One-Pot Depolymerization, Demethylation, and Phenolation of Lignin for Bioactive Polyphenol Production.

ChemSusChem·2026
Same author

Protocol for an AlCl<sub>3</sub>-induced Alzheimer's model in zebrafish larvae with optimized pH and behavioral assessment.

STAR protocols·2026
Same author

Targeted Metabolomics Combined with OPLS-DA to Analyze the Differences in Phenolic Compounds in Wampee.

Foods (Basel, Switzerland)·2026
Same author

An injectable tough hydrogel sealant enabling rapid hemostasis and promoting oral tissue regeneration.

Bioactive materials·2026
Same author

Hypothermic Treatment Alleviates Myocardial Hypertrophy in <i>bmp10</i><sup>-/-</sup> Zebrafish.

Frontiers in bioscience (Landmark edition)·2026

This study introduces a novel method combining genetic algorithms (GA) and support vector machines (SVM) for accurate multiclass cancer categorization. The approach identifies reliable gene biomarkers for improved molecular cancer diagnosis and treatment.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Microarray technology offers potential for cancer diagnosis, prognosis, and treatment prediction.
  • High-throughput gene profiling generates vast data, necessitating effective feature reduction for tumor biomarker identification.
  • Reliable serum biomarkers are crucial for advancing cancer understanding and treatment strategies.

Purpose of the Study:

  • To develop an efficient method for multiclass cancer categorization using gene expression data.
  • To identify robust sets of tumor biomarkers for molecular cancer diagnosis.
  • To improve the accuracy of predicting prognoses and effective cancer treatments.

Main Methods:

  • Combined genetic algorithm (GA) and all-paired (AP) support vector machine (SVM) for iterative feature selection.

Related Experiment Videos

  • Utilized leave-one-out cross-validation (LOOCV) to assess classification accuracy.
  • Employed nearest shrunken centroids (NSC), annotation analysis, and literature text mining for feature characterization.

    • Main Results:

    • Achieved highly accurate multiclass cancer categorization with compact, non-redundant gene biomarker sets.
    • Identified distinct gene classifier sets with comparable accuracy, indicating robustness.
    • Discovered previously unrecognized tumor subclasses and potential cancer biomarkers through advanced analysis.

    Conclusions:

    • The GA/SVM approach provides an effective tool for cancer biomarker discovery using microarray data.
    • This method facilitates molecular cancer diagnosis and aids in understanding tumor heterogeneity.
    • The identified gene biomarkers hold promise for improving cancer patient management and treatment outcomes.