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

HykGene: a hybrid approach for selecting marker genes for phenotype classification using microarray gene expression

Yuhang Wang1, Fillia S Makedon, James C Ford

  • 1Department of Computer Science, Dartmouth College, 6211 Sudikoff Laboratory, Hanover, NH 03755-3510, USA. wyh@cs.dartmouth.edu

Bioinformatics (Oxford, England)
|December 9, 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

Epizelus (Hdt 6.117): A Medical History Critique and Reappraisal.

Social history of medicine : the journal of the Society for the Social History of Medicine·2025
Same author

Stiffness evaluation of continuum robots based on the energy method and castigliano's second theorem.

Frontiers in robotics and AI·2025
Same author

Urine Metabolomics Reveals the Intervention Effects and Mechanism of Shenhua Tablets in IgA Nephropathy.

Biomedical chromatography : BMC·2025
Same author

Dynamic Cap-Mediated Substrate Access and Potent Inhibitor Design of Monkeypox Virus I7L Protease.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2025
Same author

Health risks of ambient polycyclic aromatic hydrocarbons based on the prediction of their gas-phase concentrations in the central Yangtze River Delta, China: Regional and time-resolved variability.

Journal of hazardous materials·2025
Same author

Nickel-Mediated Ring Expansion Reactions of Benzocyclobutenones for Selective Construction of 5,7-Dihydro-6<i>H</i>-dibenzo[<i>a</i>,<i>c</i>][7]annulen-6-ones and 1<i>H</i>-Isochromen-1-ones.

Organic letters·2025

This study introduces a new method to select non-redundant marker genes for disease classification using microarray data. The approach improves accuracy by identifying key genes, overcoming limitations of existing gene selection techniques.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Microarray gene expression data aids disease phenotype classification.
  • A key challenge is the high dimensionality of gene data versus limited samples.
  • Selecting informative genes enhances classification accuracy.

Purpose of the Study:

  • To develop a method for selecting a small set of non-redundant marker genes for accurate disease classification.
  • To address the issue of highly correlated genes in existing top-ranked gene sets.

Main Methods:

  • A hybrid approach combining gene ranking and hierarchical clustering was developed.
  • Feature filtering identified top-ranked genes, followed by clustering to generate a dendrogram.
  • A sweep-line algorithm analyzed the dendrogram to select marker genes by collapsing dense clusters.

Related Experiment Videos

Main Results:

  • The novel approach effectively selects a small number of marker genes.
  • Achieved comparable or superior leave-one-out cross-validation accuracy compared to methods using top-ranked genes directly.
  • Empirical validation on three public datasets confirmed the approach's efficacy.

Conclusions:

  • The hybrid gene selection method offers an effective strategy for disease classification using microarray data.
  • It successfully identifies non-redundant marker genes, improving classification performance.
  • This approach provides a valuable tool for genomic data analysis in disease research.