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

Eigengene-based linear discriminant model for tumor classification using gene expression microarray data.

Ronglai Shen1, Debashis Ghosh, Arul Chinnaiyan

  • 1Department of Biostatistics, University of Michigan Ann Arbor, MI 48109-0602, USA.

Bioinformatics (Oxford, England)
|August 24, 2006
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

Frailty phenotype transitions and functional improvements during a supervised exercise trial in older people with HIV: results from the HEALTH Trial.

Age and ageing·2026
Same author

Glycemic response trajectories on metformin monotherapy in real-world diabetes care.

medRxiv : the preprint server for health sciences·2026
Same author

Robust ranking of renewable energy alternatives handling uncertainty using novel hesitant bi-fuzzy MEREC-MOORA and Dombi aggregation approach.

Scientific reports·2026
Same author

The Impact of Social Vulnerability on Exercise Outcomes: A Longitudinal Study of Physical Function in Older People With HIV.

Journal of the International Association of Providers of AIDS Care·2026
Same author

Special issue: cell and gene causal inference in the design and analysis of gene therapy clinical trials.

Journal of biopharmaceutical statistics·2026
Same author

Mapping the last mile: Micro-stratification for sustained visceral leishmaniasis elimination in Bangladesh.

PLoS neglected tropical diseases·2026

Eigengene-based Linear Discriminant Analysis (ELDA) improves tumor classification by selecting key genes in a multivariate framework. This method creates smaller, more accurate classifiers compared to traditional univariate approaches.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Nearest shrunken centroids classifier is popular for tumor classification using gene expression data.
  • Current feature selection methods use univariate statistics, ignoring gene co-regulation networks and causing redundancy.

Purpose of the Study:

  • To introduce Eigengene-based Linear Discriminant Analysis (ELDA) for multivariate gene selection.
  • To improve the accuracy and reduce the size of tumor classifiers.

Main Methods:

  • ELDA employs a modified Spectral Decomposition (SpD) to identify 'hub' genes linked to important eigenvectors.
  • The approach incorporates a misclassification cost matrix for differential error penalization.

Main Results:

Related Experiment Videos

  • ELDA selects more characteristic genes, resulting in significantly smaller classifiers than univariate methods.
  • De-correlated gene expression profiles enhance the applicability of diagonal linear discriminant models.
  • False negative prognoses in breast cancer were controlled using a cost-adjusted discriminant function.

Conclusions:

  • ELDA offers a powerful multivariate approach for gene selection in cancer microarray analysis.
  • The method enhances classifier performance and provides better control over misclassification errors.