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

PLS dimension reduction for classification with microarray data.

Anne-Laure Boulesteix1

  • 1Department of Statistics, University of Munich. anne-laure.boulesteix@tum.de

Statistical Applications in Genetics and Molecular Biology
|May 2, 2006
PubMed
Summary

Partial Least Squares (PLS) offers high prediction accuracy for high-dimensional microarray data classification. This study compares PLS with other methods, introduces a boosting algorithm, and explores PLS for gene selection and data visualization.

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

When to Adjust for Multiple Testing: A Unifying Guiding Principle.

Biometrical journal. Biometrische Zeitschrift·2026
Same author

STrategies for developing REseArch Methods guidance (STREAM): Protocol.

Journal of clinical epidemiology·2026
Same author

The statistical software revolution in pharmaceutical development: challenges and opportunities in open source.

Drug discovery today·2026
Same author

On "Confirmatory" Methodological Research in Statistics and Related Fields.

Statistics in medicine·2025
Same author

ChatGPT as a Tool for Biostatisticians: A Tutorial on Applications, Opportunities, and Limitations.

Statistics in medicine·2025
Same author

Rethinking the Handling of Method Failure in Comparison Studies.

Statistics in medicine·2025

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Statistical Genomics

Background:

  • High-dimensional microarray data presents challenges for accurate classification.
  • Partial Least Squares (PLS) is recognized for its effectiveness in prediction accuracy with such data.

Purpose of the Study:

  • To compare PLS-based classification with state-of-the-art methods.
  • To investigate the application of a boosting algorithm to PLS classification.
  • To explore PLS for gene selection, data visualization, and determining the optimal number of components.

Main Methods:

  • Partial Least Squares (PLS) dimension reduction.
  • Linear Discriminant Analysis (LDA) on PLS components.
  • Application of a boosting algorithm.

Related Experiment Videos

  • Analysis of 9 real microarray cancer datasets.

    • Main Results:

    • PLS-based classification demonstrated competitive prediction accuracy.
    • A method for selecting the number of PLS components was proposed.
    • The relationship between PLS dimension reduction and gene selection was examined.
    • A property of the first PLS component for binary classification was proven.
    • PLS was shown to be effective for data visualization.

    Conclusions:

    • PLS, combined with LDA and boosting, is a powerful tool for high-dimensional microarray data classification.
    • PLS offers insights into gene selection and data visualization in cancer genomics.
    • The proposed methods enhance the utility of PLS in bioinformatics research.