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

Robust PCA and classification in biosciences.

Mia Hubert1, Sanne Engelen

  • 1Department of Mathematics, Katholieke Universiteit Leuven, W. De Croylaan 54, B-3001 Leuven, Belgium. Mia.Hubert@wis.kuleuven.ac.be

Bioinformatics (Oxford, England)
|February 28, 2004
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

Erythrocytosis-inducing PHD2 mutations implicate biological role for N-terminal prolyl-hydroxylation in HIF1α oxygen-dependent degradation domain.

eLife·2025
Same author

Massive Localized Lymphedema, Wound Care Without Major Surgical Excision: A Case Report.

Advances in skin & wound care·2025
Same author

Glucose deprivation impairs hypoxia-inducible factor-1α synthesis.

Discover oncology·2024
Same author

Hypoglossal Nerve Stimulation Therapy in a Belgian Cohort of Obstructive Sleep Apnea Patients.

Life (Basel, Switzerland)·2024
Same author

Deficiency in PHD2-mediated hydroxylation of HIF2α underlies Pacak-Zhuang syndrome.

Communications biology·2024
Same author

Can virtual non-contrast imaging replace true non-contrast imaging in multiphase scanning of the neck region?

Acta radiologica open·2023

Robust Principal Component Analysis (ROBPCA) offers improved outlier detection and classification for high-dimensional data. This method enhances analysis of biochemical datasets by combining robust estimation with projection-pursuit techniques.

Area of Science:

  • Bioinformatics
  • Statistical Computing
  • Data Science

Background:

  • Principal Component Analysis (PCA) is a standard dimension reduction technique for high-dimensional data.
  • Classical PCA is sensitive to outliers, compromising analysis and classification accuracy.
  • Outlying measurements significantly degrade the performance of classification methods relying on sample covariance matrices.

Purpose of the Study:

  • To introduce a robust PCA (ROBPCA) method for analyzing high-dimensional data.
  • To develop a diagnostic plot for identifying and classifying outliers.
  • To evaluate the effectiveness of ROBPCA in biochemical data analysis and classification.

Main Methods:

  • Developed a robust PCA (ROBPCA) method integrating projection-pursuit and robust estimation.

Related Experiment Videos

  • Implemented a diagnostic plot for outlier detection and classification.
  • Applied ROBPCA to biochemical datasets and combined it with robust discriminant analysis.

    • Main Results:

    • ROBPCA effectively handles high-dimensional data with outliers.
    • The diagnostic plot aids in identifying and classifying outliers.
    • The combination of ROBPCA and robust discriminant analysis improved classification accuracy compared to classical PCA and quadratic discriminant analysis.

    Conclusions:

    • ROBPCA provides a robust alternative to classical PCA for high-dimensional data analysis.
    • Robust methods, including ROBPCA, enhance the reliability of classification in the presence of outliers.
    • The proposed methods offer improved performance for biochemical data analysis.