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

Megavariate data analysis of mass spectrometric proteomics data using latent variable projection method.

Kwan R Lee1, Xiwu Lin, Daniel C Park

  • 1GlaxoSmithKline Pharmaceuticals, Collegeville, PA 19426, USA. kwan.lee@gsk.com

Proteomics
|September 16, 2003
PubMed
Summary

Latent variable projection methods, including principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), effectively analyze complex spectroscopic and chromatographic data. A PLS-DA model achieved 85% accuracy in classifying proteomic data, demonstrating its utility.

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

Real-World Treatment Patterns and Outcomes for Patients With Relapsed/Refractory Multiple Myeloma and Prior Exposure to Lenalidomide and Proteasome Inhibitors in the United States.

Clinical lymphoma, myeloma & leukemia·2026
Same author

Relative Effectiveness of Talquetamab Versus Real-World Physician's Choice Treatment Regimens in Patients With Relapsed/Refractory Multiple Myeloma: Updated Results From an Indirect Treatment Comparison.

Clinical lymphoma, myeloma & leukemia·2026
Same author

Identification of treatment effect modifiers and prognostic factors in newly diagnosed and relapsed or refractory multiple myeloma.

Journal of comparative effectiveness research·2025
Same author

Real-world frontline treatments in patients with advanced non-small-cell lung cancer harboring epidermal growth factor receptor exon 20 insertions and adjusted comparisons versus amivantamab plus chemotherapy from the PAPILLON study.

Lung cancer (Amsterdam, Netherlands)·2025
Same author

Real-world characteristics and outcomes of patients with high-risk and non-high-risk smoldering multiple myeloma using the Flatiron Health database.

Blood cancer journal·2024
Same author

Talquetamab Versus Real-World Physician's Choice Treatment: Comparative Effectiveness in Patients With Triple-Class Exposed Relapsed/Refractory Multiple Myeloma.

Clinical lymphoma, myeloma & leukemia·2024

Area of Science:

  • Chemometrics
  • Data Mining
  • Bioinformatics

Background:

  • Spectroscopic and chromatographic data are inherently multivariate and complex.
  • Traditional data mining techniques may struggle with the high dimensionality of such datasets.
  • Latent variable projection methods offer a powerful approach for analyzing complex chemical and biological data.

Purpose of the Study:

  • To evaluate the effectiveness of wavelet transformation and latent variable projection methods for multivariate data analysis.
  • To apply principal component analysis (PCA) for clustering and partial least squares discriminant analysis (PLS-DA) for classification of proteomic data.
  • To validate the performance of a PLS-DA model using internal and external cross-validation.

Main Methods:

  • Utilized latent variable projection methods, specifically PCA for clustering and PLS-DA for classification.

Related Experiment Videos

  • Applied these methods to raw data from the First Duke Proteomics Data Mining Conference.
  • Employed internal and external cross-validation techniques for model assessment.

    • Main Results:

    • PCA was used to address the clustering problem.
    • A simple two-component PLS-DA model successfully separated two distinct data groups.
    • External validation with an independent test set achieved an accuracy of 85%.

    Conclusions:

    • Latent variable projection methods, particularly PLS-DA, are highly effective for analyzing complex spectroscopic and chromatographic data.
    • The developed PLS-DA model demonstrates robust classification performance for proteomic datasets.
    • Cross-validation confirms the reliability and generalizability of the classification model.