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

Incorporating covariates into integrated factor analysis of multi-view data.

Gen Li1, Sungkyu Jung2

  • 1Department of Biostatistics, Mailman School of Public Health, Columbia University, New York 10032, New York, U.S.A.

Biometrics
|April 14, 2017
PubMed
Summary

This study introduces supervised integrated factor analysis (SIFA), a novel method for reducing multi-view data by decomposing it into joint and individual factors. SIFA effectively integrates auxiliary covariates for enhanced insights in biomedical research.

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

Clinical Predictors of Nose/throat Bacteriome and Fungal Colonization in Skilled Nursing Facility Residents.

The Journal of infectious diseases·2026
Same author

Facial Skin Quality Improvement After Treatment With CPM-HA20G: Clinical Experience in Korea.

Journal of cosmetic dermatology·2025
Same author

BAMITA: Bayesian multiple imputation for tensor arrays.

Biostatistics (Oxford, England)·2024
Same author

ADAPT: Analysis of Microbiome Differential Abundance by Pooling Tobit Models.

Bioinformatics (Oxford, England)·2024
Same author

Challenges and real-world solutions for adoption of holistic skincare routine (cleansing, treatment, moisturization, and photoprotection) in acne, rosacea, atopic dermatitis, and sensitive skin: An expert consensus.

Journal of cosmetic dermatology·2024
Same author

Simultaneous clustering and estimation of networks in multiple graphical models.

Biostatistics (Oxford, England)·2024

Area of Science:

  • Biostatistics
  • Genomics
  • Bioinformatics

Background:

  • Biomedical research frequently involves analyzing multiple datasets from the same samples (multi-view data).
  • Integrating multi-view data is crucial for leveraging comprehensive information and simplifying complex datasets for analysis.
  • Existing methods may not fully capture joint and individual variations or incorporate auxiliary information effectively.

Purpose of the Study:

  • To develop a novel statistical model, supervised integrated factor analysis (SIFA), for integrative dimension reduction of multi-view data.
  • To incorporate auxiliary covariates into the factor decomposition process.
  • To provide a robust method for analyzing complex, multi-modal biological data.

Main Methods:

  • Developed the supervised integrated factor analysis (SIFA) model.
Keywords:
Data integrationDimension reductionMulti-source dataPrincipal component analysisSupervision

Related Experiment Videos

  • SIFA decomposes multi-view data into joint and individual factors.
  • Employed a computationally efficient Expectation-Maximization (EM) algorithm for model fitting, incorporating nonparametric models for auxiliary covariates.

    • Main Results:

    • SIFA successfully decomposes multi-view data into joint and individual factors, capturing variations across and within datasets.
    • Application to Genotype-Tissue Expression (GTEx) data yielded new insights into gene expression variation across multiple tissues.
    • Simulation studies and a pediatric growth study demonstrated SIFA's superiority over competing methods.

    Conclusions:

    • SIFA offers a powerful approach for integrative dimension reduction of multi-view data.
    • The method effectively leverages auxiliary covariates to enhance the understanding of biological variation.
    • SIFA provides valuable insights in complex biomedical datasets, as shown in gene expression and growth studies.