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

Flexible empirical Bayes models for differential gene expression.

Kenneth Lo1, Raphael Gottardo

  • 1Department of Statistics, University of British Columbia, 333-6356 Agricultural Road, Vancouver, BC, Canada V6T 1Z2. c.lo@stat.ubc.ca

Bioinformatics (Oxford, England)
|December 2, 2006
PubMed
Summary

This study enhances Bayesian hierarchical models for gene expression analysis by introducing gene-specific variances, significantly reducing false positives and improving robustness in differential expression inference.

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

Resolving sensitivity, specificity and signal contamination in Xenium spatial transcriptomics.

Nature methods·2026
Same author

Multiplexed single-cell and spatial profiling reveal B cells and tertiary lymphoid structures as prognostic indicators in pleural mesothelioma.

British journal of cancer·2026
Same author

Humoral and cellular responses to a tetravalent dengue vaccine (TAK-003) in adults from a dengue non-endemic region: An open-label phase 2 trial.

Vaccine·2026
Same author

Molecular profiling of inflammatory palmoplantar disorders for diagnosis and treatment optimization.

The Journal of allergy and clinical immunology·2026
Same author

CAR-modified marrow infiltrating lymphocytes efficiently target malignant plasma cells with very low antigen density.

Molecular therapy : the journal of the American Society of Gene Therapy·2026
Same author

The ubiquitin ligase KLHL6 drives resistance to CD8<sup>+</sup> T cell dysfunction.

Nature·2026

Area of Science:

  • Bioinformatics
  • Statistical Genetics
  • Computational Biology

Background:

  • Differential gene expression analysis is crucial for understanding biological processes.
  • Bayesian hierarchical models, including Gamma-Gamma (GG) and Lognormal-Normal (LNN), are popular for gene expression analysis.
  • Existing models often rely on unrealistic assumptions, such as a common coefficient of variation across genes, potentially impacting inference accuracy.

Purpose of the Study:

  • To extend existing Gamma-Gamma (GG) and Lognormal-Normal (LNN) Bayesian hierarchical models.
  • To incorporate gene-specific variances into these models to overcome limitations of common variance assumptions.
  • To develop robust parameter estimation methods for the enhanced models.

Main Methods:

  • Extension of GG and LNN hierarchical models to accommodate gene-specific variances.

Related Experiment Videos

  • Development and application of Expectation-Maximization (EM) based algorithms for parameter estimation.
  • Evaluation of the proposed methodology using cDNA microarray and Affymetrix spike-in experimental datasets.

    • Main Results:

    • The extended GG and LNN models demonstrated a significant reduction in the false positive rate compared to original models.
    • The enhanced models maintained high sensitivity while improving accuracy in differential expression inference.
    • Simulation studies confirmed the robustness of the new frameworks against model misspecification.

    Conclusions:

    • The proposed extension of Bayesian hierarchical models with gene-specific variances offers improved accuracy and reliability for differential gene expression analysis.
    • The developed EM-based algorithms provide effective parameter estimation for these enhanced models.
    • These advancements contribute to more precise interpretation of gene expression data in various biological contexts.