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

A structural mixed model for variances in differential gene expression studies.

Florence Jaffrézic1, Guillemette Marot, Séverine Degrelle

  • 1INRA, UR337 Station de Génétique Quantitative et Appliquée, Jouy-en-Josas, France. florence.jaffrezic@jouy.inra.fr

Genetical Research
|May 23, 2007
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

Multi-omics network inference with a Gaussian copula model.

BMC bioinformatics·2026
Same author

Algorithm for Temporomandibular Disorders With Osteopathic Manipulative Therapy: An Expert Consensus.

Journal of chiropractic medicine·2026
Same author

Biological Context and Endpoint Alignment Are Critical for Interpreting Population-Based Validation of Cluster-Based MASLD Subtypes.

Diabetes, obesity & metabolism·2026
Same author

Validation of a data-driven clustering model for MASLD: Evidence from three large-scale Asian cohorts.

JHEP reports : innovation in hepatology·2025
Same author

Sex-dependent effects of maternal high-fat diet during lactation in the offspring of adult THY-Tau22 mice.

Brain : a journal of neurology·2025
Same author

3D imaging and quantitative analysis of adipocytes in situ and ex situ.

Adipocyte·2025

This study introduces a new variance model for microarray data analysis, improving statistical test power and accuracy for differential gene expression. The model offers a robust and efficient approach compared to existing methods.

Area of Science:

  • Genomics
  • Statistical Bioinformatics
  • Computational Biology

Background:

  • Accurate variance modeling is crucial for powerful and precise statistical tests in microarray data analysis.
  • Existing methods for variance modeling in gene expression analysis have limitations in robustness and adaptability.

Purpose of the Study:

  • To propose a novel structural variance model incorporating condition and random gene effects for microarray data.
  • To develop a simple estimation procedure for variance model parameters using empirical variances.
  • To evaluate the performance of the proposed model against established and recent methods.

Main Methods:

  • A structural model was applied to variances, including condition and random gene effects.
  • A simple estimation procedure was developed by working on empirical variances.

Related Experiment Videos

  • The proposed model was compared with gene-by-gene analysis, homogeneous variance models, SAM, VarMixt, and Limma using real and simulated data.

    • Main Results:

    • The proposed model demonstrated superior power compared to gene-by-gene analysis.
    • It showed greater robustness to false positives than the homogeneous variance model.
    • Performance was comparable to Limma and superior to SAM and VarMixt, even with few replicates.

    Conclusions:

    • The structural variance model offers an efficient and robust approach for differential gene expression analysis in microarrays.
    • Its advantage lies in easily incorporating various factors of variation using linear mixed models on log-transformed variances.
    • The method is computationally fast and suitable for comparing more than two conditions.