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

Statistical significance threshold criteria for analysis of microarray gene expression data.

Cheng Cheng1, Stanley B Pounds, James M Boyett

  • 1Department of Biostatistics, St. Jude Children's Research Hospital. cheng.cheng@stjude.org

Statistical Applications in Genetics and Molecular Biology
|May 2, 2006
PubMed
Summary

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

Egg recognition and chick discrimination in colonial breeding birds.

Behavioural processes·2019
Same author

Association of total sedentary behaviour and television viewing with risk of overweight/obesity, type 2 diabetes and hypertension: A dose-response meta-analysis.

Diabetes, obesity & metabolism·2019
Same author

AAV-ie enables safe and efficient gene transfer to inner ear cells.

Nature communications·2019
Same author

The effect of the overexpression of miR-490 on the biological function of prostate cancer cell PC-3.

Journal of B.U.ON. : official journal of the Balkan Union of Oncology·2019
Same author

Treatment response and outcome of children with T-cell acute lymphoblastic leukemia expressing the gamma-delta T-cell receptor.

Oncoimmunology·2019
Same author

Hypertension is a modifiable risk factor for osteonecrosis in acute lymphoblastic leukemia.

Blood·2019

New statistical criteria improve microarray analysis when standard false discovery rate (FDR) methods yield no significant findings. These methods balance statistical significance and FDR for better gene target discovery.

Area of Science:

  • Bioinformatics
  • Statistical Genetics
  • Computational Biology

Background:

  • Microarray data analysis commonly uses false discovery rate (FDR) control, with q-value plots offering multiple perspectives.
  • Existing FDR methods offer limited guidance for adjusting significance thresholds when customary FDR levels (0.01, 0.05, 0.1) do not yield significant results.

Purpose of the Study:

  • To develop novel statistical significance criteria for large-scale multiple testing that complement existing FDR methods.
  • To provide a framework for balancing significance thresholds and FDR levels using sound statistical and biological considerations.

Main Methods:

  • Introduced three new statistical significance criteria: profile information criterion, total error proportion, and guide-gene driven selection.
  • Investigated the error properties of these criteria theoretically and through simulations.

Related Experiment Videos

  • Applied and compared the proposed methods using genomic screening data for Arf gene targets.

    • Main Results:

    • The profile information criterion relates to FDR control and minimax estimation.
    • The total error proportion is linked to the asymptotic properties of FDR control and total error risk.
    • The guide-gene driven selection integrates statistical significance with existing biological knowledge.
    • Simulation studies demonstrated the operating characteristics of the new criteria compared to q-values, using a gene regulatory pathway model.

    Conclusions:

    • The developed statistical significance criteria offer meaningful alternatives for large-scale multiple tests, especially when standard FDR control is insufficient.
    • These methods provide a guideline for investigators to make informed trade-offs between significance thresholds and FDR levels.
    • The proposed approaches enhance the interpretability and utility of microarray data analysis for biological discovery.