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 method to increase the power of multiple testing procedures through sample splitting.

Daniel Rubin1, Sandrine Dudoit, Mark van der Laan

  • 1Division of Biostatistics, School of Public Health, University of California, Berkeley, CA, USA. drubin@stat.berkeley.edu

Statistical Applications in Genetics and Molecular Biology
|October 20, 2006
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

African American Dementia Caregiver Strategies Inventory: Initial descriptive and psychometric evaluation.

Rehabilitation psychology·2026
Same author

Development and Application of a Desirability of Outcome Ranking (DOOR) Endpoint for Acute Bacterial Skin and Skin Structure Infections.

Open forum infectious diseases·2026
Same author

UNLOCKING MULTI-SAMPLE DIFFERENTIAL EXPRESSION FOR SPATIAL TRANSCRIPTOMICS DATA WITH TESSERA.

bioRxiv : the preprint server for biology·2026
Same author

Transportability to the European Population of Efficacy of Belumosudil as Compared With Physician's Choice of Best Available Therapy for the Treatment of Chronic Graft Versus Host Disease.

Transplantation and cellular therapy·2026
Same author

Methodological and regulatory considerations for causal AI in drug development.

NPJ digital medicine·2026
Same author

Discovery of critical thresholds in mixed exposures and estimation of policy intervention effects.

Journal of causal inference·2026
Same journal

Balanced mediated pathway detection in genomic data.

Statistical applications in genetics and molecular biology·2026
Same journal

Annealed variational mixtures for disease subtyping and biomarker discovery.

Statistical applications in genetics and molecular biology·2026
Same journal

Performance of the permutation test approach with base calling errors for detecting changes in variant allele frequencies in ctDNA for a single patient.

Statistical applications in genetics and molecular biology·2026
Same journal

BLOG: Bayesian longitudinal omics with group constraints.

Statistical applications in genetics and molecular biology·2026
Same journal

AI-driven risk prediction and categorization in cystic fibrosis leveraging AttentiveLSTM and Fox Wolf Optimizer.

Statistical applications in genetics and molecular biology·2026
Same journal

Perfect collinearity not created equal: measuring and visualizing the severity of multi-collinearity of modern omics data.

Statistical applications in genetics and molecular biology·2026
See all related articles

This study introduces a novel sample-splitting method for multiple testing procedures to enhance true discoveries. By estimating optimal cutoffs, this approach significantly outperforms traditional common cutoff methods while maintaining Type-I error control.

Area of Science:

  • Statistics
  • Statistical inference
  • Hypothesis testing

Background:

  • Multiple testing procedures aim to control Type-I errors (false positives) when testing numerous hypotheses.
  • Existing methods often use a common cutoff, which may not be optimal for maximizing true discoveries.
  • The ideal optimal cutoffs are unknown as they depend on the true data distribution.

Purpose of the Study:

  • To develop a practical multiple testing procedure that maximizes true positives while controlling false positives.
  • To investigate the performance of a sample-splitting approach for estimating optimal test statistic cutoffs.
  • To compare the proposed method against the benchmark common cutoff procedure.

Main Methods:

  • A sample-splitting technique is employed to estimate optimal test statistic cutoffs from a subset of data.

Related Experiment Videos

  • These estimated cutoffs are then applied to the remaining data for hypothesis testing.
  • The performance is evaluated based on the expected number of true positives and Type-I error control.

    • Main Results:

    • The sample-splitting method, using estimated optimal cutoffs, can significantly increase true discoveries compared to the common cutoff benchmark.
    • This improvement is achieved while strictly adhering to user-supplied thresholds for the expected number of false positives.
    • The proposed method demonstrates superior power in certain scenarios, leading to more reliable scientific findings.

    Conclusions:

    • Sample splitting offers a practical and powerful approach to multiple testing, enhancing the ability to detect true effects.
    • The method provides a viable alternative to traditional common cutoff procedures, especially when maximizing discoveries is critical.
    • This research advances statistical methodology for robust hypothesis testing in complex data scenarios.