Controlling the false discovery rate under a non-parametric graphical dependence model

2025/06/30 by Drew T. Nguyen, Nguyen, Drew T., William Fithian +1
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · #Cancer Genomics and Diagnostics #FOS: Computer and information sciences #FOS: Mathematics #Methodology (stat.ME) #Software Testing and Debugging Techniques #Statistical Methods in Clinical Trials #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.2506.24126

openalex publication_date 2025/06/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We propose sufficient conditions and computationally efficient procedures for false discovery rate control in multiple testing when the p-values are related by a known dependency graph -- meaning that we assume independence of p-values that are not within each other's neighborhoods, but otherwise leave the dependence unspecified. Our methods' rejection sets coincide with that of the Benjamini--Hochberg (BH) procedure whenever there are no edges between BH rejections, and we find in simulations and a genomics data example that their power approaches that of the BH procedure when there are few such edges, as is commonly the case. Because our methods ignore all hypotheses not in the BH rejection set, they are computationally efficient whenever that set is small. Our fastest method, the IndBH procedure, typically finishes within seconds even in simulations with up to one million hypotheses.

Citations

• A short review on the maximum clique problem algorithms with classical, AI, and quantum methods
• On Controlling the False Discovery Rate in Multiple Testing of the Means of Correlated Normals Against Two-Sided Alternatives
• Multiple testing under negative dependence
• False discovery rate control with e-values
• Conditional calibration for false discovery rate control under dependence
• The International Genome Sample Resource (IGSR) collection of open human genomic variation resources
• The FDR-Linking Theorem
• A unified treatment of multiple testing with prior knowledge using the p-filter
• Adaptive Controls of FWER and FDR Under Block Dependence
• PLINK: A Tool Set for Whole-Genome Association and Population-Based Linkage Analyses
• Normal approximation under local dependence

Related

• Dependence-Aware False Discovery Rate Control in Two-Sided Gaussian Mean Testing
• Conditional calibration for false discovery rate control under dependence
• Filtering the rejection set while preserving false discovery rate\n control
• Inference with approximate local false discovery rates
• Inequalities for the false discovery rate (FDR) under dependence