Two-Stage Multiple Test Procedures Controlling False Discovery Rate With Auxiliary Variable and Their Application to...

Area of Science:

• Statistics
• Bioinformatics
• Genomics

Background:

• Multiple hypotheses testing is crucial in analyzing large datasets.
• Controlling the false discovery rate (FDR) is essential for reliable results.
• Traditional methods may lack power when dealing with complex dependencies.

Purpose of the Study:

• To develop novel methodologies for multiple hypotheses testing using auxiliary variables.
• To enhance statistical power and control the FDR effectively.
• To provide a framework for improved gene selection in genomic studies.

Main Methods:

• Incorporation of auxiliary variables into multiple hypotheses testing frameworks.
• Development of two distinct approaches based on the role of auxiliary variables.
• Utilizing the copula method to model the dependence between primary and auxiliary variables.
• Deriving joint distributions from marginal distributions to understand variable relationships.

Main Results:

• The proposed methodologies effectively control the FDR.
• The new methods demonstrate greater statistical power compared to traditional approaches.
• Numerical studies confirm the efficacy of the proposed techniques.
• Application to Set4 Δ mutant data set shows potential for improved gene selection.

Conclusions:

• Auxiliary variables offer a powerful tool for enhancing multiple hypotheses testing.
• The proposed methods provide a statistically robust and more powerful alternative to existing techniques.
• This approach has significant implications for gene selection and other complex data analyses.