Improving power for testing genetic association in case-control studies by reducing the alternative space

Area of Science:

• Genetics
• Statistical Genetics
• Epidemiology

Background:

• Cochran-Armitage trend test is standard for genetic association studies.
• Optimal trend test performance relies on correct genetic model specification.
• Underlying genetic models are often unknown in real-world studies.

Purpose of the Study:

• To adapt existing genetic model selection (GMS) methods for unknown risk alleles.
• To introduce a novel approach using genetic model exclusion.
• To enhance the power of detecting genetic associations in case-control studies.

Main Methods:

• Modification of the existing GMS method to handle unknown risk alleles.
• Development of a new method excluding unsupported genetic models.
• Application of proposed methods to genome-wide association study (GWAS) data.

Main Results:

• The modified GMS and exclusion approaches increase the power to detect true associations.
• The genetic model exclusion method demonstrates superior performance compared to existing methods.
• Proposed methods were validated using simulation results and real GWAS data.

Conclusions:

• The proposed genetic model exclusion approach offers improved power for genetic association detection.
• These methods provide robust tools for analyzing genetic markers when the genetic model is uncertain.
• The study contributes to more effective analysis of complex disease genetics.