A rank-based normalization method with the fully adjusted full-stage procedure in genetic association studies.

Area of Science:

• Genetic Epidemiology
• Statistical Genetics
• Human Complex Trait Genetics

Background:

• Genotype-phenotype association studies are crucial for understanding complex human traits.
• Statistical methods are needed to analyze associations between traits and genetic variants, accounting for environmental and social factors.
• Linear regression is common for quantitative traits, but often assumes normality, which is frequently violated.

Purpose of the Study:

• To propose and evaluate a novel statistical test for genotype-phenotype association analysis involving rare variants.
• To address the challenges posed by non-normal trait distributions and confounding factors in genetic association studies.
• To improve the accuracy and power of rare variant association testing.

Main Methods:

• Development of a fully adjusted full-stage inverse normal transformation (INT) test.
• Utilizing simulation studies to compare the proposed method with existing INT approaches (fully adjusted two-stage INT, INT-based omnibus test) and other methods (quantile/median regression, Yeo-Johnson transformation).
• Theoretical analysis to support the desirable properties of the proposed method.

Main Results:

• The fully adjusted full-stage INT demonstrates superior performance compared to existing INT methods for rare variant association testing, particularly when genotypes are uncorrelated with covariates.
• The proposed method effectively handles non-normal trait distributions, retaining the benefits of the fully adjusted two-stage INT while mitigating its limitations for rare variant analysis.
• Simulation results confirm the theoretical advantages of the fully adjusted full-stage INT.

Conclusions:

• The fully adjusted full-stage INT is a robust and effective method for genotype-phenotype association studies involving rare variants and non-normal traits.
• This new approach offers improved statistical power and Type I error rate control in genetic epidemiology.
• The method provides a valuable tool for advancing the study of complex human traits through genetic analysis.