Case-Control Genome-wide Joint Association Study Using Semiparametric Empirical Model and Approximate Bayes Factor

Area of Science:

• Genetics
• Biostatistics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) are crucial for identifying genetic variants associated with diseases.
• Existing methods may not fully leverage all available data, potentially limiting power and reliability.
• Joint modeling of disease status, covariates, and genotype data offers a more comprehensive analytical framework.

Purpose of the Study:

• To develop a semiparametric approach for analyzing case-control GWAS.
• To enable direct and joint modeling of case status, covariates, and genotype counts for improved understanding of disease mechanisms.
• To incorporate external information like disease prevalence for enhanced statistical power.

Main Methods:

• A semiparametric model combining parametric and nonparametric components for case status, covariates, and genotype distributions.
• Empirical likelihood approach to integrate side information such as disease prevalence.
• Profiling to remove nuisance nonparametric components, yielding consistent and asymptotically normal estimates.
• Approximate Bayes Factor (ABF) for hypothesis testing of disease association, considering deviations from Hardy-Weinberg Equilibrium (HWE).

Main Results:

• The proposed method provides a direct and joint modeling framework, leading to more reliable conclusions.
• Incorporation of disease prevalence via empirical likelihood enhances the efficiency of estimates and the power of tests for detecting disease-associated SNPs.
• The developed ABF method is computationally efficient for large-scale GWAS and effectively detects marker-disease associations, even considering HWE deviations.

Conclusions:

• The semiparametric approach offers a robust and powerful methodology for case-control GWAS analysis.
• The integration of disease prevalence and consideration of HWE deviations improve the accuracy and efficiency of identifying disease-associated genetic markers.
• The proposed method enhances the understanding of disease mechanisms and provides reliable conclusions in genetic association studies.