Family-based association analysis with ordered categorical phenotypes, covariates and interactions

Area of Science:

• Statistical genetics
• Human genetics
• Genetic epidemiology

Background:

• Genetic association studies with ordered categorical phenotypes often use suboptimal methods for quantitative or binary traits.
• Existing methods may not fully capture the complexity of family-based genetic data.

Purpose of the Study:

• To present a novel likelihood-based analysis method for single nucleotide polymorphism (SNP) genotypes and ordered categorical phenotypes in nuclear families.
• To extend previous work for binary phenotypes to accommodate more complex genetic models.

Main Methods:

• Developed a likelihood-based method for ordered categorical phenotypes in nuclear families of any size.
• Incorporated covariates, gene-gene, and gene-covariate interactions, ascertainment models, and family-specific effects.
• Validated the method with simulated data and applied it to a polymorphism in the c-reactive protein (CRP) gene.

Main Results:

• The proposed method provides interpretable parameter estimates and valid confidence intervals.
• Analysis of CRP gene polymorphism in systemic lupus erythematosus families revealed a sex-specific association.
• The CRP polymorphism was associated with anti-nuclear autoantibody (ANA) production in females, but not in males.

Conclusions:

• The new method offers a more robust and flexible approach for genetic association studies with ordered categorical traits.
• Demonstrated the utility of the method in identifying a sex-specific genetic association relevant to systemic lupus erythematosus.
• Highlights the importance of considering interactions in genetic analyses for complex diseases.