Candidate-gene association analysis for a continuous phenotype with a spike at zero using parent-offspring trios

Area of Science:

• Genetics
• Biostatistics
• Statistical genetics

Background:

• Non-normally distributed phenotypes, particularly those with a spike at zero, pose challenges for traditional genetic association tests.
• Coronary artery calcification (CAC) often exhibits a mixed discrete-continuous distribution with a significant proportion of individuals having zero calcium, making it a relevant example.

Purpose of the Study:

• To propose and evaluate a novel statistical test for genetic association with non-normally distributed phenotypes featuring a spike at zero.
• To generalize the quantitative transmission disequilibrium test (QTDT) for complex trait analysis in family-based studies.
• To assess the performance of the proposed test using simulations and real genetic data.

Main Methods:

• Development of generalized additive models for location, scale, and shape (GAMLSS).
• Adaptation of the quantitative transmission disequilibrium test with mating type indicator for mixed distributions.
• Parent-offspring simulations to evaluate statistical properties, including power and Type I error rates.
• Application to Framingham Heart Study data (Genetic Analysis Workshop 16) with simulated CAC phenotypes.

Main Results:

• The proposed generalized additive model test demonstrates robust performance in detecting genetic associations with phenotypes having a spike at zero.
• Simulations confirm the validity and power of the new test compared to existing methods.
• Application to real family data successfully identified associations between genetic markers and simulated CAC traits.

Conclusions:

• The developed statistical framework provides a powerful tool for genetic association studies involving complex, non-normally distributed phenotypes.
• This approach enhances the ability to identify genetic contributors to diseases like coronary artery disease, where phenotypes often present with zero-inflated distributions.
• The generalized additive model test offers a flexible and effective alternative for analyzing genetic data in family studies.