A variance component based multi-marker association test using family and unrelated data.

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Family data integration enhances genetic association studies, particularly for rare variant analysis.
• Current methods may face power limitations when testing multiple variants with varying effects.

Purpose of the Study:

• To develop a novel variance-component based association test.
• To enable joint testing of multiple common or rare variants using both family and unrelated samples.

Main Methods:

• The approach aggregates genetic information based on similarity, not genotype scores, to avoid power loss.
• Leverages linkage disequilibrium (LD) information and allows adjustment for covariates and non-genetic familial influences.
• Implements an adaptively estimated degrees of freedom for the test statistic.

Main Results:

• The method demonstrates robust Type I error control in simulations.
• Achieves good statistical power for detecting genetic associations.
• Effectively handles variants with differing effect directions and strengths.

Conclusions:

• The developed method provides a powerful and flexible tool for genetic association analysis.
• The R package "fassoc" facilitates data analysis and exploration in genetic studies.
• The approach is valuable for incorporating familial data and testing complex genetic architectures.