A general class of association tests for family-based data using weight functions

Area of Science:

• Genetics
• Statistical Genetics
• Human Genetics

Background:

• Genetic association studies are crucial for identifying genes linked to diseases.
• Existing methods for family-based association testing have limitations, especially with complex family structures.

Purpose of the Study:

• To develop a general statistical framework for association tests using nuclear families with multiple affected children.
• To propose a class of association test statistics for diallelic and multiallelic markers.
• To evaluate the power and performance of the proposed tests compared to existing methods.

Main Methods:

• The framework is based on the symmetry of transmitted/nontransmitted alleles from heterozygous parents under the null hypothesis of no association.
• Proposed test statistics generalize existing tests like the transmission disequilibrium test (TDT) for trios and methods for affected sib pairs/sibships.
• The most powerful test was selected for simulation and real data analysis.

Main Results:

• The proposed association test demonstrated superior power in detecting genetic associations compared to previously published methods when using affected sibships.
• In the analysis of sitosterolemia data, the most significant association (P=0.0012) was found for a marker locus on the same bacterial artificial chromosome as the disease locus.

Conclusions:

• The developed statistical framework provides a robust and powerful approach for family-based genetic association studies.
• The proposed tests are effective in identifying disease-associated loci, particularly in complex family structures.
• The findings in sitosterolemia highlight the utility of the new methods in real-world genetic research.