Haplotype association analysis for late onset diseases using nuclear family data

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Haplotype-based association studies are crucial for identifying genetic risk factors in late-onset diseases.
• Using unaffected spouses as controls is a common design, often analyzed with the expectation-maximization (EM) algorithm.
• Existing methods for incorporating offspring data into spouse-pair case-control studies are inefficient or biased.

Purpose of the Study:

• To develop a unified, likelihood-based method for haplotype inference that efficiently utilizes offspring genotype information.
• To improve the accuracy of haplotype frequency estimation and probabilistic haplotype reconstruction in case-control studies.
• To provide robust statistical tests for disease-haplotype association.

Main Methods:

• A novel EM algorithm-based approach for haplotype inference using spouse pairs and available offspring genotypes.
• Probabilistic apportionment of ambiguous haplotypes, leveraging offspring data where available and standard EM for others.
• Development of likelihood ratio and permutation tests for disease-haplotype association, including three novel test statistics.

Main Results:

• The proposed method efficiently estimates haplotype frequencies and enables probabilistic haplotype reconstruction using the entire sample.
• It effectively incorporates offspring genotype information, overcoming limitations of existing methods.
• The developed statistical tests are designed to enhance the detection of disease-haplotype associations.

Conclusions:

• This unified method offers a more efficient and statistically sound approach to haplotype inference in genetic studies with spouse-pair designs and offspring data.
• The findings facilitate more accurate identification of genetic variants associated with complex diseases.
• The study provides valuable tools for statistical geneticists and researchers in disease association studies.