Robust rare variant association testing for quantitative traits in samples with related individuals

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• High-throughput sequencing necessitates advanced statistical tests for rare genetic variants.
• Family-based sequencing studies offer advantages like error reduction and increased power for rare alleles.
• Accounting for familial correlation is crucial for accurate type I error control and power in genetic association studies.

Purpose of the Study:

• To develop a robust rare-variant association test for family data.
• To generalize existing methods like SKAT-O for related individuals.
• To provide a computationally efficient and powerful tool for genetic association analysis in sequencing studies.

Main Methods:

• Proposed MONSTER (Minimum P-value Optimized Nuisance parameter Score Test Extended to Relatives), a novel statistical test.
• Utilized a mixed effects model to account for covariates and additive polygenic effects.
• Developed an analytical approach for P-value assessment, avoiding complex permutations.

Main Results:

• MONSTER effectively accounts for family structure and demonstrates computational efficiency.
• Simulation studies show MONSTER offers superior power compared to existing tests for related individuals.
• Applied MONSTER to Framingham Heart Study data, replicating associations with three genes for high-density lipoprotein cholesterol.

Conclusions:

• MONSTER is a powerful and efficient statistical test for rare-variant association analysis in family-based sequencing studies.
• The method successfully addresses limitations of existing tests for related samples.
• MONSTER facilitates the discovery of genetic variants influencing complex human traits.