A general framework for detecting disease associations with rare variants in sequencing studies

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Rare variants are increasingly recognized for their substantial role in complex human diseases.
• High-throughput sequencing technologies now enable extensive rare variant data generation in large populations.

Purpose of the Study:

• To develop a generalized framework for association testing of rare variants.
• To integrate mutation information across multiple sites within genes for enhanced genetic analysis.
• To accommodate diverse study designs, phenotypes, and covariates.

Main Methods:

• A novel framework combining mutation information from multiple variant sites within genes.
• Utilizing appropriate regression models to link genetic data to disease phenotypes.
• Deriving theoretically optimal procedures for combining rare mutations and constructing test statistics.
• Allowing for fixed or variable allele-frequency thresholds and varied mutation effect directions.

Main Results:

• The proposed methods demonstrate superior statistical power and computational efficiency compared to existing approaches.
• Successfully applied to a deep-resequencing study, identifying rare variants associated with total cholesterol.
• The framework is versatile, supporting various study designs (case-control, cohort, etc.) and phenotypes (binary, quantitative).

Conclusions:

• The developed framework provides a robust and efficient approach for rare variant association testing.
• This methodology facilitates the discovery of novel genetic associations for complex diseases.
• Associated software is publicly available, promoting wider research application.