A covering method for detecting genetic associations between rare variants and common phenotypes

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association (GWA) studies often struggle to identify associations with common variants.
• Multiple rare variants (RVs) may collectively influence disease etiology, posing a challenge for traditional GWA approaches.

Purpose of the Study:

• To develop and validate a new algorithm, RareCover, for detecting associations involving multiple rare variants.
• To assess the performance of RareCover compared to existing methods like collapsing and weighted-collapsing strategies.

Main Methods:

• Developed RareCover, an algorithm that aggregates disparate rare variants with low effect and modest penetrance, irrespective of their genomic location.
• Conducted extensive simulations to evaluate RareCover's power across various penetrance and population attributable risk (PAR) values.
• Applied RareCover to re-sequencing data from individuals with extreme Body Mass Index (BMI) values, focusing on FAAH and MGLL genes.

Main Results:

• Simulations demonstrated RareCover's superior power over collapsing-based methods.
• Analysis of BMI cohort data identified significant associations in the upstream regulatory regions of FAAH and MGLL genes.
• These associations suggest that rare variants disrupt gene expression, impacting cannabinoid metabolism.

Conclusions:

• RareCover offers a powerful approach for identifying genetic associations involving rare variants.
• Incorporating rare variants into genetic association studies is crucial for a comprehensive understanding of disease etiology.
• The findings highlight the potential role of rare variants in metabolic regulation and complex traits like obesity.