Pooled association tests for rare variants in exon-resequencing studies

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Deep sequencing generates vast genetic data in disease studies.
• Detecting associations with individual rare variants is challenging.
• Pooling rare variants offers a strategy to identify susceptibility genes.

Purpose of the Study:

• To develop a statistical method for detecting associations between multiple rare variants in protein-coding genes and traits.
• To incorporate functional predictions of missense variants into association testing.
• To evaluate the method's performance using simulations and real-world data.

Main Methods:

• Regression analysis of phenotypic values on genotype scores.
• Variable allele-frequency thresholding.
• Incorporation of computational predictions for missense variant functional effects.
• Permutation testing for statistical significance assessment.

Main Results:

• The developed statistical method demonstrates power in detecting associations with quantitative and dichotomous traits.
• The approach effectively analyzes pooled rare variants.
• The method was successfully applied to empirical sequencing data from three disease studies.

Conclusions:

• The proposed statistical method provides a robust framework for identifying disease susceptibility genes using rare variants.
• This approach enhances the utility of deep sequencing data in genetic association studies.
• The method offers a valuable tool for complex trait genetics research.