AP-SKAT: highly-efficient genome-wide rare variant association test

Area of Science:

• Genetics and Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) link single-nucleotide polymorphisms (SNPs) to phenotypes.
• Rare variant analysis often involves grouping variants by gene or pathway.
• The sequential kernel association test (SKAT) is a common method but suffers from biased p-values due to asymptotic approximations.

Purpose of the Study:

• To develop an efficient and accurate adaptive SKAT procedure (AP-SKAT).
• To address the computational limitations of traditional permutation tests for p-value calculation in genetic association studies.

Main Methods:

• An adaptive permutation test procedure (AP-SKAT) was devised.
• The procedure adaptively stops permutation testing based on confidence intervals for binomial distribution p-value estimation.
• Performance was evaluated using genotype data from the SKAT R package and the 1000 Genome Project, including whole genome sequencing and SNP array data.

Main Results:

• AP-SKAT efficiently classifies significant SNP sets and ranks them by permuted p-values.
• The proposed procedure demonstrates high efficiency and accuracy comparable to standard methods.
• Type I error rates were controlled, and power and sample size calculations were favorable across different datasets.

Conclusions:

• AP-SKAT achieves competitive power and sample size for various genetic data types and sample sizes.
• It accurately estimates p-values for significant SNP sets within realistic computational times.
• The method is suitable for current whole genome sequencing and SNP array data and adaptable to other association tests.