Enhancing power of rare variant association test by Zoom-Focus Algorithm (ZFA) to locate optimal testing region

TL;DR

The paper introduces the Zoom-Focus Algorithm (ZFA), a method that improves the power of rare variant association tests by identifying optimal testing regions within genomic data, leading to more accurate detection of genetic markers.

Contribution

The paper presents ZFA, a novel wrapper algorithm that enhances rare variant test power by adaptively locating the most relevant genomic regions for association testing.

Findings

ZFA increased test power over 10-fold in simulations.

Applied ZFA to real data, uncovering biologically relevant markers.

ZFA outperforms traditional fixed-region methods.

Abstract

Motivation: Exome or targeted sequencing data exerts analytical challenge to test single nucleotide polymorphisms (SNPs) with extremely small minor allele frequency (MAF). Various rare variant tests were proposed to increase power by aggregating SNPs within a fixed genomic region, such as a gene or pathway. However, a gene could contain from several to thousands of markers, and not all of them may be related to the phenotype. Combining functional and non-functional SNPs in arbitrary genomic region could impair the testing power. Results: We propose a Zoom-Focus algorithm (ZFA) to locate the optimal testing region within a given genomic region, as a wrapper function to be applied in conjunction with rare variant association tests. In the first Zooming step, a given genomic region is partitioned by order of two, and the best partition is located within all partition levels. In the next Focusing step, boundaries of the zoomed region are refined. Simulation studies showed that ZFA substantially enhanced the statistical power of rare variant tests by over 10 folds, including the WSS, SKAT and W-test. The algorithm is applied on real exome sequencing data of hypertensive disorder, and identified biologically relevant genetic markers to metabolic disorder that are undiscoverable by testing using full gene. The proposed algorithm is an efficient and powerful tool to increase the effectiveness of rare variant association tests for exome sequencing datasets of complex disorder.