Imputed Rare‐Variant Aggregate Analysis Is a Powerful Approach to Detect Genetic Associations

Abstract

ABSTRACT Imputation can cost‐effectively generate genotypes for millions of variants. This study evaluated the power of performing rare‐variant aggregate association tests using imputed genotypes. UK Biobank exome‐sequence data and variants obtained via imputation using the HRC (r1.1) ( N = 32,488 samples) or TOPMed (r3) ( N = 133,597 samples) reference panels were analyzed. Simulations were used to compare the power of performing rare‐variant aggregate association analysis using either sequence or imputed data. The number of genes available for analysis, that is, those with ≥ 2 rare variants, was approximately the same for exome‐sequence and TOPMed imputed data, but HRC imputed data had ~10% fewer genes. For imputed data, using a less stringent R 2 threshold (i.e., > 0.3 vs. > 0.8) led to greater power to detect aggregate associations due to additional rare variants included in the analysis. Exome‐sequence data provided the highest power to detect rare‐variant aggregate associations. TOPMed imputed variants usually had < 20% reduction in power in comparison to exome‐sequence data, whereas HRC imputed variants provided substantially less power. Rare‐variant aggregate association analyses using exome‐sequence data, and imputed variants for PCSK9 and low‐density lipoprotein and APOC3 and triglyceride levels were also performed. For these analyses, even when only ultra‐rare variants (minor allele frequency < 0.001) were analyzed, significant aggregate associations could be detected for all data types. Although there is a decrease in power for rare‐variant aggregate association tests when analyzing imputed variants compared to sequence data due to missing variants and uncertainty in genotypes, analysis of imputed data is a viable approach to detect rare‐variant aggregate associations when sequence data are unavailable.