Statistical methods for SNP heritability estimation and partition: A review

In GWAS studies, SNP heritability measures the proportion of phenotypic variance explained by all measured SNPs. Accurate estimation of SNP heritability can help us better understand the degree to which measured genetic variants influence phenotypes. Over the last decade, a variety of statistical methods and software tools have been developed for SNP heritability estimation with different data types including genotype array data, imputed genotype data, whole-genome sequencing data, RNA sequencing data, and bisulfite sequencing data. However, a thorough technical review of these methods, especially from a statistical and computational viewpoint, is currently missing. To fill this knowledge gap, we present a comprehensive review on a broad category of recently developed and commonly used SNP heritability estimation methods. We focus on their modeling assumptions; their interconnected relationships; their applicability to quantitative, binary and count phenotypes; their use of individual level data versus summary statistics, as well as their utility for SNP heritability partitioning. We hope that this review will serve as a useful reference for both methodologists who develop heritability estimation methods and practitioners who perform heritability analysis.