SumHer

SumHer estimates confounding bias, SNP heritability, heritability enrichments, and genetic correlations from genome-wide association study (GWAS) summary statistics while supporting user-specified heritability models.

Key Features:

• Customizable Heritability Models: Enables specification of user-defined heritability models, distinguishing it from LD Score Regression (LDSC).
• Estimation of Confounding Bias: Estimates confounding bias in GWAS summary statistics that can influence interpretation of association results.
• SNP Heritability Estimation: Calculates SNP heritability to quantify the proportion of phenotypic variance attributable to common SNPs.
• Heritability Enrichment Analysis: Assesses enrichments of heritability across SNP categories defined by functional annotations.
• Genetic Correlation Estimation: Estimates genetic correlations between traits using GWAS summary statistics.

Scientific Applications:

• Large-scale GWAS reanalysis: Applied to 24 association studies (mean sample size 121,000) to reassess heritability estimates and confounding corrections.
• Enrichment reassessment: Reanalyzed reported enrichments and found contrasting results to previous LDSC-based analyses, including no categories exceeding twofold enrichment versus prior reports of a 13-fold enrichment in conserved regions and multiple categories >3-fold.

Methodology:

Applies user-specified heritability models to GWAS summary statistics to estimate confounding bias, SNP heritability, heritability enrichments, and genetic correlations, and reanalyzes reported enrichments under alternative analytical frameworks.