A method to estimate the contribution of rare coding variants to complex trait heritability

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Complex traits exhibit significant heritability not fully explained by common variants.
• Rare coding variants (MAF < 0.01) are hypothesized to contribute to this missing heritability.
• Accurate estimation of rare variant heritability is crucial for understanding genetic architecture.

Purpose of the Study:

• To develop and validate a novel framework, the Rare variant heritability (RARity) estimator, for assessing rare variant heritability (h²_RV).
• To quantify the contribution of rare variants to the heritability of 31 complex traits using UK Biobank data.
• To explore gene-level rare variant heritability and identify potential gene-phenotype relationships.

Main Methods:

• Development of the RARity estimator framework for heritability analysis.
• Application of RARity to 31 complex traits in a large UK Biobank cohort (n=167,348).
• Comparison of gene-level aggregated vs. unaggregated rare variant heritability estimates.

Main Results:

• Gene-level aggregation of rare variants resulted in a substantial loss of heritability (79%).
• Unaggregated rare variants explained >5% heritability for 27 traits, with height showing the highest h²_RV (21.9%).
• RARity recovered pedigree-based heritability estimates for 11 traits and identified 11 novel gene-phenotype associations.

Conclusions:

• Rare coding variants significantly contribute to the heritability of complex traits.
• The RARity estimator provides a robust method for assessing rare variant heritability at both variant and gene levels.
• Current in silico prediction methods are insufficient for identifying functional rare variants impacting complex traits, necessitating new approaches.