HEIDI

HEIDI partitions total heritability of complex traits into contributions from specific genomic regions using a linear mixed model-based approach to characterize genetic architecture.

Key Features:

• Linear mixed model-based estimation: Uses linear mixed models to estimate variance components underlying regional heritability.
• Heritability quantification: Quantifies the proportion of phenotypic variance attributable to genetic variation for complex traits.
• Partitioning by genomic regions: Partitions heritability into an arbitrary number of specified genomic regions, including chromosomal and subchromosomal segments.
• Computational efficiency and accuracy: Implements a computationally efficient algorithm with improved accuracy for estimating regional contributions in large-scale genomic datasets.

Scientific Applications:

• Genome-wide association studies (GWAS): Enables partitioning of heritability in GWAS to dissect the genetic architecture of complex traits.
• Understanding genetic architecture: Provides insight into how different genomic regions contribute to overall genetic variance of traits.
• Chromosomal and subchromosomal analysis: Demonstrated on GWAS data for human height to estimate heritability contributions at chromosomal and subchromosomal levels.

Methodology:

Uses a linear mixed model-based partitioning approach that focuses on partitioning rather than joint estimation; validated by simulations and applied to real GWAS data (human height) to estimate regional heritability.