Decomposing genomic variance using information from GWA, GWE and eQTL analysis.

Area of Science:

• Genomics
• Quantitative Genetics
• Bioinformatics

Background:

• Traditional genome-wide association (GWA), genome-wide expression (GWE), and expression quantitative trait locus (eQTL) analyses use a bottom-up approach to link molecular variants to complex traits.
• A top-down modeling approach partitioning overall genomic variance may offer deeper insights into the genetic architecture of complex traits.

Purpose of the Study:

• To test a whole-genome variance components analysis for partitioning genomic variance.
• To characterize the genetic architecture of growth-related traits in a mouse F2 population using GWA, GWE, and eQTL data.
• To introduce and utilize Area Under the Curve (AUC) measures for genomic variance profiles.

Main Methods:

• Performed whole-genome variance components analysis on growth-related traits in a mouse F2 population.
• Partitioned genomic variance using GWA, GWE, and eQTL data.
• Ordered single nucleotide polymorphisms (SNPs) by P-values and calculated AUCs to characterize genetic architecture.

Main Results:

• Observed traits exhibited a genomic variance profile significantly deviating from the infinitesimal model, particularly body weight and body fat.
• SNPs with high trait-specific regulatory potential explained more genomic variance than those with high overall regulatory potential.
• AUC measures effectively quantify SNP importance and deviation from the infinitesimal model.

Conclusions:

• The top-down variance partitioning approach provides a global understanding of trait genetic architecture.
• The shape of the genomic variance profile curve indicates the number of SNPs controlling phenotypic variance.
• Trait-specific regulatory SNPs play a crucial role in the genetic basis of complex traits.