varGWAS

varGWAS detects genetic loci that influence trait variance by prioritising variance associations to reveal gene-interaction effects on biomarker concentrations in large-scale datasets such as the UK Biobank.

Key Features:

• Variance Prioritisation Approach: Implements a regression-based Brown-Forsythe test together with variance effect estimates to detect associations between single nucleotide polymorphisms (SNPs) and trait variance rather than mean trait values.
• Genome-wide SNP-variance Analysis: Performs genome-wide association analysis specifically targeting SNP-variance effects to identify variance quantitative trait loci (vQTLs).
• Application to Blood Biomarkers: Applied to 30 blood biomarkers in the UK Biobank, identifying 468 variance quantitative trait loci across 24 biomarkers.
• Interaction Detection: Identified 82 gene-environment and six gene-gene interactions independent of strong scale or phantom effects, replicating known findings and discovering novel epistatic effects including TREHrs12225548 x FUT2rs281379 and ZNF827rs4835265 x NEDD4Lrs4503880 affecting alkaline phosphatase and gamma glutamyltransferase.

Scientific Applications:

• Genetic interaction discovery: Identifies loci influencing biomarker variance to reveal gene-environment and gene-gene interactions that modulate trait variability.
• Preclinical and biomarker stratification: Enables detection of subgroups with differential biomarker responses relevant to preclinical drug development and biomarker-based stratification.
• Personalized medicine: Informs targeted therapeutic strategies by characterising variance-associated loci that may drive heterogeneous treatment responses.

Methodology:

Uses variance prioritisation via a regression-based Brown-Forsythe test, estimates of variance effects, and genome-wide association analysis targeting SNP-variance effects.