LocusPackRat: a Semi-Automated Framework for Prioritizing Candidate Genes from Large GWAS Intervals

Area of Science:

• Genetics
• Bioinformatics
• Systems Biology

Background:

• Genome-wide association studies (GWAS) often identify large genomic regions (loci) containing numerous genes, complicating the identification of causal genes.
• Reduced mapping resolution in model organisms further challenges gene prioritization from GWAS loci.

Purpose of the Study:

• To develop a semi-automated, extendible package named LocusPackRat for accelerating candidate gene prioritization from GWAS loci.
• To integrate diverse evidence types, including gene expression and regulatory data, with functional and disease annotations.

Main Methods:

• LocusPackRat assembles standardized 'packets' of evidence for each gene within a locus.
• These packets merge study-specific data (e.g., differential expression, cis-eQTLs) with external annotations from InterMine and Open Targets.
• The package facilitates side-by-side comparison and team review through identically structured packets.

Main Results:

• Demonstrated LocusPackRat's effectiveness using a GWAS study of cardiac hypertrophy and failure in the Collaborative Cross.
• The tool successfully shortened the pathway from statistical association to mechanistic hypotheses.

Conclusions:

• LocusPackRat enhances the likelihood of successful experimental validation by improving candidate gene prioritization.
• The package is adaptable for various genetic reference populations and human cohorts, offering broad applicability.