Improved functional mapping of complex trait heritability with GSA-MiXeR implicates biologically specific gene sets

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) identify genomic loci for complex traits but lack biological interpretation.
• Standard gene set analysis (GSA) methods struggle to pinpoint specific biological pathways from GWAS data.

Purpose of the Study:

• To develop GSA-MiXeR, an analytical tool for gene set analysis (GSA).
• To improve the biological interpretation of genomic loci associated with complex human traits and disorders.
• To quantify partitioned heritability and fold enrichment for small gene sets.

Main Methods:

• GSA-MiXeR fits a heritability model for individual genes, considering linkage disequilibrium.
• The tool quantifies partitioned heritability and fold enrichment, particularly for small gene sets.
• Method validation was performed using extensive simulations and sensitivity analyses.

Main Results:

• GSA-MiXeR was applied to diverse complex traits and disorders, including schizophrenia.
• The tool prioritized gene sets with higher biological specificity than standard GSA approaches.
• For schizophrenia, GSA-MiXeR implicated voltage-gated calcium channel function and dopaminergic signaling.

Conclusions:

• GSA-MiXeR enhances the biological interpretation of GWAS findings for complex diseases.
• Biologically relevant gene sets, even small ones, offer insights into disease pathobiology.
• The tool highlights potential therapeutic targets for complex disorders.