FABIO: TWAS fine-mapping to prioritize causal genes for binary traits.

Area of Science:

• Genetics
• Bioinformatics
• Statistical Genetics

Background:

• Transcriptome-wide association studies (TWAS) integrate gene expression and genome-wide association studies (GWAS) to find gene-trait links.
• Current TWAS fine-mapping methods often analyze genes individually or focus on quantitative traits, limiting their scope.
• Existing methods may be suboptimal for binary traits and large genomic regions.

Purpose of the Study:

• To introduce FABIO, a novel TWAS fine-mapping method designed for binary traits.
• To enable joint modeling of all genes on a chromosome for improved causal gene identification.
• To enhance the power and accuracy of TWAS for disease-related genetic studies.

Main Methods:

• FABIO utilizes a probit model to connect genetically regulated expression (GReX) with binary outcomes.
• It accounts for the correlation of GReX among all genes within a chromosome.
• The method jointly models all genes across an entire chromosome.

Main Results:

• Simulations demonstrated FABIO's effectiveness in controlling false discoveries and increasing power.
• Application to UK Biobank data for six binary diseases showed significant reduction in causal gene set size (27.9%-36.9%).
• FABIO identified potential causal genes, including GATA3 (asthma), ABCG2 (gout), and SH2B3 (hypertension).

Conclusions:

• FABIO is an effective tool for TWAS fine-mapping of binary disease traits.
• The method offers substantial power gains and improved precision over existing approaches.
• FABIO facilitates the prioritization of potentially causal genes for complex diseases.