Bayesian Genome-wide TWAS Method to Leverage both cis- and trans-eQTL Information through Summary Statistics

Area of Science:

• Genetics
• Genomics
• Computational Biology

Background:

• Transcriptome-wide association studies (TWASs) integrate gene expression and genetic data to study complex traits.
• Existing TWAS methods are computationally intensive and often exclude distant trans-expression quantitative trait loci (eQTLs), which significantly influence gene expression.
• Incorporating trans-eQTLs is crucial for a comprehensive understanding of gene regulation in complex diseases.

Purpose of the Study:

• To develop a computationally efficient Bayesian genome-wide TWAS (BGW-TWAS) method that leverages both cis- and trans-eQTL information.
• To improve the power of TWAS by including trans-eQTLs, which are often overlooked due to computational challenges.
• To identify novel genetic associations for complex traits, such as Alzheimer dementia (AD), by utilizing comprehensive eQTL information.

Main Methods:

• Developed the Bayesian genome-wide TWAS (BGW-TWAS) method using Bayesian variable selection regression.
• The method efficiently utilizes summary statistics from standard eQTL analyses to incorporate both cis- and trans-eQTLs.
• Applied BGW-TWAS to individual-level and summary-level GWAS data for Alzheimer dementia.

Main Results:

• Simulation studies demonstrated that BGW-TWAS achieves higher statistical power than existing methods that do not consider trans-eQTLs.
• BGW-TWAS identified significant associations between genetically regulated gene expression (GReX) of ZC3H12B and Alzheimer dementia (AD) pathology, driven by trans-eQTLs.
• The GReX of KCTD12 was associated with β-amyloid, driven by both cis- and trans-eQTLs. Analysis of AD GWAS data identified 13 significant genes, including known risk genes HLA-DRB1 and APOC1.

Conclusions:

• BGW-TWAS is a powerful and computationally efficient method for integrating cis- and trans-eQTLs in genome-wide association studies.
• The method successfully identified novel genetic associations for Alzheimer dementia, highlighting the importance of trans-eQTLs in complex trait genetics.
• BGW-TWAS provides a valuable tool for discovering gene-trait associations driven by distant genetic effects.