"Enhanced Genetic Analysis Tool Boosts Discovery of Disease-Causing Genes"
Originally Published 1 year ago — by Nature.com
Researchers have proposed a new statistical framework called causal-TWAS (cTWAS) to address limitations in existing methods for transcriptome-wide association studies (TWAS). cTWAS aims to control for genetic confounders and improve the discovery of causal genes from genome-wide association studies (GWAS). Through simulations and real data applications, cTWAS demonstrated accurate parameter estimation, well-calibrated posterior inclusion probabilities (PIPs), and reduced false discoveries compared to standard TWAS, colocalization, and Mendelian randomization-based methods. In an application to GWAS of LDL cholesterol, cTWAS outperformed standard TWAS in distinguishing known LDL-related genes from nearby bystander genes, demonstrating its potential for reliable gene discovery in complex traits.