MultiXcan

MultiXcan is a statistical method that enhances our understanding of the biological mechanisms underlying genome-wide association studies (GWAS) hits and bolsters our capacity to identify therapeutic targets. This method achieves its objectives by efficiently integrating expression quantitative trait loci (eQTL) and GWAS studies. MultiXcan addresses limitations such as restricted eQTL sample sizes and the absence of relevant developmental and disease contexts that can hinder association detection.

At the heart of MultiXcan is its ability to capitalize on the extensive sharing of eQTLs across different tissues and contexts, thereby significantly improving the identification of potential target genes. It accomplishes this through a sophisticated multivariate regression approach that naturally accounts for the correlation structure across multiple data panels. This integrated analysis offers a more comprehensive view than examining each panel in isolation, thereby increasing the likelihood of detecting significantly associated genes.

To extend its applicability and cater to a broader range of research needs, a summary result-based extension known as S-MultiXcan was developed. S-MultiXcan delivers highly concordant results with the individual-level version of MultiXcan, provided that linkage disequilibrium (LD) is accurately matched. This extension ensures that researchers can apply MultiXcan's robust approach even when individual-level data are unavailable, leveraging summary statistics instead.