CoMM-S2: a collaborative mixed model using summary statistics in transcriptome-wide association studies.

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) identify genetic variants associated with complex traits but do not fully elucidate the underlying mechanisms.
• Existing methods like the collaborative mixed model (CoMM) integrate GWAS and expression quantitative trait loci (eQTL) data but require individual-level GWAS data, limiting their use with widely available summary statistics.
• There is a need for statistically efficient methods that leverage transcriptome information using only summary statistics from GWAS.

Purpose of the Study:

• To propose CoMM-S2, a novel probabilistic model for investigating the mechanistic role of genetic variants using only GWAS summary statistics.
• To enable the use of readily available GWAS summary statistics for understanding genetic contributions to complex traits.

Main Methods:

• Developed CoMM-S2, a probabilistic model that integrates GWAS summary statistics with gene expression data.
• CoMM-S2 employs a two-model approach: one linking genotype to gene expression, and another linking phenotype to predicted gene expression.
• Evaluated CoMM-S2 performance using simulation studies and real-world data analysis.

Main Results:

• CoMM-S2 effectively analyzes genetic variants using only GWAS summary statistics, eliminating the need for individual-level GWAS data.
• Performance of CoMM-S2 is comparable to CoMM, which requires individual-level GWAS data.
• Demonstrated the utility of CoMM-S2 in real data analyses.

Conclusions:

• CoMM-S2 provides a statistically efficient and powerful approach to explore the mechanistic links between genetic variants and complex traits.
• The model expands the utility of GWAS summary statistics by integrating transcriptome information.
• CoMM-S2 facilitates a deeper understanding of the genetic architecture of complex traits.