Structural equation modeling for hypertension and type 2 diabetes based on multiple SNPs and multiple phenotypes.

Area of Science:

• Genetics
• Biostatistics
• Epidemiology

Background:

• Genome-wide association studies (GWAS) identify genetic variants for complex diseases but struggle with intricate genetic and environmental interactions.
• Multiple correlated phenotypes further complicate genotype-phenotype association analyses.

Purpose of the Study:

• To develop a novel structural equation modeling (SEM) framework to systematically analyze genetic associations with multiple complex diseases.
• To incorporate intermediate phenotypes and genetic variants (SNPs) into a unified model for a more comprehensive understanding of disease mechanisms.

Main Methods:

• A four-step SEM process: informative single-nucleotide polymorphism (SNP) selection, latent variable extraction, intermediate phenotype-disease relationship analysis, and SEM construction.
• Application of the SEM method to hypertension and type 2 diabetes (T2D) using GWAS data, with obesity-related traits as intermediate phenotypes.

Main Results:

• Identified 24 informative SNPs and generated 10 latent variables from GWAS data.
• Constructed a quantitative map illustrating the relationships between SNPs, intermediate obesity phenotypes, hypertension, and T2D.
• The final SEM model demonstrated excellent goodness-of-fit.

Conclusions:

• The proposed SEM approach effectively models complex relationships among multiple SNPs, intermediate phenotypes, and correlated diseases.
• This method provides a quantitative map for a deeper understanding of disease etiology and genetic architecture.