Modeling the complex gene x environment interplay in the simulated rheumatoid arthritis GAW15 data using latent...

Area of Science:

• Genetics
• Rheumatology
• Statistical Modeling

Background:

• Rheumatoid arthritis (RA) is a complex autoimmune disease influenced by genetic and environmental factors.
• Current genetic association studies often use single nucleotide polymorphism (SNP)-at-a-time approaches, which can be limited by statistical challenges like multicollinearity.

Purpose of the Study:

• To develop and validate a novel latent gene construct approach using structural equation modeling for analyzing complex diseases like RA.
• To model individual genes as latent variables defined by multiple SNPs to improve genetic analysis.

Main Methods:

• Employed structural equation modeling (SEM) framework on simulated rheumatoid arthritis data from the Genetic Analysis Workshop 15.
• Utilized a latent gene construct approach, defining genes as latent variables based on dense and non-dense single-nucleotide polymorphisms (SNPs).
• Integrated genes, sex, and smoking as predictors within the SEM framework to model rheumatoid arthritis.

Main Results:

• The latent gene construct approach produced valid gene constructs, especially with dense SNPs.
• The models incorporating latent gene constructs and other factors showed good fit according to certain indices.
• Gene F, C, DR, sex, and smoking were identified as significant predictors of rheumatoid arthritis, while Genes A and E were not.

Conclusions:

• The novel latent gene construct approach shows promise for unraveling complex diseases by reducing statistical tests and mitigating multicollinearity.
• This method offers enhanced control of confounding, potentially leading to less biased effect estimates for gene-disease associations.
• Further research is needed to refine the approach, particularly in quantifying bias, evaluating model fit, and resolving complex gene interactions.