Framework for identifying quantitative trait loci in association studies using structural equation modeling

Area of Science:

• Genetics
• Statistical genetics
• Population genetics

Background:

• Identifying quantitative trait loci (QTL) is crucial for understanding genetic contributions to complex traits.
• Association studies are powerful tools for QTL detection, but can be confounded by population structure.
• Structural equation modeling offers a flexible framework for genetic analyses.

Purpose of the Study:

• To propose a novel framework for QTL detection in association studies using structural equation modeling.
• To develop and evaluate two distinct statistical tests for identifying QTLs and estimating their explained variance.
• To assess the robustness of the proposed methods against population admixture and compare their statistical power.

Main Methods:

• Development of a structural equation modeling framework for QTL analysis.
• Introduction of two tests: one assuming no admixture, and a Transmission Disequilibrium Test (TDT)-like test robust to admixture.
• Power calculations and simulations to evaluate test performance and sample size requirements.

Main Results:

• The first test, assuming no admixture, requires fewer subjects (e.g., 100 for 10% variance explained) compared to the TDT-like test (1.7x larger sample size for equivalent power).
• The first test demonstrated robustness against population admixture, with good power to detect admixture when present.
• The TDT-like test effectively controls for false positives arising from population admixture.

Conclusions:

• The proposed structural equation modeling framework provides effective methods for QTL detection in association studies.
• The developed tests offer flexibility in handling population structure, with options for different genotyping requirements.
• The findings suggest that admixture can often be detected and managed, mitigating potential false-positive findings in genetic association studies.