Naturalgwas: An R package for evaluating genomewide association methods with empirical data

Area of Science:

• Population Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Large-scale genetic association studies face challenges due to geographic variation in genotype frequencies, leading to false positives.
• Existing methods for mitigating these issues lack tools to evaluate the achievable statistical power for specific study designs.
• Population structure and gene-by-environment interactions are significant confounders in genetic association analyses.

Purpose of the Study:

• To present an R program for simulating phenotypes from genotypes to estimate the upper bounds on achievable power for association tests.
• To provide a tool for evaluating the performance of different association testing methods in the presence of population structure and environmental factors.
• To guide researchers in methodological choices for genetic association studies and assess test performance.

Main Methods:

• Development of an R package to simulate phenotypes incorporating population structure and gene-by-environment interactions.
• Implementation of a gold-standard test to evaluate association test power using confounder information.
• Application of the program to simulated data for Arabidopsis thaliana to compare association methods and assess power.

Main Results:

• The developed program successfully simulates phenotypes and estimates power, providing upper bounds for association test performance.
• Simulated data analysis showed that new association methods performed comparably to the gold-standard test in removing confounding factors.
• Statistical power to detect causal variants was influenced by the number of variants and their effect sizes, with specific thresholds identified.

Conclusions:

• The R program offers valuable guidance for selecting appropriate genetic association testing methodologies.
• The tool provides crucial insights into the performance of association tests within user-specific genetic contexts.
• This approach aids in optimizing study design and interpreting results in large-scale genetic association studies.