Accounting for population stratification in DNA methylation studies

Area of Science:

• Epigenetics
• Genomic Epidemiology
• Computational Biology

Background:

• DNA methylation is a key epigenetic mechanism linking genome, environment, and disease.
• Population stratification is a significant confounder in genetic studies, often overlooked in DNA methylation research.
• Failure to adjust for population stratification can lead to false positive findings in association studies.

Purpose of the Study:

• To propose and evaluate methods for correcting population stratification in DNA methylation studies.
• To demonstrate the confounding effect of population stratification on DNA methylation using race as a proxy.
• To compare the performance of principal component-based approaches with other existing methods.

Main Methods:

• Utilized principal components (PCs) derived from genome-wide methylation data to adjust for population stratification.
• Illustrated confounding by assessing DNA methylation associations with race in a cohort of 388 individuals.
• Evaluated PC-based methods against surrogate variable analysis and genomic control through simulations.

Main Results:

• All tested methods effectively removed inflation caused by population stratification.
• Single-nucleotide polymorphism (SNP)-based PCs offered maximum power, followed by methylation-based PCs.
• Methylation-based PCs, particularly those using CpG sites proxying SNPs, proved powerful and efficient, especially when SNP data is absent.

Conclusions:

• Principal component analysis is a viable strategy for correcting population stratification in DNA methylation studies.
• Methylation-based PCs offer a robust and computationally efficient alternative for adjusting population stratification.
• These methods are crucial for enhancing the reliability of DNA methylation association studies and understanding disease etiology.