Correcting for cell-type composition bias in epigenome-wide association studies

Area of Science:

• Genetics and Genomics
• Computational Biology
• Human Disease Etiology

Background:

• Epigenetic variations are increasingly implicated in the development of complex human diseases.
• Analyzing whole blood for epigenetic changes is complicated by variations in cellular composition.
• Distinguishing true epigenetic signals from cell type differences is a major challenge in epigenome-wide association studies (EWAS).

Purpose of the Study:

• To address the confounding issue of cellular composition in whole blood epigenetic studies.
• To explore bioinformatics solutions for accurately identifying epigenetic variations related to disease.
• To overcome the limitations of cell sorting in large-scale EWAS.

Main Methods:

• Review and analysis of bioinformatics methodologies applied to EWAS data.
• Evaluation of computational approaches for correcting or accounting for cell type heterogeneity.
• Focus on strategies applicable to large sample sizes where cell sorting is impractical.

Main Results:

• Two recent publications offer bioinformatics strategies to disentangle epigenetic variation from cellularity.
• These computational methods aim to improve the reliability of EWAS findings in whole blood.
• The identified approaches provide potential solutions for large-scale studies.

Conclusions:

• Bioinformatics offers a viable and scalable solution to the cellular composition confounder in EWAS.
• Accurate identification of disease-related epigenetic marks requires addressing cell type variability.
• Future EWAS in whole blood can benefit from these emerging computational tools for more robust results.