Statistical challenges in the analysis of blood-derived DNA methylation data from environmental epidemiology studies

Abstract

Title: Statistical challenges in the analysis of blood-derived DNA methylation data from environmental epidemiology studiesAbstract: One of the foremost challenges in the analysis and interpretation of epigenomic data is the issue of cellular heterogeneity and its potential confounding effects on studies that aim to understand the relationship between DNA methylation and some phenotype or exposure. This is especially true of whole blood, which represents one of the most commonly used tissue types for environmental epigenomic studies due to its relatively easy accessibility, and because of a heightened interest in the immonomethylomic effects of environmental insults. Whole blood is a heterogenous and complex mixture of many different cell types each with distinct functions and roles. The cellular landscape of whole blood may be sensitive to certain environmental insults and such insults may impact patterns of DNA methylation in a cell-specific manner. Thus, in addition to the well-recognized potential for confounding, environmental exposures may elicit cell-specific effects, the detection of which represents a major challenge in the analysis whole-blood methylation signatures. The identification of cell-specific effects of methylation based on bulk-tissue methylation signatures is an ongoing area of research with important implications for environmental epidemiology studies of DNA methylation. This talk will orient the audience to the above challenges and will include a discussion of recent work that aims to address these challenges. Insight into the application of such methods for environmental epidemiology studies of DNA methylation will be given.