This study will employ a systematic approach to investigate the effects of childhood glycemic profile on the metabolome and the methylome. We propose that blood glucose level is linked to changes in the metabolome, which results in variation in DNA methylation between individuals. Using Mendelian randomization techniques it will possible to investigate the directionality of this relationship. It is possible that there is a causal relationship between blood glucose level as an environmental exposure due to diet and altered DNA methylation which may alter gene expression.

This systematic approach requires several steps. Firstly, using existing GWAS results, a genetic proxy for blood glucose levels will be used to perform a Mendelian randomisation analysis investigating the effect of blood glucose on the metabolome treating blood glucose as an exposure and metabolome as an outcome. GWAS results will then be used again to find a genetic proxy for the observed metabolome effects, thus obtaining a filtered list of SNPs that can be used in a second MR step to identify differentially methylated CpGs that are associated with changes in the metabolome. It would then be possible to investigate the transcriptional effects of these modifications, linking exposures to molecular mediators and phenotypic outcomes.

This systematic method has is advantageous over traditional EWAS due to the stepwise filtering of SNPs associated with the exposure of interest, and may be more likely to provide biologically interesting data relating to disease phenotypes. ALSPAC provides an ideal opportunity to test the utility of this approach, due to the availability of genotype, NMR metabolomics, DNA methylation and some transcriptional data from a large cohort.