Analysis of blood-based DNA methylation scores reveals associations with shared cohabitation duration in cohabiting pairs

Individuals who share their environment for long periods of time, that is, spouses, are likely to share behaviors and exposures, and longer spouse-pair cohabitation durations have been suggested to increase methylation pattern similarity on a methylome-wide scale. Risk scores and aging clocks based on blood DNA methylation use subsets of the methylome that share functional associations with disease and phenotype outcomes to estimate exposure and disease-specific epigenetic effects. By using the CpG sites contained within methylation risk scores and aging clocks, it is possible to examine functionally-linked methylation patterns of long-term shared exposure effects. In a population cohort of paired adults, we observed that methylation disease and phenotype risk scores show no significant correlation between partners, and neither do aging clocks after adjustment for age, which is already correlated in pairs. We also did not find evidence of an absolute difference in scores being significantly associated with cohabitation duration. However, when we examined the correlation in normalized methylation values between partners, we observed that correlations were higher in pairs with longer cohabitation durations for many, but not all, functionally-linked CpG sets. This supports the hypothesis that shared cohabitation's association with methylome-wide similarity in pairs is the result of many shared exposures, not a generic effect of cohabitation. Further research is required to establish the precise exposure-methylation links implicated.