From Methylation to Metabolism: Linking Molecular Layers in Diabetes Risk

Type 2 diabetes risk has traditionally been assessed using fasting glucose, BMI, and lipid panels (1). Although these measures are easily accessible and scalable, they capture disease only after metabolic dysfunction has already been established, limiting the potential for prevention and hindering the effectiveness of interventions (2). This gap has motivated efforts to develop molecular biomarkers to enable detection of earlier stages of risk and refine stratification through approaches such as epigenomic, proteomic, and metabolomic profiling (3–5). While these “single-layer” analyses have each yielded important insights, such as DNA methylation sites associated with insulin resistance (6), circulating proteins linked to diabetes incidence (7), or amino acid signatures that accompany early metabolic dysfunction (8), these approaches are limited in bridging specific outcomes with additional layers of biological information. Further complicating discovery of robust biomarkers is the heterogeneity of plasma measures due to environmental and genetic factors.