Cell-type heterogeneity can be a major source of confounding and reverse causation in epigenome-wide association studies (EWAS). Adjustment for cell-type composition is therefore critical for an ...