Some SNPs could influence a phenotype by altering gene regulation in cell types relevant to that phenotype. Starting from this hypothesis, the authors of this study used statistical approaches to show that some cell-type-specific chromatin modifications can be used for mapping phenotype-associated SNPs to regulatory variants. They found that chromatin marks linked to gene activation — particularly histone H3 lysine 4 trimethylation — overlap approximately one-quarter of trait-associated variants in relevant cell types (such as the liver and pancreatic islet cells for type 2 diabetes). This approach could provide leads for functional follow-up.