The epigenome, a collection of chemical modifications that alter the way genetic information is used in different cells, has important roles in normal development and disease. A collection of papers from the Roadmap Epigenomics Program, published in Nature this week, charts these modifications in a number of human cell types. A better understanding of the epigenome offers a framework for studying human disease and may assist in the design of new treatments.

Although all cells in the body have the same DNA, they have different characteristics because they express different sets of genes. Which genes are expressed is in large part dictated by epigenetics. This is why, for example, heart cells look different and function in different ways from brain cells.

To explore how cells achieve their specific identities and functions, the Roadmap Epigenomics Program analysed samples taken directly from human cells and tissue - embryonic and adult. Researchers produced maps detailing how the epigenome varies and operates in different settings. In one paper, which integrates 111 reference human epigenomes generated by the Roadmap Epigenomics Program, Manolis Kellis and co-authors highlight similarities and differences in a range of different tissue and cell types, and provide a resource for interpreting the molecular basis of human disease, by revealing the epigenetic signatures of genetic variants associated with 58 complex traits.

Alexander Meissner and colleagues show that DNA methylation - the addition of chemical groups to the genome that sets the stage for genes being turned on or off - affects how stem cells differentiate into specialized neural cells. Bing Ren and colleagues show that another form of epigenomic regulation - chromatin modification, which alters the three-dimensional configuration of DNA and affects gene expression - also has a general role in stem-cell differentiation. Shamil Sunyaev and colleagues demonstrate that epigenomic profiles are linked to mutation patterns associated with cancer, and that this information can be used to predict the cell of origin of an individual tumour. Manolis Kellis and co-authors explore the epigenetic profile of Alzheimer's disease in a mouse model of neurodegeneration similar to Alzheimer's disease, and show that genetic predisposition to Alzheimer's disease may be primarily associated with immune functions, while changes in neuronal activity associated with memory and learning may be affected primarily by non-genetic factors. They further show that these changes are conserved between the mouse model and human Alzheimer's disease.

Just as the Human Genome Project provided a map of the genes of the human genome, the Roadmap Epigenomics Program offers a resource for understanding how our genetic blueprint is interpreted in different cell and tissue types. The next step will be to map the epigenetic profiles of individuals to understand more about how they vary from person to person and to establish causality between any of these epigenomic marks and disease.

The research will be freely available to all, and the papers from Nature, along with papers from Nature Communications, Nature Biotechnology, Nature Methods, Nature Neuroscience, Nature Immunology and Nature Protocols, can be explored on the Epigenome Roadmap Site at