The Roadmap Epigenomics Program has published the first comprehensive maps and analyses of human epigenomes across more than 100 tissues and cell types, providing a window into the links between DNA and disease. One study using data from the program found that it's possible to pinpoint where a cancer originated by examining the distribution of mutations along its genome.

Researchers supported by the NIH Common Fund's Roadmap Epigenomics Program have published the first comprehensive maps and analyses of human epigenomes across more than 100 tissues and cell types, providing a trove of data to aid biomedical research into the links between DNA and disease.

The Roadmap Epigenomics Program builds on work being done by the Encyclopedia of DNA Elements (ENCODE) Project. ENCODE researchers are attempting to define specific elements within human DNA that are important for biologic function by leveraging epigenomic signatures, with a large component of the data deriving from malignant or transformed cell lines grown in culture.

In contrast, the Roadmap Epigenomics Program catalogs cell type–specific DNA and histone modifications. Its investigators have focused exclusively on primary cells and tissues, chiefly from healthy people. A description of the maps and 24 papers published in Nature and its related journals demonstrate how the new data can be used to understand cell differentiation and disease development.

“A major accomplishment of this project was developing a public resource for research,” says Lisa Helbling Chadwick, PhD, Roadmap Epigenomics Program team leader and program director at the NIH's National Institute of Environmental Health Sciences. “Just as researchers used the human genome data to understand genetic variation in the context of disease, this project provides a similar resource to help us understand how epigenetic variation might relate to disease.”

In one of the studies published in Nature, researchers used Roadmap data to gain insight into the distribution of genetic mutations that arise during carcinogenesis. They found that the location of mutations closely parallels chromatin features, and surprisingly that the chromatin features of normal cells from which a cancer originates are a much stronger predictor of cancer mutational profiles than the chromatin features of matched cancer cell lines.

The researchers also found it possible to determine where a cancer originated simply by looking at the distribution of mutations along its genome. Such information could be vital to making informed treatment decisions in patients with metastatic disease from an unknown primary site, says study coauthor John Stamatoyannopoulos, MD, associate professor of genome sciences and medicine in the Department of Genome Sciences and Department of Medicine, Division of Oncology, at the University of Washington School of Medicine in Seattle.

“By looking at the pattern of mutations in the cancer we could accurately predict what normal cell type it matched up with,” he says. “Being able to determine where a cancer originated is very important for selecting initial therapy for patients who present with metastatic tumors.”

The findings also have implications for basic research into cancer biology, says Stamatoyannopoulos.

“On a more general level, we now have an unexpected window into the very early events that may be happening in the life of a cancer,” he says. “We are continuing to explore what other information may be hidden in these mutational patterns.”