Published online 2 February 2010 | Nature 463, 596-597 (2010) | doi:10.1038/463596b


Project set to map marks on genome

Consortium sets sights on the differences that make us different.


Click for larger image.M. Green & S. Forsburg, Univ. Southern California

At one time, sequencing the whole human genome seemed almost impossible. But even as it was being completed, biologists were realizing that the genes encoded within it would reveal little about what makes each of us unique.

By 2003, it was clear that the true key to understanding our differences was more likely to be found in epigenomics — the reversible chemical modifications to DNA and its associated proteins that determine when genes can be expressed. And epigenomic studies could open up new research avenues in cancer and other complex diseases.

So it was with some fanfare that top biologists launched the International Human Epigenome Consortium (IHEC) in Paris last week, a global consortium that aims, as its first phase, to map 1,000 reference epigenomes within a decade.

"Epigenomes are changeable, programmable and will feed us the bottom line on how the genome works," says Rob Martienssen of Cold Spring Harbor Laboratory in New York, a member of the consortium's steering committee. "IHEC is an exciting prospect."

IHEC is now recruiting members among funding agencies and other organizations to raise US$130 million for its first phase. Executive-committee members must commit $10 million over five years, and participants must agree to make their data freely available.

The initiative was spearheaded by the US National Institutes of Health, which launched its $190-million, five-year Roadmap Epigenomics Program in 2008, and the European Commission, which in July will publish a call for proposals for a €30-million (US$42-million) epigenetics consortium, including a large component for reference epigenomes. Representatives from Canada, France, Germany and several Asian countries voiced interest at the launch in joining IHEC.

The consortium faces a much greater challenge than did the Human Genome Project when it was launched in 1990. "The human genome is singular and finite, but the human epigenome is almost infinite — the epigenome changes in different states and different tissues," says Philip Avner of the Pasteur Institute in Paris, a member of IHEC's steering committee.

Epigenomics is indeed a complex business. The human genome comprises strings of coiled DNA supercoiled around histone proteins into dense material called chromatin. Genes can be transcribed only when the chromatin is opened up to expose them, and this process depends heavily on chemical marks — the epigenome. Methyl groups, small non-coding RNAs and other small molecules can all serve to mark the DNA and the chromatin.

Whereas the hard-wired genome is virtually identical in all of a person's roughly 250 different types of tissue and is essentially stable during that person's lifetime, the epigenome changes during development such that each cell type has its own characteristic set of marks. These marks change with age and may also change, possibly in a heritable way, in response to environmental stress. Some diseases, such as cancer, may be caused by harmful changes to the epigenome, or changes in the epigenome may be a consequence of disease.

The normal variation in the many epigenomes a single person will have in a lifetime, or even during the circadian day, is unknown, and technologies to detect the diverse types of epigenomic mark are still developing. Some scientists at the meeting were therefore concerned that IHEC might be premature.

"But this is exactly the point of launching IHEC now — to avoid confusion and move the field forward more quickly," says Peter Jones, director of the Norris Comprehensive Cancer Center at the University of Southern California in Los Angeles.

The number of publications on epigenomics is growing exponentially, he says, which is why researchers need to agree now on how to standardize data. "We need a set of high-quality reference epigenomes that everyone can trust, we need to standardize technological approaches and we need to avoid unnecessary duplication," he told the meeting.

Already, the costs of some epigenome technologies are falling rapidly. The first genome-wide map of DNA methylation, or methylome, was published last November (R. Lister et al. Nature 462, 315–322; 2009) at a cost of $100,000, according to lead author Joseph Ecker of the Salk Institute for Biological Studies in La Jolla, California. "I expect the cost of a similar methylome will fall to $10,000 in the next six months," he says. So IHEC's target budget of $130 million could allow more than 1,000 epigenomes to be mapped.


At the meeting, project leaders decided that the vast majority of reference epigenomes should be from normal human tissue. "When we have all the references, we'll be able to compare diseased tissue against them," says Avner. The new consortium may ally itself with large genomics consortia such as the International Cancer Genome Consortium to share samples.

IHEC plans to have a final policy document by April, and nominations for the executive and scientific boards by June. The boards will choose the reference epigenomes and decide how to implement open-access policies. They will also ensure quality control and a realistic level of bioinformatics support, given that the scale of the data will dwarf that of the human genome. 

See Editorial, page 587.

Commenting is now closed.