Big Science had a busy time this past month, with one large-scale project, the Cancer Genome Atlas, announced on Dec. 13 and another, the Human Epigenome Project, called for by cancer researchers Dec. 15.

“It's amazing how the timing worked out,“ Peter Jones, professor at the University of Southern California, told Medical Device Daily's sister publication, BioWorld Today. While the paper calling for a Human Epigenome Project was published in the Dec. 15 issue of Cancer Research, and its authors make much of epigenetic contributions to cancer (the article noted that at least 50% of changes seen in cancers are epigenetic), the scope of an epigenome project would not be restricted to cancer.

“We need to know what the normal human epigenome looks like,“ Jones said. Different labs are gaining a good understanding of isolated parts of the epigenetic puzzle, but only a large-scale effort will enable a more comprehensive understanding. “If I just focus on cancer and Joe down the road just focuses on globin genes, we're never going to understand how the whole thing works,“ he said.

The study of epigenetics — heritable variations in gene expression that are not encoded in DNA sequence, but instead in structural modifications of DNA such as methylation and acetylation — also is supposed to play a role in the Human Cancer Genome Atlas, a joint effort to understand cancer genomes announced Dec. 13 by the National Cancer Institute and the National Human Genome Research Institute, both member institutes of the National Institutes of Health (Bethesda, Maryland).

For now the atlas is a pilot project, but the two sponsoring institutions seem to pin high hopes on it.

At a press conference announcing the atlas, the word “revolutionary“ was bandied about liberally by several of the speakers. Francis Collins, MD, PhD, director of the National Human Genome Research Institute and a veteran of the Human Genome Project, put the atlas in the tradition of the Human Genome Project and the HapMap Project. But, he told reporters, “in many ways, it's like doing thousands of genome projects.“

The reason, said NIH Director Elias Zerhouni, is that “cancer is different, because you don't inherit the genetic mutations that lead to cancer.“ Instead, mutations acquired over a lifetime of cell divisions and environmental mutagens are what most often leads to cancer. And by their nature, such mutations are more varied than inherited ones.

The scientists hope that new technologies can be developed that will help them deal with the massive quantities of information. While epigenomic technologies are among those mentioned in the announcement for the Cancer Genome Atlas, at the press conference, participants seemed somewhat skeptical that most epigenomic methods are currently mature enough to be used on a genome-wide scale.

Collins said that the question was whether epigenomics can be robust enough to drive sequencing. “But,“ he added, “if investigators can convince peer reviewers of that, then epigenomics will be a big part of this.“

Unsurprisingly, the epigenetics crowd takes a more sanguine view of the capabilities of their technologies. Nathan Lakey is CEO of Orion Genomics (St. Louis), which uses epigenetic methods for cancer diagnostics, among other things. He said that both genomic and epigenomic approaches are important in the quest to understand cancer.

“It certainly makes sense to do the sequencing, especially because that infrastructure already exists,“ he told BioWorld Today. But at the same time, “everybody knows epigenetics plays a huge role“ in cancer. Jones echoed that sentiment.

“I'm thrilled that there is an epigenetics component to the atlas, but worried that its importance is sometimes overlooked,“ he said.

Asked about whether epigenetic technologies are high-throughput enough to make a genome-wide approach feasible, Lakey replied that while such technologies are “less mature“ than sequencing technologies overall, there are some approaches that have potential for use in genome-wide studies. While base-by-base studies are not a realistic option at this point, “regional-based methylation studies can be very high-throughput,“ Lakey said. Likewise, “bisulfite based-methods are tedious, but enzyme and array-based methods also can be very high-throughput.“

Jones concurred, pointing in particular to tiling arrays and chromatin immunoprecipitation as high-throughput epigenetic technologies. He also added that critical epigenetic mutations are likely to be restricted to the start sites of genes, and that gives scientists an idea of which DNA regions to look at.

“There are problems — it's not as easy as sequencing — but there are also solutions,“ Jones said.

Besides, he added, when the human genome project started, sequencing also was nowhere near as high-throughput as it is today. “If you look back 15 years ago, everybody said you can't sequence that much DNA,“ he said. “The reason sequencing technologies evolved was because there was a need for them.“