Scientists have generated the most comprehensive map of the structural variation that exists among normal, healthy humans, according to a study published online today in Nature. Understanding normal variation between individuals is critical to identifying abnormal changes that may contribute to a wide variety of heritable diseases.

Image: Wikimedia commons

"I think it's considered to be a landmark paper," said geneticist Frank Speleman of the Center for Medical Genetics at Ghent University Hospital in Belgium, who was not involved in the work. "It's quite important in the complete context of genome wide association studies and genetic predisposition."

Using microarrays that contained more than 42 million probes, genome scientist Stephen Scherer of The Hospital for Sick Children in Toronto and the University of Toronto and his colleagues searched the genome of 40 healthy individuals for copy number variants (CNVs) -- areas of the genome that come in varying quantities as a result of deletions, insertions, or duplications. The researchers identified 11,700 CNVs 443 base pairs or greater in size, with an average of approximately 1,000 CNVs differing between any two individuals. "[That's] an important amount of normal variation that happens in the genome," Speleman said.

The team then genotyped more than 5,000 of the CNVs from 450 individual genomes pulled from the International HapMap project to determine the population frequency and distribution of these variable regions. This generated a database of normal variation that can be correlated to specific populations and examined for patterns of inheritance among related individuals.

"There is this paradigm shift that CNVs are being incorporated into genetic studies," Scherer said. "And I think it's going to enlighten a lot of our interpretations" of genome wide association studies and other human genetics research.

Scherer and his colleagues generated a similar map in 2006 which, "while comprehensive," Scherer said, "was pretty low resolution." At that time, the researchers could only confidently detect CNVs of 50 kilobases or more. "We had no real idea of what the characteristics looked like," Scherer said. "Some of the CNVs were in fact overlapping with other CNVs and we couldn't really tell what was what." Now, just three years later, they've upped their resolution by two orders of magnitude -- a goal coauthor Matthew Hurles quoted to The Scientist after publishing the first map.

With this higher resolution technology, the researchers believe they have documented about 70% of the common CNVs (those that occur in more than 5% of the population). The CNVs they identified, however, failed to explain the high heritability of many complex diseases. The contributing factors, Scherer suggested, are likely to be the rare CNVs, which are more difficult to identify.

While the results of this study clearly fill in some important detail to the map of human CNVs, there are still many CNVs left to be discovered, the authors admit. "I'm excited by the fact that they've finally released this data," said human geneticist Evan Eichler of the University of Washington. "[But] I think this is far from being comprehensive at this point." The next step, he said, is sequencing individual genomes with long reading frames to further increase the resolution. "You're only as good as you can genotype," he said.

The study "tells us something about the frequency of CNVs and sheds a little bit of light on their role in common disease," said molecular biologist George Zogopoulos of the University of Toronto, who also did not participate in the research. "[Now] we need to look at diseased genomes." Comparing the genomes of individuals affected by certain diseases to healthy controls may identify important CNV-disease associations, he explained.

Despite all that remains to be done, this study -- along with other ongoing efforts to further characterize the human genome -- have added tremendously to our understanding of human genetics, Speleman said. "In the past years, an enormous leap has been taken. We're looking at huge amounts of information, which are generating a lot of new views on the variability of the genome."


Related stories:

Genomic Alterations 2.0
[October 2007]

Tips for CNV Detection
[October 2007]

Structural variations common in human genome
[22nd November 2006]