A consortium of researchers in Europe is spearheading an initiative to sequence the genomes of 1,000 Europeans, with the goal of finding genetic links to 10 diseases.
The consortium, which will require additional sequencing capacity to be established in Europe, comprises researchers from the Center for Genomic Regulation in Barcelona, Spain; the Wellcome Trust Sanger Institute; the University of Geneva; and the University of Nijmegen in the Netherlands. Last week it met in Barcelona during the 2008 European Society of Human Genetics conference to discuss the project.
Called Investigating European Profiles of Structural and Sequence Variation of the Human Genome in Disease, or EUVADIS, the consortium initially plans to generate whole-genome sequence data for 1,000 Europeans diagnosed with common disorders — 100 for each of 10 diseases.
This is also the principal way in which the initiative differs from the recently launched international 1,000 Genomes Project (see In Sequence 1/22/2008), which focuses on discovering low-level sequence variation in healthy individuals.
“This is, if you want, sort of a follow-up of what has been initiated with the whole-genome-association scans that have been performed, like the Wellcome Trust Case Consortium and several others,” Xavier Estivill, director of the genes and disease program at the Center for Genomic Regulation in Barcelona and coordinator of EUVADIS, told In Sequence last week.
Though these GWA studies have revealed some of the genetic variability linked to a number of diseases, they remain incomprehensive. “If we want to extract the complete information that may be responsible for the genetic basis of common disorders, we will not do it well with just targeting the variants that are in HapMap,” he said. “The next step is … to go to the detailed analysis of the sequence.”
He said EUVADIS will take advantage of “lots of good clinical data” that has been collected at various centers throughout Europe. The vast majority of samples studied in GWA scans performed by the Wellcome Trust Case Consortium and others, are from European populations, he added.
According to an abstract for an upcoming EUVADIS workshop in September at a conference in Strasbourg, France, that is sponsored by the European Science Foundation, the European population “has a long history, is extremely rich in genetic variability, and has left many marks in its diverse geographic regions, which have materialized in different disease prevalences and phenotypes.”
EUVADIS aims to integrate population and geographic data; analyze epidemiological aspects of disease and common traits; use high-throughput technologies to sequence 1,000 subjects for 10 common human disorders; evaluate phenotypic consequences of structural and sequence changes of the genome; and integrate computational biology data of medical conditions for prevention, diagnosis, and treatment of common diseases, according to the abstract.
Initially, the study will only involve patient samples but no separate controls; samples from different disease cohorts and those sequenced in the 1,000 Genomes will serve as controls.
The current estimated cost of sequencing a human genome with “reasonably good coverage” to be able to call low-level variants and structural variants is €100,000, according to Estivill, while sequencing 1,000 samples will cost €80 million to €100 million, which does not include the analysis.
However, since sequencing costs are “dropping significantly and rapidly,” he said, this price tag is likely to get smaller — a prediction that could help efforts by the unfunded consortium to raise capital.
The project has not secured funding yet, but during the Strasbourg conference, EUVADIS organizers plan to pass the hat to possible funding agencies, charities, and interested academic and industry groups.
Funding might come, for example, from the European Commission’s 7th Framework Program, which has a call for applications for related projects next year.
That conference will be followed by a meeting early next year, after which EUVADIS organizers hope to secure at least part of the funding that will enable the at least three-year effort to begin in late 2009.
“The next step is … to go to the detailed analysis of the sequence.”
The project is open to more participants, Estivill said, and the founding members have spoken with interested research groups in France and Germany. In principle, countries outside Europe could join with their own disease-centered projects, he added.
Similar to how the International Cancer Genomics Consortium is organized (see In Sequence 4/29/2008), each EUVADIS group or country may devote its efforts to a specific disease, according to Estivill.
The consortium has not yet decided which 10 diseases to focus on, but they are “likely to include multiple sclerosis, diabetes, and depression,” according to a preliminary program of the Strasbourg workshop.
EUVADIS would require additional sequencing capacity to be established in Europe, Estivill said, which is currently dominated by two institutions: the Wellcome Trust Sanger Institute and Genopole, the French National Sequencing Center.
The Sanger Institute, for example, “with the commitment [it] already has with different projects they are involved in,” will likely only be able to provide a fraction of the sequencing for EUVADIS. “It’s clear that they will need additional machines, or we will need other groups that are participating in this analysis,” he said.
But according to Estivill, by early 2009 it is “likely that there will be at least a couple of other centers that will have a capacity of maybe 15 or 20” Illumina Genome Analyzers or Applied Biosystems SOLiD instruments.
One of these centers is expected to be in Spain and funded by the Spanish government, although it has not been decided yet where it will be located, Estivill said. The Catalan Autonomous government is favoring Barcelona because “the vast majority of the scientific contributions to genome projects [in Spain] have been produced by investigators from the region,” he added.
Estivill’s institution, the Center for Genomic Regulation, which is part of the Biomedical Research Park of Barcelona, currently owns one Illumina Analyzer and one 454 Genome Sequencer and plans to buy a second Illumina instrument.
Companies developing new sequencing technologies could also play a role in the project, according to Estivill. “In Europe, as far as I know, there is no single Helicos instrument,” he said. “That could be a very good project to test how well the system is doing, and we could establish a partnership.”
Some researchers who are not part of EUVADIS caution that such a project poses a number of challenges.
For example, before the project even starts its participants must sufficiently understand the new sequencing technologies’ performance and cost, and the systems must be accurate enough to recognize rare mutations.
“To obtain high sensitivity and specificity for rare mutations will require accuracy beyond that of what has traditionally been known as finished genome sequence,” David Altshuler, a professor of genetics and medicine at Harvard Medical School, told In Sequence by e-mail last week. “It is not clear to anyone in the field what coverage will be needed to obtain this, but based on current knowledge, it will be substantial.”
He also pointed out that the sample size of a study like EUVADIS needs to be “very large to make sense of any individual gene, as many genes will be involved, there is a genome-wide significance threshold to be overcome, and the mutations will not always be obvious.” According to most estimates, he said, “many thousands” of samples would be needed, similar to genome-wide association studies.
Researchers also still need to develop analytical methods to relate rare mutations and disease risk, according to Altshuler. “Where the diseases are Mendelian in nature and the mutations obvious in effect, this is easy,” he said. “But that will not be the general case, and much work is needed on analysis.”
In addition, the “nature of the selected phenotypes” will be critical, according to Alan Michelson, associate director for basic research at the National Heart, Lung, and Blood Institute. Whether the study will include quantitative traits across the entire spectrum or only extreme traits, for example, will also influence the power of the study, he said.
Michelson added that analyzing and managing “massive amounts of data” generated in such a large-scale sequencing project is another challenge, one that the 1,000 Genomes Project is currently grappling with (see In Sequence 5/13/2008).
It will probably be several years before researchers find out what can be learned about disease from whole-genome sequencing studies that cannot be gained by other means, such as studying Mendelian diseases, common variants in genome-wide associations, or by candidate gene sequencing, according to Altshuler.
“It will be worth it — but there is a long way from proposing such studies to having done them well enough to learn general lessons,” he said.