Rare mutations linked to disease may hide in common variants.
The genetic underpinnings of many diseases remain elusive, and computer simulations and experimental data suggest a reason why: the true culprit [coupable] may be masked by a more obvious suspect.
The findings, by David Goldstein, a geneticist at the Duke Institute for Genome Sciences & Policy in Durham, North Carolina, and his colleagues follow simulated genome-wide association (GWA) studies, in which genome variants among many individuals with and without a specific disease are compared and linked to the risk of developing the disorder.
The GWA technique is based, in part, on the assumption that diseases that are common in the population — such as diabetes and heart disease — are due to genetic factors that are also common. However, Goldstein speculated that rare mutations that are too low in frequency to be uncovered [à découvert] by GWA studies might be hiding within the common variants and throwing off the signal. His group's simulations back up this possibility. The results are published in PLoS Biology1.
Leonid Kruglyak, a geneticist at Princeton University in New Jersey, says the work confirms what many had suspected. "I think everyone who thought hard about genome-wide association studies knew that this was a possibility," he says.
The work addresses a concern that has been brewing in the genetics community for several years. More than 2,000 common variants have been linked to common diseases through GWA studies, but explain only a tiny fraction of the estimated heritability of these diseases.
Researchers have speculated as to the source of this 'missing heritability' (see 'The case of the missing heritibility'). One explanation is that a rare variant with a strong role in disease arises more frequently in people who share a common variant.
Goldstein and his colleagues looked at mock populations [populations modèles] of thousands of people. Some individuals had rare variants — mutations that occurred in 0.5–2% of the population — that were very likely to cause an unspecified disease. The authors simulated the genealogy of this population and ran GWA studies. They found common genetic variants that associated with the disease, but only a proportion of individuals with the common variant also had the rare variant that theoretically caused it. So, the association with the common variant was indirect. Goldstein calls it a "synthetic association".
Synthetic associations make it seem as if many people share a genetic sequence that confers some small risk of disease when, in fact, a few of those individuals have a rare variant that confers a much higher risk. This can lead to false assumptions [hypothèses] about the common variant, says Goldstein, it could also mean that rare variants account for much of the missing heritability of disease.
Geneticists knew that many common variants picked out by GWA studies would be proxies — indicators of a genetic cause that might exist near the common variant or somewhere else in the genome — and, indeed, many of the associations made so far don't seem to have an explanation. Synthetic associations could be one factor at play. Goldstein speculates that, "a lot, and possibly the majority [of these unexplained associations], are due to, or at least contributed to, by this effect".
Goldstein and his colleagues also did a real-world experiment as a proof of the principle. They compared individuals with diseases caused by a single-gene mutation, such as sickle-cell anaemia or a genetic form of deafness [surdité], with a control population. Sure enough, in GWA studies on these populations, common variants seemed to associate with the disease, although in some cases these were as far as 2.5 million base pairs away from the known causative mutation.
Sarah Tishkoff, a geneticist at the University of Pennsylvania in Philadelphia, says "this may make it challenging to identify the functional variant within an association". Teri Manolio, a population geneticist at the National Human Genome Research Institute in Bethesda, Maryland, writes via email that "if their simulations are correct, and I suspect they are," they suggest researchers will have to sequence a lot of DNA, up to 10 million bases, surrounding common variants.
Goldstein says that the work suggests more whole-genome sequencing will be needed in more targeted populations of affected individuals and families. In a sense, he says the issues being raised signal the need for shift from the powerful statistics of GWA studies to work more focused on specific genes in affected families and how they function biologically. As Goldstein puts it, "the importance of the family has really come back again."
http://www.nature.com/news/2010/100126/full/news.2010.33.htm
/https%3A%2F%2Fassets.over-blog.com%2Ft%2Fcedistic%2Fcamera.png)