Uncommon Carriers

If variety lends life flavor, then humans are kicking things up to a previously unrecognized notch on the spice-o-meter.

New efforts to decipher the genetic blueprints of thousands of people have turned up more than half a million tweaks in human DNA, many more than scientists expected. Most of these tweaks are new to science, and a majority fall into a class called “rare variants,” found in 0.5 percent of the population or less. Some of the variety recently uncovered is so uncommon that it shows up in people living in a single geographic region, or even in only one person.

Despite their limited spread, the newly discovered rare variants could profoundly affect susceptibility to disease or how well drugs work. They may also help researchers reconstruct recent human migrations around the world.

For years, scientists have been examining the chemical units of DNA called nucleotides that act as letters in the human genetic instruction book. So researchers thought they had a good handle on how often to expect single-letter changes in the A’s, G’s, T’s and C’s in that book. Such changes stem from errors in copying and are spotted via comparison with some majority-rule blueprint. They can go by terms like “single nucleotide polymorphisms” or “mutations” depending on where and when they show up.

When looking at 202 genes predicted to be important in diseases from 14,002 people, John Novembre of the University of California, Los Angeles and colleagues unearthed five times as many rare genetic variants as expected.

Based on what’s known about human diversity, the researchers thought they would find one letter change for every 90 nucleotides. Instead, the team reported in the July 6 Science, a variant showed up for every 17 nucleotides. Not only did these variants include more rare ones than expected, but also more really rare ones. About 74 percent of the rare variants are practically secret family recipes. Others reveal the distinct flavor of geographic regions, much like wines or cheeses.

Figuring out how these exotic variants contribute to disease will probably be more challenging than it has been for widespread tweaks. Currently researchers rely on statistics to sort out links between common variants and disease, but rare variants won’t make the cut in these types of analyses. Instead of linking a particular DNA change to a disease, scientists will probably have to pinpoint where variants strike to see if certain genes are hit over and over again.

“The hope is that even though the variants themselves will be unique to each population, the underlying genes … will be the same,” says Jeffrey Kidd, a human population geneticist at the University of Michigan in Ann Arbor.

Probing the rare

Combing the roughly 3 billion DNA units that make up each person’s genetic blueprint is a daunting task. Novembre’s team focused on just a tiny portion: 864,000 DNA units containing 202 genes. Another team took a broader look, concentrating on about 60 million units containing 15,585 genes. The project was ambitious, reading each of those genes for thousands of people 100-plus times to be sure every letter was correctly identified.

“This is, I think the technical term is, a ginormous project,” says study coleader Joshua Akey, a population geneticist at the University of Washington in Seattle.

People in the study carried an average of 13,595 single DNA nucleotide variants, with about six rare variants for every common one, Akey and colleagues reported in the July 6 Science. Rare variants were about four times as numerous as scientists expected.

A typical participant had more than 300 genes carrying variants that impair the function of the protein produced by that gene. In almost all those cases, the culprit was a rare variant. Rare variants probably tend to be the bad ones because natural selection weeds out harmful alterations before they can become common, Akey says. Rare, adverse variants found in the study probably arose recently, giving evolution little chance to counteract them.

Although many of these rare variants are predicted to be harmful, their youth makes them a treasure trove for scientists reconstructing how humans have migrated around the world. Population geneticists have used common genetic variants to decipher human movement before, but most of that data represents ancient history. A variant common enough to appear in 5 percent or more of the population is probably more than 10,000 years old, says Alon Keinan, a computational biologist and human population geneticist at Cornell University.

But rare variants are younger, with new ones popping up in every generation, so they can tell geneticists what has happened in the last several hundred to several thousand years. That’s a time period scientists know little about, genetically speaking.

Tracing the past

Archaeological finds and, more recently, census data reveal that the human population has been booming, expanding from a few million people 10,000 years ago to more than 7 billion today. But, “the jury is still somewhat out on the exact signature history has left on our genetic variation,” Keinan says.

Previous studies had suggested that populations grew in size but not in genetic diversity, with variants squeezed out by natural selection and population bottlenecks as humans colonized the world. Data compiled from a small number of people seemed to indicate that human genetic diversity had remained fairly stagnant for the last 30,000 years or longer. So genetic records could outline basic migration paths showing that humans moved from point A to B, but wouldn’t record all the interesting byways and side trips along the way.

But it now looks like those studies underestimated the number of rare variants in the human population. With colleague Andrew Clark, Keinan reported in the May 11 Science that rapid population growth has produced an abundance of rare genetic variants in humans.

Studies of thousands of people reveal that a measure of genetic diversity that includes new variants has grown at a rate of 5 to 14 percent per generation for at least the last 900 to 2,800 years. Armed with the new information, Keinan and Clark simulated different scenarios for the colonization of Europe. Simulations that included bottlenecks and population growth mirrored patterns of rare variants seen in modern Europeans better than did scenarios that neglected population expansion.

The pair’s measurements of genetic diversity offer a broad view, mostly of Europeans. Novembre advocates drilling deeper to look at rare variants regionally or in certain ethnic groups.

Populations sampled so far (all three recent studies focused mainly on Europeans and African-Americans) have had different patterns of diversity, shaking up population geneticists’ assumptions that one measure of diversity should fit all human groups. “There isn’t one number you can plug in and predict how much diversity we’ll see,” Novembre says.

As scientists begin studying smaller and smaller groups of people, they may discover yet different levels and patterns of variation. In fact, researchers may need to examine every single person on the planet to find all the diversity.

Previous studies that didn’t consider population growth predicted that scientists would find all the new variants they could expect to uncover after examining the genetic instruction books of 1,000 people. One more genome wouldn’t add much to the mix.

But by Keinan and Clark’s calculations, which take into account variants introduced as a population grows, a 1,001st genome would yield at least a thousand new variants, making the flavor of human diversity ever more complex.