Genetics: Scientists seek diversity in genome study subjects

Genomics research, in which researchers scan subjects’ DNA in search of the genetic basis of many diseases, has focused too narrowly on studying subjects of European descent, write a team of genetics experts in the journal Nature this week.

The Los Angeles Times spoke with one of the authors of the piece, Stanford population geneticist and 2010 MacArthur Fellow Carlos D. Bustamante, about why scientists should focus on sequencing genomes of people from other ethnic groups, too.

What got you interested in this problem?

This commentary stems from conversations [coauthors] Esteban [Gonzalez Burchard], Francisco [De La Vega] and I have been having over the past couple of years. How do we think about taking a lot of what’s been in development in European populations and apply it to other populations?

Genome-wide association studies -- when you go out and take individuals with a disease and those without and find genetic changes that predict who is in which group -- have been very successful. The vast majority of those studies have been conducted with subjects of European descent.

Now there’s also a burgeoning explosion of these in East Asian populations, led in part by the BGI (formerly Beijing Genomics Institute) in Shenzhen, China, which we mention in the Nature commentary.

But the rest of the world is being left behind, and we view that as problematic. Something like 96% of the participants in medical genomics studies are of European descent. It’s a hugely lopsided representation.

Here in the U.S. some of the people with the worst health outcomes are members of minority populations. They need to be included in these kind of studies. We don’t think this is just an issue of political correctness -- rather it’s one of addressing widening health disparities. If you don’t include a broad representation of people in this next wave of studies, you’ll only benefit a small set of individuals.

Explain why such research only benefits a small group.

Because we’re moving toward studying rare variants.

So far, genome-wide association studies have looked at common genetic variants that occur in at least 5% of the population. These variants are often shared across populations as well and explain a lot of disease risk. However, they don’t explain most of the genetic underpinning. There’s a lot that’s still left to be explained.

That’s why people are now looking at these rare genetic variants, which occur in smaller numbers of people. But if those rare variant studies don’t sample a broad representation of the population, we’ll be in a situation where we only have information for the people studied. We’ll miss associations that exist in other subgroups.

For a long time, people only included men in medical studies. And that was problematic. You can’t study breast cancer, for example, in men. This is similar.

How is genomics research conducted?

Most of it basically happens through large, organized federally funded studies. You need tens of thousands of people. You’re not going to have tens of thousands of people with a certain disease come into a single hospital.

It is not easy to go out and recruit many, many people. But it can be done. And the best place to do it is in the countries of origin. We argue for a strong engagement between investigators in developing countries and investigators in countries with resources to do these studies.

What needs to happen to expand the scope of this genetic research?

No. 1, we need to make sure that data on underrepresented groups that have already been collected get put into these sequencing pipelines. If there are phenotypes collected of African Americans, Indian Asians, Latinos, etc., they should be prioritized in the next group of studies.

In prospective studies, things that are being designed, we need to make sure diversity is built into the study. If you’re looking at disease associations in Americans, your sample should at least be representative of people in the U.S.

It’s also really important that we create links between the developing world and the U.S. and China and Europe in ways that go beyond, “just give me your samples,” which can make people suspicious. There has to be a good engagement model for that to be successful. You get large numbers of participants by bringing investigators together. How do we get researchers in developing countries into the fold? How do we make them part of the larger project?

I think the philanthropic sector has a role to play.

I think the other part is, we need to explain to the public why this stuff matters. In this era, there’s a real temptation to say, do we really need to invest so much in science and medicine? If this will help us better predict diagnose and treat disease, we’ll be better off.

But ultimately, it’s because that’s going to be one of the main economic engines in the future economy, where we have an aging [population] with exploding healthcare costs. That $1 billion or so that got spent on the human genome got translated to $400 billion in the economy. It’s really clear that investing in science and tech is a sure way to keep your economy going.

But in order for the public to buy in, we need to explain why this matters. We have an obligation and duty to explain.

Read the comment article here (subscription required.)