Some aspects of DNA research are decidedly modern: think genetic engineering or genetic testing. Another breed of DNA researchers, though, aim to blend new and old. By harnessing the power of modern sequencing and analytical techniques, they are working to uncover the genetic secrets hidden in the DNA of ancient specimens collected during archaeological expeditions or curated in museums. These ancient DNA researchers are analyzing preserved DNA from ancient hominids, plants, and animals to learn about their evolutionary histories, offering a level of detail that inferences based on modern DNA simply cannot match.
But despite the successes, some of which are described in this week’sspecial issue, ancient DNA research is far from routine. The samples are frequently degraded and prone to contamination by DNA from other sources, and coaxing data out of the ancient material is costly and painstaking work. At a more fundamental level, part of coming up with a research question requires determining whether the necessary samples even exist and, if so, where they are and how to get access to them.
One of the great difficulties is trying to find your own way and trying to find your own realm of research where someone has not already been out trying to find the samples.
These challenges aren’t enough to keep some intrepid researchers away. Currently the field is dominated by a relatively small number of established, well-funded labs, but there is another generation of researchers who are coming into their own and looking to carve out some space for themselves in this dynamic field. Here are the stories of three of these scientists.
The animal lover
In March 2010, Eline Lorenzen was hunting for polar bears … sort of. Guided through a dark, maze-like basement at the Russian Academy of Sciences (RAS) in St. Petersburg, she reached her quarry: a box of ancient bones. With utmost care, she drilled off a small piece of each bone, putting it into a plastic bag to take back to Eske Willerslev’s ancient DNA lab at the University of Copenhagen, where she was a postdoc. She hoped that sequencing and analyzing the DNA in these ancient samples would offer new information about how polar bears evolved.
A few years prior, Lorenzen had been working with modern DNA alone. An animal lover—as a child growing up in South Africa she frequently went on safaris in the savanna—she studied the population genetics of African hoofed mammals for her Ph.D., also at the University of Copenhagen. There, she met Willerslev and felt inspired by his enthusiasm and the exciting research going on in the ancient DNA field. She decided she wanted to join his lab and dive into ancient DNA herself—leading her to RAS’s basement.
But collecting ancient samples was just the first step. Once she got the fragments of bone back to the lab, she donned a full body suit, face mask, hair net, and a double layer of gloves. In a fume hood in a lab designated just for ancient DNA work, she cleaned the bones and scraped off the outer layers to get rid of as much DNA contamination as possible before grinding up a portion of each of the bones and conducting the DNA extraction. Next, she used amplification and sequencing methods to find out whether the DNA was too degraded or contaminated to yield useful results. If the sample was usable, it was back to step one to optimize the sequencing results. “It’s super difficult to get data,” Lorenzen says. “It takes a lot of work and a lot of time.” She is still working with her ancient polar bear samples more than 5 years after collecting them.
Complicating matters further is the fact that, as the field matures, researchers need to collect and analyze sequences from multiple specimens to prove that their results are not based on the idiosyncrasies of a single sample. “When you’re the person at the bench, it can be quite disheartening how long it takes to generate data at a scale that can be published in a high-impact journal,” she says.
Luckily for her, the bones she collected samples from were free from other drill marks, indicating that no other researchers had yet sampled them. It had been a different story at RAS’s zoological museum a few days earlier, where she wasn’t allowed to sample the bones she had come for because they were already covered with drill marks from other researchers who, like her, hoped to mine the relatively small number of ancient polar bear samples to reveal their evolutionary history.
“It’s a very competitive area because ancient DNA is a hot topic,” says Lorenzen, now an assistant professor at the Centre for GeoGenetics at the University of Copenhagen’s Natural History Museum of Denmark and a visiting scholar in Rasmus Nielsen’s lab at the University of California (UC), Berkeley. “It’s a tightknit community, but it’s also a tight field, in that a lot of people are interested in the same overall questions”—for example, what are the evolutionary histories of humans, charismatic large animals like polar bears, and essential agricultural crops like maize. “There are lots of people who are interested in working on these ancient samples. One of the great difficulties is trying to find your own way and trying to find your own realm of research where someone has not already been out trying to find the samples.”
Junior researchers trying to establish their own niche may want to focus on the less studied species, even if they are not as glamorous, Lorenzen says—as long as the research question is important. In addition, by developing relationships with museums and curators who have ancient specimen collections, they can stay abreast of what samples are available, which can help them develop novel research questions.
Communication and cooperation with the other researchers in the ancient DNA field are crucial steps toward finding the balance between competition and collaboration. Lorenzen recommends conferences as an important way to build these relationships, especially as you are starting out. “You need to navigate who else is in the field and what they’re working on. It’s a bit of a minefield, but … if you have a good working relationship with them and have met them many times at conferences and workshops, it makes life much easier. That networking can never start too early.”
Lorenzen has not restricted her postdoctoral research to ancient DNA samples. She has done some work on polar bear evolution by working with modern polar bear DNA, for example. “I keep going back and forth and saying, ‘never again’” to ancient DNA, she says. But despite the technical difficulties, she always finds herself going back to it. “I’m not so impassioned by ancient DNA that I have to work with ancient samples for the rest of my life, but right now, for my work, I can get some unique insights that I wouldn’t be able to without incorporating ancient samples.”
After doing an ancient DNA Ph.D. in Europe and a postdoc blending modern and ancient DNA work in the United States, María Ávila-Arcos knows that her decision to return to her native Mexico is a risky career move. It will mean fewer resources and more challenges to doing research in an already challenging field. But when she got an offer to start her own research group studying the evolutionary history of Latin Americans at Mexico’s new human genome research institute, less than 3 years after finishing her Ph.D., she couldn’t turn it down. “I don’t feel quite ready to make this big step, but … I know I would always be sorry if I said no and didn’t give myself the opportunity to at least try.”
This isn’t Ávila-Arcos’s first bold career move. As an undergraduate studying genome sciences at the National Autonomous University of Mexico, she was immediately fascinated when she read some of the early papers about sequencing Neandertal mitochondrial DNA. After interviewing for a position as a Ph.D. candidate at one of the top ancient DNA labs, she was offered a spot—but in the meantime she had also received an invitation from Thomas Gilbert, who was just starting out at the time and has since become a leader in the field, to work with him at the University of Copenhagen. Even though she felt that working in the famous lab would be “a guarantee for a perfect ancient DNA career,” she opted for Gilbert’s lab, which felt like a better fit for her personality. “It was a decision I made from the heart,” she says, “and I never ever regretted my decision.” During her Ph.D., she worked with ancient DNA from a variety of organisms—including whales, koalas, and maize—before settling on humans, to study their evolution.
Nowadays, as a postdoc at Stanford University in Palo Alto, California, she focuses on premodern humans’ population movements. For example, she was involved in work studying the recent population history of Native Americans, which just this week was published in . Such questions “are relevant for living populations. … We all ask about our origins. We’re always curious about knowing where we’re from.” Currently she is working on a project to learn more about African ancestry in Mexico. She hopes to analyze the DNA from archaeological samples of the earliest known Africans to be enslaved in Mexico. In the meantime, she has just completed fieldwork collecting the DNA of living Mexican participants to complement the older specimens.
One of the primary challenges she faces in starting her new lab will be working with limited funds. “It’s a super-competitive field right now, and you have to have a lot of money to sequence as many samples as possible,” she says. “I don’t know if I can be in the position to compete when I’m in Mexico. … I will have to be creative and think carefully about where I put my money.”
She hopes that her lab will be part of a larger movement to “democratize” the discipline. “Right now there are kind of monopolies running the field,” she says, referring to the large, established, well- funded labs that are responsible for most of the major ancient DNA projects. Sequencing ancient DNA is expensive, so it can be hard for new labs to enter the field, but efforts are underway to develop methods to bring down the cost and make the field more accessible to newcomers.
Another challenge—which she is trying to turn into an opportunity—is the lack of geographical diversity in the field. “The focus so far has been mostly on samples found in a European context,” she says. “That has some DNA preservation reasons, but it also represents a bias. Just working on samples from Europe basically leaves the rest of the world unattended. There are a lot of interesting questions that can be answered beyond Europe,” such as the evolutionary histories of humans, animals, and plants from elsewhere in the world, which is one of her goals for her new lab.
While the experimentalists in the field labor in the lab to coax sequences out of ancient specimens, Sriram Sankararaman rarely strays far from his computer. He is working at the other end of the analysis pipeline, developing methods to extract meaning from the increasing amount of genetic data, ancient and modern, that his wet lab colleagues are generating.
A computer scientist by training, he became interested in applying his expertise to biological problems when he was exposed to some bioinformatics research as an undergraduate intern at IBM Research. When he moved on to his Ph.D. in computer science at UC Berkeley, he wanted to explore this field further, so he took a few bioinformatics courses. “I really loved the kinds of problems and kinds of challenges” bioinformaticians were tackling, he says. For his Ph.D., he focused on biological questions, for example developing an algorithm to study populations that are mixtures of other populations, which is the result of a process called admixture, and infer properties of the original populations.
As he was finishing his Ph.D., in 2010, the first full Neandertal genome sequence was released, and he realized his expertise might have an exciting application in ancient DNA. “To me, one of the most exciting discoveries to come out of that analysis [of the Neandertal genome] was that there was admixture between modern humans and Neandertals. It seemed like both a really exciting opportunity and a challenge to see whether we can learn about not just admixtures in recent populations but also these admixtures in ancient populations,” he says. His interest led him to a postdoc with David Reich in the Harvard Medical School genetics department.
Although Sankararaman works on computational method development, he emphasizes the importance of being connected to the experimental action. After completing his Ph.D., he attended a 3-week genetics boot camp at Cold Spring Harbor, where he got his hands dirty with some wet lab experiments. “Methods need to be developed in close proximity to the biological questions and the technologies driving these questions,” he says. “That’s a very valuable expertise to build, figuring out what the technologies are, their limitations, and developing methods that can work within the limitations of these technologies.”
In November, he will start his own research group as an assistant professor in the computer science department at UC Los Angeles. He plans to use both ancient and modern sequences to investigate population genetics and evolution. As the amount of available sequence data continues to grow, he thinks he has his work cut out for him.
“Ancient DNA is a really valuable resource that needs to be incorporated into our [genetic analysis] methods,” he says. For example, “in population genetics we’ve been building models of human population history, and ancient DNA gives us a way of testing these models. … Ancient DNA is going to be an integral component of refining those models, testing and validating them.”
But, Sankararaman says, “there are still important problems for which we need good methods.” For example, new methods developed specifically for ancient DNA could allow researchers to use both the sequence and available information about the sample’s age in their analysis to build more accurate models of a species’ evolutionary past. There’s also the fact that access to ancient DNA is often limited, so researchers sometimes have just one or two sequences to conduct analyses with to try to represent an entire population. Analytical approaches must be developed that can work around such constraints. “We have the data but the methods are playing catch-up,” he says. “We’re far away from having methods that can really leverage all the information out there. … There’s still plenty of interesting work to be done.”