A recent publication in eLife captures a sequencing project led by the Pritchard lab, in which thousands of genomes were analyzed to reveal how and where people moved during the Roman Empire. Lead authors Margaret Antonio and Clemens Weiss gave us a look behind the scenes of this paper coming to be.
How did your group decide upon this topic?
Antonio: The Pritchard Lab is broadly interested in human population genetics and complex traits. (Just before I joined the lab) Ron Pinhasi’s lab (University of Vienna) was interested in understanding the evolution of human traits (e.g. lactase persistence) through analyzing ancient DNA. Their lab specializes in methods for extracting and processing ancient DNA. The Pritchard Lab, with its expertise in human genetics and computational work, was a natural fit for that kind of collaboration. Before getting into the evolution of individual genes and traits, it’s important to first understand the population structure and how it changes over time. In other words, in order to understand how genes and traits are being selected for by environmental pressures, we have to first understand how these genes naturally vary between populations and are passed down. Therefore, we focused on using ancient DNA to first understand how population structure varies across time, across Europe, the Mediterranean, and the Roman Empire.
Weiss: Much of this research was motivated by a previous study in the lab, which investigated genomes from individuals buried in and around Rome over the last ~12,000 years (Science, 2019). This study showed surprising demographic diversity in Rome at the height of the Empire. To learn more about the demography of the Empire beyond just the city of Rome, we sampled individuals from that time period, but across the range of the Empire and beyond. Part of the motivation for this was to see if the demographic diversity observed in Rome was observed elsewhere, or if it was specific to the capital of the Empire. Another motivation was to fill a gap in sampling during the historical period: much of ancient DNA research has focused on pre-history, where there’s no written record, and where we have learned a lot from being able to study genomes with the help of ancient DNA. However, I think it’s becoming increasingly clear that we can learn a lot from genomes from the historical period as well, with the added benefit that we have a much richer body of work from historians that can help us put our findings into context.
What made you interested in ancient genomics – is that something you arrived to Stanford with, or discovered during your time here?
Antonio: I happened upon this project by coincidence. I was in the Biomedical Data Science PhD Program, which has students studying everything from electronic health care records, to medical imaging, to human genetics. I had no idea what I wanted to study. I went to a few research presentations by Pritchard Lab graduate students – I was captivated by the way they communicated the story of their research and their approach to tackling scientific problems. That was something I wanted to be able to do. When I met with Jonathan about doing a rotation, he wrote a list of ongoing projects in the lab on a whiteboard. To be honest, I didn’t know what most of the words meant. However, being a fan of the movie “Gladiator”, one stood out to me: “Ancient DNA from Roman Empire.” That led to 6 years of work in ancient DNA, a field I had never heard of.
Weiss: I arrived at Stanford with a background in ancient genomics. During my PhD, I worked on ancient DNA primarily from plants and microbes, but also ancient hominins. I was attracted to moving towards ancient DNA of humans for my postdoc in part because of the richness of data. One of my research interests for my postdoctoral work was developing novel statistical and computational methods to make sense of the increasing wealth of genomic data available, so moving towards studying humans made sense for that. My primary fascination with ancient DNA is that it allows us to temporally sample genomes. Usually, generating such data is reserved for prospective studies for example through experimental evolution, where samples are taken through time. Since evolution is inherently a temporal process, this allows us to watch evolutionary forces in action. However, doing such experiments is not possible or practical for most systems, so ancient DNA is a great opportunity to study these processes retroactively, with the added benefit of being able to study the evolutionary consequences of past events (such as human migration, plant responses to climate change…).
What was the most challenging aspect?
Antonio: Both projects – the one that focused on the population of Rome (Science, 2019) and our recent one on the Roman Empire (eLife, 2024) turned out to be very involved. As a graduate student, it was challenging to stay focused on a project that lasted 3 years. We were constantly looking at the data from different angles, testing hypotheses, challenging observations – it was exciting and fast-paced, but also required endurance and team support!
Weiss: I’d say the most challenging aspect for me was to do right by the incredible body of work that exists by archaeologists and historians who have studied the Roman Empire for decades. I think there’s a risk with genomics to get lost in the data and lose track of the bigger picture somewhat. Staying cognizant of the historical context in which these humans (that we now have the great opportunity to study) lived, I think helped us to tell a compelling story that enriches previous work. I’d like to give thanks and credit to Walter Scheidel here at Stanford for some great conversations Margaret and I had with him about a lot of this context, which for me really changed the way I thought about both interpreting the data and telling the story we lay out in the paper.
The most surprising?
Antonio: I’m surprised by how it all came together. Midway through the project, we communicated the preliminary results to groups of collaborators via Zoom. There were archaeologists and museum directors from all over – Armenia, Lithuania, Croatia, and many other regions. Without the samples they contributed from their collections, we wouldn’t have been able to carry out this project.
Processing the samples was another huge endeavor, led by Ron Pinhasi’s lab. Extracting DNA from bones requires a lot of time and attention to detail. Those are just a couple examples of the pieces that had to fall into place. I look back and think, “Wow, so many people stepped up in different ways and at different times to contribute and make this happen.”
Why is it important for modern people to know about the mobility of ancient people? What personally excites you about it?
Antonio: In the last decade there’s been an uptick of interest in finding out one’s ancestry (through direct to consumer ancestry tests, for example). The results are usually quite broad – e.g. you’re 30% Italian or 20% Spanish. Even though we can trace ancestry back to a specific population, this work on ancient DNA has shown us that those populations of the past were extremely dynamic and mobile. The Italians didn’t stay in Italy and the Spaniards didn’t stay in Spain – they were moving, trading, interacting with other people across sea and land. With that context, ancestry is less like a category and more like a puzzle piece that makes up the more complex story of the past.
Weiss: What excites me about mobility, especially in the region and time period we studied here, is that it feels like a shift in the type of mobility towards something that more resembles mobility as we think about it today. In the context of the population history and genetics of western Eurasia, we often talk about large-scale migrations such as the Farming migration ~8000 years ago and the Steppe migration ~5000 years ago. These are large-scale migrations of populations that took hundreds of years to spread across the continent. In contrast, the Roman Empire facilitated “travel” more akin to how we think about it today. Being able to track this shift through studying genomes of individuals that lived at these times I think is fascinating.
Although you are no longer in the field, what could be some avenues for future research along these lines?
Antonio: We learned a lot about human population structure and ancestry from ancient DNA. As more ancient DNA is collected and methods are developed, I’m excited to see what kinds of mysteries ancient DNA can unlock.
Weiss: I think this research made me curious about two follow-up avenues: One would involve more targeted sampling of individuals that lived during and after the fall of the Empire. A possible explanation for the maintenance of population structure into present day would be a decrease of dispersal across the region after the Empire fell, but we did not have the samples necessary to answer this question. A second avenue would involve sampling Roman settlements more densely. I think there are interesting stories to tell about the demographic diversity of the Roman Empire, and this would be a way to really dig in.
the CEHG blog 