Kathrine

Thank you for joining me today. To start, would you mind introducing yourself, your field, and the main questions your research group studies?

Professor Hey

Good morning, I'm Jody Hay. I'm a professor, actually professor emeritus now at Temple University where I do research on evolutionary genetics and I teach courses in introductory biology and in human evolution and computational biology. In my research, we develop mathematical and statistical models for trying to use genetic data to figure out evolutionary history, how species have diverged over time.

So, I'm a computational biologist these days. I used to run what we call a wet lab where we have reagents and did a lot of DNA sequencing, but I was always pretty mathematically inclined and computationally inclined. And about 20 years ago, I switched my research to being all computational.

And so, I've since then been taught a lot of courses in computational biology and started graduate programs in computational biology. So, yeah, and I write research papers and... Sure.

It's all very what we call basic research. That is, it's not applied in the sense of trying to cure a particular disease or invent something or get patents. It's all directed at very basic questions to serve our own curiosity about how evolution happened.

Kathrine

So, next, what first made you interested in evolutionary genetics?

Professor Hey

Well, I was always interested in biology and biological diversity. I had a great biology teacher when I was in high school. And, you know, I came then to understand that evolution is how we understand how organisms came into being and how new species come to exist.

So, when I went to college, I majored in biology. And I did a research project for one semester, a couple semesters, and that made me think that maybe I could be a scientist and study biology somehow. I didn't really know at the time what it would mean to study the things I was interested in.

But my first year in grad school, I had a professor who taught me how you could use genetic data to figure out some of the things about how evolution happened. And that was in a course called population genetics, where you apply understanding of how genes get passed on from generation to generation. You apply that understanding to populations.

And that professor ended up being my PhD advisor, and I went on to become what you call a population geneticist or an evolutionary geneticist. So, that was a quick overview of my path.

Kathrine

Awesome. So, next, could you walk us through a study that you conducted involving evolution, from the question to what the genomic evidence revealed?

Professor Hey

So, one of the big things that I've worked on for a long time, for many years, is, you know, for species that have recently diverged from each other, that have a common ancestor that was not very long ago, say, you know, less than 10 million years. Could you use genetic data to figure out how and when they diverged? And so, we began developing models of this process.

And in those models, you build in things like population size and hybridization rates between the diverging species. And if you have enough data from each of the species, from the organisms that are alive today, from the patterns of differences in those genomes, you can, in principle, use that to reconstruct the history of their divergence, how they came out of a single ancestral species. And we developed those models to a pretty sophisticated point, and we were looking around for good data sets to apply them to.

And we found a good data set from a group of chimpanzee species and subspecies. And there's two species of chimpanzees, a common chimpanzee and what's sometimes called a pygmy chimpanzee, though they're not really very much smaller than the common chimpanzee. They're also called the bonobo.

Kathrine

Oh, yeah.

Professor Hey

Yeah. And so, we applied the model to subspecies and to those two species. And we figured out, we estimated how long they diverged.

And it looked like the bonobo and the common chimp diverged about a million years ago, and that they had been exchanging genes, hybridizing to some degree, though not a lot, over that period of time. So, that ended up being a pretty useful study for a lot of people. That work gets cited a lot.

So, that was a lot of fun.

Kathrine

Anyway. Oh, I just want to mention, when I was little, I think I actually saw the study that you did on the chimpanzees and bonobos.

Professor Hey

Oh, that's cool.

Kathrine

Yeah. And when you mentioned it, I feel like it kind of dragged out that memory. So, that was interesting.

It's like a full circle moment, I think. Yeah. So, next, your lab develops computational models and software for estimating population history.

So, how do you test whether a model's reconstruction of the past is reliable when it's difficult to really verify it?

Professor Hey

That is a great question. It's tricky because you can't go back in time and see what the truth is. And for most situations where you'd like to use that model, like with the chimpanzees and bonobos, we don't know the actual truth.

We don't have anything else that really helps corroborate. In those cases, we don't even have like fossil evidence of their ancestors. So, in cases like that, we have to come up with some data that we do know the true history to test the methods.

And the main way that we do that is that we actually simulate. We make up some fake data. We write a computer program that generates DNA sequence data from a bunch of individuals that had a very specific history of when the two populations diverged and how much hybridization they might have had, what their population sizes were.

And we can use that computer program to build data sets that look kind of like real data sets and that they're DNA sequences. And then, but what's special about them is that we know the real numbers for population size and splitting time and hybridization rates. And so, we can run the methods that we've developed for estimating all those things and coming up with numbers for those things on the simulated data.

And that tells us if they work. It doesn't prove that they will work on real data, but it's a critically important validation. And the more you do that and you try out different kinds of histories, the more confidence you can get that your method actually works fairly well.

So, that's the main way we do it.

Kathrine

Mm-hmm. Incredible. So, when scientists study human migration ancestry and population history, what ethical responsibilities do they have when they're presenting their conclusions?

Professor Hey

So, the ethical implications pretty much only come in when you're studying human population. I mean, there are ethical issues for other species like, you know, are you being fair to the people who generated the data? Because we're not generating the data.

We're using other people's data. Are you citing people fairly? Are you collaborating in an ethical way?

And those kinds of things. But when you're studying human populations, which we do, then a whole other set of issues come in. And they're mainly around how you identify human populations.

And because for the models and understanding history, you would like to use names of populations and groupings of present-day individuals and populations that have some real connection to the history. And in the case of human populations, that history is much younger than, say, the chimpanzees and bonobos. Human populations have been spreading around the world only around 60,000 years or so.

And within Africa, you know, maybe a hundred thousand years of moving around in Africa. Maybe a couple hundred thousand years since the origin of our species. And so, if you're going to identify people, you need to have names and ways of grouping them that make sense in that historical context.

But humans have all other kinds of ways of grouping themselves that have historical meaning to those individuals. You know, you might think, okay, you might say, you know, I'm, you know, I'm from Africa. Or in my personal case, you know, I'm from Great Britain or something like that.

And, but that might not, you know, my ancestors might be from, they might mostly have been from the British Isles, but they might have been from other places. And people use, you know, they also use religious groupings. Sometimes they use groupings that turn out not to actually have much historical reality.

So, these are issues that come up, like if you have your ancestry examined by sending your DNA off to one of these companies like ancestry.com or 23andMe, though they're not doing this anymore. And they'll give you a breakdown of your ancestry, but it's really hard to interpret those numbers. They'll give you names of regions in the world that don't necessarily give you the timing of things and stuff.

So, this is a big issue when you're doing human population genetics. You want to have, you know, geography can help a lot and say what region of the world, because our ancestors didn't move as much, nearly as much as we do in our lives today. Today, humans are moving all around and, you know, human populations are mixing up genetically, you know, in ways that they didn't really used to.

So, that's a big issue that my research intersects with. And we just need to be really careful about how we identify groups and give names to groups.

Kathrine

Absolutely. So, what does a typical day in your life look like? And has your time divided amongst like developing models, analyzing data, perhaps writing and teaching or other activities?

Professor Hey

So, I probably spend like 80% of my time sitting at the computer. And most of that is just doing research. That could be developing the math of a particular model, or a particular way of doing inference, writing the programs to implement those models, running the analyses, that is running those programs, or simulating data.

I also spend a good bit of time writing papers. And then working on lectures for my courses. And other sundry stuff, writing recommendation letters, doing reports for some committee I'm on.

And then there's teaching. I would usually teach one course a semester. I also lead a journal club, which is like a seminar.

Other time is meeting people in my lab, or doing Zoom meetings with collaborators, and there's committee meetings and stuff. It's a pretty full day.

Kathrine

That's exciting. So, next, what do students commonly misunderstand about studying evolution or genetics?

Professor Hey

I think probably the biggest thing that it takes a while for students to get their head around is the way that evolution and genetics go together. And just how much that combined set of concepts is absolutely fundamental to every single thing in biology. They fail to understand the scope of it all.

So, the modern theory of biology, of how creatures came to be, plants and animals and microorganisms and humans and everything, is a combination of Darwin's theory of evolution by natural selection, and Mendel's theory of inheritance, of how genes get passed from offspring to parent. And then the modern version of that, which is how it happens at the level of DNA sequences, and that those theories fit perfectly together, which is a kind of cool way of suggesting that they might really be pretty accurate. And that modern evolutionary genetic theory is the theory of life.

It is the theory of how life arose. You know, at the very beginning, some early kind of molecule started leaving descendants, modified descendants, and natural selection started acting and leading to more complicated replicators, which eventually evolved chromosomes and proteins and cells and multicellular life and all the rest over 4 billion years. So, we really do have this amazing theory.

And it takes a while for students to really get their head around all that and just have that sink in and realize just how explanatory that theory is and how strong all the evidence is in support of it. And once I do, it's kind of an eye-opener. It's pretty cool to see.

Kathrine

Yeah, that's great. So, finally, what is one realistic step a high school student could take this summer to begin exploring evolutionary genomics, population genetics, or bioinformatics?

Professor Hey

Well, if they're really, really serious about delving into those things, then they want to get comfortable kind of teaching themselves things. What it sometimes students takes them a little while to realize, but, you know, if they're going to go into college, particularly if they're going to, you know, get a master's degree or something like that, is that most of the teaching that they do, most of the teaching, and the problem-solving, they're guided by their teachers. And teachers can get you through some of the most challenging parts.

So, there's, if you're really, students really up for that, then there's a wealth of resources online for learning about evolution, for learning about genetics, for learning computer programming. I don't have specific suggestions, but there are some good online courses in all of these things. And now there's very good AI tools to help guide you in that as well and to answer your questions.

They'll usually be fairly accurate in their response. And I think, you know, the biggest thing that kind of sort of separates students who really go on to be scientists, say, from those who don't, is those who really get comfortable that they're beating their head against hard things that they don't understand until they finally figure it out. And teaching themselves things.

And the specific resources are not the big question there. There's lots of resources. The big question is, you know, are they really prepared to overcome all those stumbling blocks and dig down and make sense of it when they're just working by themselves and they don't have someone to hold their hand.

Kathrine

That's an inspiring note to end on. Thank you so much for sharing your work and advice with me. I've really appreciated your time and learned a lot from what you've had to share, and I think students will also learn a lot from hearing your perspective.

Professor Hey

Oh, I'm very glad to share all this with you, Kathrine. Thank you so much for doing these videos, and it's been great talking with you.