Latest / Polymath World Channel / Polymath World # Episode 24 - Dr Zachary Ardern
Transcript
- Polymath World (00:01.52) Hello and welcome back to the Polymath World channel. I am ready with my decaf El Grey to talk to a very, very dear friend today, as I often get to do on this channel. Get to meet so many wonderful, amazing people doing really good research. And it's not always about the older professors who've been in the career for a very, long time. We also like to highlight some rising stars and some young people who are blazing their own trail and working on the frontiers of their discipline. I'm delighted today to be joined by Evolutionary geneticist from Cambridge, Dr. Zach Arden. Zach, how are you today? It's great to see you again. Zachary Ardern (00:39.138) Yeah, doing well. It's great to be here, Sam. Polymath World (00:41.762) Excellent. Well, I can't wait to get into your field because we've not had an evolutionary geneticist here. I do often get accused of bit of biases towards genetics for obvious reasons, but why don't we, before we get into what that is, start with your own career journey. You know, were you playing with bacteria or microorganisms when you were a child? Zachary Ardern (01:02.518) Yeah, I actually wasn't that interested in biology until kind of my later teens. So I was into science, I was keen on physics and chemistry, but I thought biology was kind of a soft science and kind of not, maybe not really rigorous or not really very kind of detailed or mechanistic. Because the kind of biology that I kind of came across at school was kind of quite kind of whole organism ecology. whereas I was interested in, guess, molecules and mechanisms. So I got into molecular biology at university. Polymath World (01:39.462) Gosh, really that late? So, it didn't sort of capture you before you were 18. Zachary Ardern (01:47.042) No, I started to get interested in it maybe 16, 17 and realized that there was such a thing as molecular biology. that's, I was planning to do biology and chemistry in my undergrad but ended up just doing biology for my science degree. Polymath World (02:02.374) Wow, that's amazing. was there anything that triggered you making the jump from being interested in physics all the way over to biology? Anything inspire you in particular? Zachary Ardern (02:13.794) Yeah, I think it was starting to read about molecular biology. So starting to learn about proteins and genes and an evolutionary perspective on those and big questions about the origin of life and evolution and just realizing that, we can start to understand some of this stuff scientifically and it's really interesting. it's kind of related to at least to say organic chemistry which I was starting to be familiar with. So I realized that yeah there's areas of biology there I'm actually quite interested in and kind of went into that in university. Polymath World (02:48.912) Now people will be able to tell from your accent that you're not originally from Cambridge. tell us a bit more about where you're from and what education looked like over there. Zachary Ardern (03:00.47) Yes, I'm from New Zealand where I did all my studies. my, from my under, all my schooling and my undergrad and my post-grad diploma and my PhD were all in New Zealand and Auckland. My schooling was a fairly standard state school and yeah, that was a good, good background for me to go into university. In university, I did both arts and sciences. So I did a conjoint degree in majoring philosophy and economics and the arts. And I wanted to major in chemistry and biology in the sciences, but I couldn't quite make it work. So I just majored in biology in the end. So I was interested in both kind of the humanities side of things and the science side of things. And fortunately, it was fairly common in New Zealand to do a ConDriant degree to kind of take a kind of say shorter degree, but able to do both. So shorter than doing two full degrees, but kind of being able to major across both areas. Polymath World (04:04.326) Wow, that's, I mean we love that on the Polymath World channel, people studying multiple disciplines, but that sounds extremely rigorous. Is it a four-year program? Zachary Ardern (04:14.112) Yeah, so that was four years and I had two majors in the arts and one in the science. I did approximately eight or so courses in each of the majors plus some other kind of general education courses. Polymath World (04:27.92) Well, I love that. I mean, it sounds like a huge amount to put on someone's plate, but I do really like that. sort of wish we had a bit more. Obviously in America, you can have a minor as well as a major. And I think there's great advantages to students being able to do that. Whereas in Britain, we go all in, but we do it for three years. And we sort of specialize early. But you enjoyed sort of balancing the philosophy and the biology? Zachary Ardern (04:41.187) Mm-hmm. Zachary Ardern (04:54.84) Yeah, I realized I probably wasn't gonna make a career in philosophy kind of sometime during that degree. So I was really glad I had the biology and I got really interested in the kind of theoretical side of biology, which I realized had some overlap with philosophy. So that's kind of where I eventually ended up. But I was really interested in both the philosophy and biology. So I also did economics in the humanities side, but decided I wasn't actually that interested in that. That was where I had thought I might make the career, but then realized actually I was really interested in the science, and so did post-grad in the science. Polymath World (05:30.842) I imagine it's quite helpful in terms of helping to train your mind to ask big questions and think outside the box and think big pictures. Have you found that that sort of helped your career? Zachary Ardern (05:45.047) Yeah, I'm not sure. I think it definitely helped my thinking to do a philosophy degree and to just to get feedback from experts in thinking critically and really forming arguments carefully and thinking about what counts as evidence. There's lots of really interesting differences between the humanities and the sciences. So I think I struggled with that. as I started in science and kind of writing scientific papers, it's quite a different style to writing a philosophy paper or writing even a philosophy essay in undergrad. So, yeah, so, but I think it gave me kind of at least an interesting angle on the science. Polymath World (06:12.57) Yeah. Zachary Ardern (06:23.95) I might have done better in some senses if I'd just focused on the science, but I think doing both has allowed me to kind have a broader perspective on the theoretical side of biology, which has been, eventually has been really useful. Polymath World (06:36.602) Yeah, I don't think it's really held you back. Where did you go for your post-grad? Zachary Ardern (06:43.288) So this was all in Auckland, so University of Auckland. So post-grad was just a post-grad diploma initially, which is just a one-year kind of taught. It's like the first half of a two-year master's. And then I started a master's degree and then kind of turned into the PhD. So I just kept going with that project. Didn't actually finish the master's because I just kind of transitioned into the PhD. Working with a... Polymath World (06:54.075) Yes. Polymath World (07:04.75) that's cool. They let you jump to the PhD. Zachary Ardern (07:07.938) Yeah, I was really fortunate to get to do that. So was working with a professor who gave, who taught a really inspiring course on kind of theoretical, on evolutionary theory, I guess. And that's what I ended up kind of working on for the PhD. Polymath World (07:19.566) Right, and where was that and what was your particular research proposal? Zachary Ardern (07:24.918) Yeah, so this was also in Auckland and the research I brought my thesis was called Small Change, Adaptation and its Costs and Experimental Evolution. So I was studying... Yeah, no one's going to read it, please don't, it's not that good. But I was studying... Polymath World (07:35.182) available in all good bookshops. Zachary Ardern (07:44.079) the genetics of change over reasonably small time scale. So was working with data from a previous experiment, which a previous PhD student had worked on. And I was analyzing, was preparing the DNA to be sequenced and getting that sequence and analyzing the data in the context of the broader kind of theoretical field to try and understand what kind of changes we were seeing over a small time. This was in a yeast, so in baker's yeast. Saccharomyces cerevisiae, which has a reasonably small genome. It's kind of tractable to kind of look through the changes that have occurred over a few hundred generations. You had to try and understand under different selector pressures what was happening in populations of this microorganism. And that gave me a really solid basis in evolutionary theory and genetics and molecular evolution and various related fields that kind of set me up for future research. Polymath World (08:40.538) Yeah, at this point, why don't you sort of really help define and outline what evolutionary genetics is? And you've mentioned here a model organism, which is very helpful to understand. What are you trying to do and what does it uncover for you? Zachary Ardern (08:51.032) Sure, yeah. Zachary Ardern (08:59.456) Yeah, so actually my work has kind of covered in some sense the full scale of evolutionary genetics. I've always, basically, nearly always been working with microorganisms. So the PhD was in yeast, a single-celled eukaryote. And since then I've worked with bacteria and viruses, but over quite different timescales. So the PhD was looking at evolution over the course of a few months. So this is changes in the DNA. that occurs in a population are partly due to selective pressure and partly due to other other factors. So these organisms have grown in quite a very stressful environment where a lot of different changes would in principle be advantageous when they occur. So those changes that were advantageous kind of grow in frequency in the population over time. But since then, some of more recent work has been looking across almost the whole history of Earth. So billions of years rather than a few months of evolution. But in both cases, it's evolutionary genetics because we're comparing genomes that have arisen through an evolutionary process. And we're comparing them to each other to try and trace back what changes occurred and what kind of factors were involved in. producing those changes, whether that was natural selection or mutation or potentially other factors as well. Polymath World (10:27.408) So how would you exert selective pressures? You mentioned a very stressful environment. What does that look like? Zachary Ardern (10:34.956) Yeah, so during the PhD, the previous student had grown in populations of this microorganism under conditions which had load of nutrients, basically. So they kind of were just barely surviving. They weren't growing very fast. But if they were able to change some aspect of their physiology, they could grow a bit better under those conditions. So. One thing that they tend to do and that lot of researchers has seen is they tend to duplicate a gene which transports sugar. And so the environment they in was a low sugar environment. And if they duplicate that gene, it helps them kind of suck the sugar into the cell more effectively. though the... members of the population that have that duplicated gene tend to do better and that actually quite quickly kind of dominates the population that those cells reproduce a lot faster and kind of take over the population. So that's a nice example of just in some sense a really small change that is really important for the rate of growth of the population. Polymath World (11:39.238) So the most famous experiment, an example that comes to mind here is the Lenski long-term evolution experiment. I think this is amazing. I think it's really extraordinary. You mentioned over billions of years and looking at the early Earth, you sort of trying to create in lab conditions, simulating what the earlier would be like or different. Zachary Ardern (11:47.95) Right. Zachary Ardern (12:02.085) yeah, I should clarify that. my recent work hasn't been experimental and it hasn't been directly from experiments. It's been from the kind of real life experiment of evolution in the real world. So we're comparing genomes of different species like the bacteria E. coli and the bacteria Salmonella, which are quite similar to each other. And then other bacteria kind of working outwards in terms of their relatedness to say E. coli. So E. coli is a bacteria that Polymath World (12:29.542) of these. Zachary Ardern (12:31.479) lives in our gut. Polymath World (12:32.76) Yes, yeah, like the flashiest of the micro-organism model organisms. These are these ones where sort of the clades are really well known and you can sort of trace back where they they shared a common ancestor and you can sort of Zachary Ardern (12:39.459) break. Zachary Ardern (12:50.925) Yeah, so because we have all the full genomes, we can kind of infer the full evolutionary history with reasonable confidence. So I'm interested in particular aspects of that. I'm particularly interested in the origin of new genes, so genes that don't seem to have a protein coding ancestor, which is kind of a very kind of niche topic within evolutionary genetics. I'm the question, how do you get new genes, given that, say, each bacterium has something like 4,000 5,000 genes, maybe on average? And different bacteria sometimes have different genes that don't seem to be related to the genes you find in some other bacteria. So this process is involved, and people can look at my papers if they want to learn about those, but this process is involved in what seem to be creating new genes from sequences which didn't used to encode proteins. So just a quick recap that the proteins are the kind of functional molecules in the cell or that they do a lot of the functions in the cell and they're encoded in sequences of DNA. And not every sequence of DNA encodes a protein, but some of them do. And so I'm interested, one of my research questions is what are the processes that get you from a sequence that doesn't encode protein to a sequence that does? And I've looked at that in both viruses and bacteria. Polymath World (14:06.132) that's amazing. I would really love to get into that in a sec. Just want to make sure I finish your career arc first. Did you start teaching when you were a PhD student or when you were a postdoc? Zachary Ardern (14:13.004) Sure. Zachary Ardern (14:19.791) I did a little bit of teaching, well I actually did quite a bit teaching during the PhD just because I wasn't earning much money as a PhD student and I was happy to have the extra income. So I ran labs, so I taught kind of first year students how to do basic lab techniques and also a little bit of basic computational stuff but mostly lab stuff. I was also teaching some philosophy actually since I had graduated in philosophy. I was teaching some critical thinking courses and others. So yeah, I was a graduate teaching assistant, partly just to earn some money and also because I enjoy teaching. And then finishing the PhD, I moved to Germany to Munich for a postdoc, which ended up being approximately four years in Munich. And there I was just doing research. I didn't do any teaching while I was there. Polymath World (15:09.574) And your research there, I understand, was on overlapping genes. So could you tell us about that? That's really interesting. Zachary Ardern (15:13.283) That's right. That's right. Yeah, it's a little bit hard to do this without a diagram, but it turns out that... So it's reasonably well known that DNA is read in triplets. So it's read in series of threes, starting usually with an ATG code. It turns out that if you shift the words... So you can imagine that each word is three letters. If you shift the words that you're reading by one letter, so instead of reading... ATG, CGG, you read TGC, GG something. So you can shift by just one letter, then you read a completely different set of triplet messages, and each triplet message encodes an amino acid, which goes on to build a protein. So it turns out that the same bit of DNA can just being shifted by one or two or reading from the anti-sense because DNA is double-stranded, it's read in both directions, you read a completely different protein. And this is reasonably well known to be quite important for some viruses and it also exists in bacteria. So I was looking at this in bacteria mostly during my postdoc but then when the COVID pandemic happened, we switched to also looking at SARS-CoV-2, the virus that causes COVID, because it turns out that in that virus there were a bunch of these overlapping genes that not many people had studied. that turned out to be kind of an interesting project that took a year or so of my life was kind of focusing on overlapping genes in SARS-CoV-2 and in COVID instead of in bacteria. Polymath World (16:57.286) Yeah, I think it took over a lot of people's research from what I gather. This could be pretty devastating if you have sort of a frame shift mutation, but I imagine also provides a lot of raw material for evolution. Was that sort of... Sorry, go ahead. Zachary Ardern (17:01.272) Yeah. Zachary Ardern (17:11.607) Yeah, so it turns out you can have an existing gene and you just read it from a different... if the ribosome just slips by one and reads a completely different thing. you can also get frameshift mutations and they are traditionally thought to be super deleterious. It turns out that they're not as deleterious as they could be due to some interesting properties of the mapping between the DNA and the amino acids. So I've written a few papers related to the different ways in which either frame shifts or just reading a different set of triplets can produce new proteins, which I think is kind of really interesting raw material for evolution to work with. Usually it's not very useful and it doesn't go anywhere. It's not selected, but occasionally it looks like it is selected. And kind of an ongoing sub-theme of my research is exploring ways in which that contributes to the creation of new genes. Polymath World (18:11.79) It's not too hard to see from here the immense benefit of understanding how evolution works through your discipline. Obviously, these don't just apply to microorganisms. This works for everything. Was it from there that you went to Cambridge? Zachary Ardern (18:33.582) Yes, I was in Karlsruhe for a few months. So that was another university in Germany just because my professor in Munich had retired and I needed to find a new position. But at the same time I had been applying in Cambridge and kind of moved then from Karlsruhe to Cambridge where I was kind of had the freedom to explore the research I'd been studying to develop while I was in Munich. Polymath World (18:55.888) And this is moving into the evolution of new genes. Is that right? Yeah. Zachary Ardern (19:01.048) Right, so Immunica was largely focusing on overlapping genes and just some of properties, whether they're expressed or not. I wanted to look at the evolution, I did a little bit, but I didn't have that much time for that. And when I got to Cambridge, I kind of designed a project around exploring the origin of novelty in proteins, including through overlapping genes, but also through other mechanisms. And topics related to that is what I've been working on over almost the last four years. So I've been working with various students along the way on various aspects of genetics of bacteria and the origin of new genes in bacteria and also in viruses. Polymath World (19:44.858) Now, you work at my absolute favourite place in the whole world. I think it's just sort the coolest building I've ever seen in my life. The Sanger Centre in Cambridge. Well, just outside Cambridge. And it's such a magnificent place. It seems like there's bits of architecture to do with research there over the years. It's like this all-inclusive place where a scientist could be happy and have everything they need. and it's very famous for the work on the human genome project there. So tell us a bit about your work at the Sanger and what being a researcher there looks like. And you've talked about the kind of freedom that you have to research. How do you come up with the questions that you want to ask? Zachary Ardern (20:38.404) Yeah, lots of questions there. So the Sanger Institute is fairly famous because it was massively involved in the original sequencing of the first draft of the human genome. So I think approximately a third of the sequence was sequenced. So it was kind of inferred through machines at the Sanger Institute in the 90s. I think at that time it was the Sanger Center. And now it has various programs exploring different areas of genomics. So all of it's related to genomes from kind of evolution of animals and plants, which is called the Tree of Life program, through to cancer and various areas of human genetics and trying to do genetic engineering. So what happens when you change things in different organisms and to understand the relationship between genetics and the effects on the organism. The program I work in is called Parasites and Microbes. So it's basically microorganisms. And the group I work within is doing bacterial genomics. So I was fortunate to get a fellowship there. And I did that by proposing a project that was related to bacterial evolutionary genomics. And I come up with that project kind of during my time in Munich and in Kalsora. It was kind of a fairly natural next step from the kind of research I had been doing before, but wanting to kind of expand in a more evolutionary direction because that's what I was really interested in. Polymath World (22:14.886) Yeah, amazing, amazing place. Really extraordinary research happening there all the time. See, it now involves the European Bioinformatics Institute as well, over the road. so, evolution of new genes. I researched DNA repair and a lot of the human DNA repair proteins have homologs in bacteria. Zachary Ardern (22:29.584) Mm-hmm. Polymath World (22:43.63) So you're looking at that and thinking this is really, really ancient. It hasn't changed very much between bacteria and humans. So is this a case of selective pressure or just a growing level of complexity in organisms that new genes will come about and get co-opted? Zachary Ardern (22:43.727) Mm-hmm. Zachary Ardern (23:04.674) Yeah, it's, that's probably a contentious question. I think there's a lot of this kind of really fundamental stuff is not that well understood. So as you say, like DNA repair, like I don't know those proteins particularly, but most of these kinds of core proteins are super ancient, that they are shared across either eukaryotes and bacteria or eukaryotes and archaea or both, or like all three. Because Polymath World (23:08.45) I'm sorry. Zachary Ardern (23:31.82) really interesting, a hugely interesting event in evolution was the merger of a bacterium and an archaeal cell to produce the first eukaryote, or least if that wasn't the first eukaryote it was fairly relatively soon afterwards. So it's kind of the development of modern eukaryotes is due to this ancestral merger of these two single-celled organisms. And so that means that most of the the genes in the human genome or the mouse genome or a plant genome or even the yeast genome can be traced back to either bacteria or archaea, these other single-celled organisms. But not all of them. It seems like there's a few, and it's really not that clear how many, but at least a few percent in any given genome don't seem to be that ancient. They seem to have arisen more recently over evolutionary history. So they may have arisen, so the origin of, or least the... last universal common ancestor was something like 3.5, 4 billion years ago. And some genes have arisen since then. So some genes might have arisen 3 billion years ago, some genes might have arisen 2 billion years ago, some genes might have just arisen 100 million years ago. This doesn't seem to be something that's happening like every day, but it does seem to be kind of happening at a low rate. Often when it does happen, the genes are not functional, so they're lost. So there's a reasonably high creation rate of genes, but there's not many, it seems like not many of those are functional. So they don't stick around very long. there's just the rare case that is both kind of translated, which turned into a protein and somehow that turns out to be functional in a particular context. And the, many areas of this are really not that well understood. This has been studied in different organisms and the field is still very young trying to piece this together and kind of give a single coherent picture. Or maybe there's not a single coherent picture. Maybe evolution is quite messy and it's involving different processes and different organisms. So there's a lot of like really fundamental questions around what mechanisms are involved in creating new genes, how much is selection involved, how much is different kinds of mutation involved. Zachary Ardern (25:45.392) Are there other factors like genetic drift involved? Yeah, many, many questions. And research in the field will have different perspectives on how much they think selection is important, for example. Polymath World (25:59.27) Hmm. Now you talked about sort of three billion years, two billion years, a hundred million years. People in the general public often forget that evolution doesn't work in years, it works in generations. And so like a hundred million years in bacterial terms, where you've got a short lifespan and massive reproduction is very different from a hundred million years in other higher organisms. are you sort of, do you have kind of a Zachary Ardern (26:06.075) Mm-hmm. Zachary Ardern (26:12.881) Mm-hmm. Polymath World (26:29.134) some sort of clock, like a molecular clock or something you're working with when you're dealing with microbes. Zachary Ardern (26:37.935) Yeah, there's definitely a lot of work in trying to work out when different events happened in the history of life. And that can be done through this father genetic approach of doing the genome comparisons and building an inferred tree based on those genomes. And some genes seem to kind of evolve at a reasonably constant rate over time. you can, for genes and animals at least, you can kind of... correlate that with certain evolutionary events, you might know that a certain animal group evolved, say, 70 million years ago. And so you know that that kind of point in the tree must be approximately 70 million years. So you can kind of, you can use that to help work out the different evolutionary rates. So this is a very complicated field and something I'm trying to get into a bit with my work on protein evolution, because there's some recent developments with protein structure that might help with some of these kind of tree methods. Yeah, but there's a lot of work on trying to work out what's the relationship between evolutionary differences in terms of number of changes it would take to get from this genome to that genome and how long that would take. And that's kind of the art of science of phylogenetics. Polymath World (27:56.772) Yeah, absolutely. Do you have a favourite micro organism, model organism to work with? Zachary Ardern (28:04.183) not really. E. coli is great. but yeah, all bacteria and archaea are really interesting. And because the work I'm doing is not experimental, I don't need to be working in model organisms, particularly. I'm just looking at all the genomes across all the bacteria. so there's some nice databases now of, there's approximately 140,000 bacterial species in the largest databases. and a few thousand archaeal species. So each of those has a genome, which is something like four million base pairs long. So it's a lot of data that you're working with, working with billions of data points to try and understand evolutionary changes. Polymath World (28:43.92) Yeah, I don't know how ancient yeast is, but obviously you're dealing with sort animals there, not bacteria. Zachary Ardern (28:56.569) Yeah, so they're eukaryotes, yeah, so they have a nucleus. So I think they're... Polymath World (28:59.578) Yeah, I- I- No, go ahead, sorry. Zachary Ardern (29:04.213) I, well, I was going to say, well, yeah, I was going to say I think they're a couple of billion years old, but in that the first eukaryotes are something like a couple of billion years old. But modern yeast is only, you know, diverged from other yeast species just a few million years ago. Polymath World (29:20.29) Alright. Okay, right. That negates the question I was going to ask about sort of how much one could learn from comparing like yeast and bacteria in that sense just because they're... Zachary Ardern (29:32.403) Yeah, so what's the same is comparing humans and bacteria in some sense. So there will be quite a few of the genes like the ribosome, for instance, will be shared with some interesting differences across them. So these really ancient genes are really useful if you want to study the whole history of life, which is something I'm interested in. And there's something like 100 or 200 genes that are pretty widely conserved across basically all organisms. Polymath World (30:02.48) Yeah, that is them. Zachary Ardern (30:02.642) Yeah, so those are good to look at for really old stuff. Polymath World (30:06.566) Yeah, that is amazing. I bet that's really, really interesting. On the non-experimental front, you were the first person to tell me about alpha fold. I think the first incarnation of alpha fold had come out, and you sent me message like, have you heard of alpha fold? I was like, no, what is this? So the computational methods, you've mentioned them already. Do they play a big part in your research? Zachary Ardern (30:18.417) Okay. Zachary Ardern (30:35.812) Yeah, so my research is all at the, I work with people who are doing experiments to test some hypotheses about the, say the function of specific proteins or to work out like how much this protein is expressed under different conditions. But I'm not doing that work. The only experimental work I've done is related to kind of growing the populations and the sequencing during my PhD, which didn't kind of get very far. So my work is all computational, which means I'm relying on data that other people have produced. And because there's so much of this data, actually hasn't been... There's a lot of information that data that hasn't been understood or used. So a lot of my work is compiling massive data sets of data that other people produced individually, but haven't been collated together. or using massive data sets other people have collated and trying to learn something about evolution that hasn't been studied before. Alpha Fold is one of the really interesting tools that's been around since 2020, 2021, roughly. And it was attempting to infer the three-dimensional structure of a protein. And it does that by... using information from protein sequences. So because all proteins apart from one such very, very new have quite a long evolutionary history and that they have a lot of relatives that you can compare them to. And so the information from doing that comparison, so comparing a protein from E. coli to a protein from Salmonella to a protein from Klebsiella, all these different bacterial species that are quite similar to each other. You can start to see what positions in that protein change, which ones don't. And the magic of alpha-fold is based on what's called coevolution. So if there's two positions in a protein that tend to, that if one changes, the other one also tends to change, looking across hundreds of thousands of versions of that protein called homologs. If those two positions tend to change together, that implies that they're quite likely to be close in three-dimensional space, even if they're not close, Zachary Ardern (32:53.122) along the kind one-dimensional string of amino acids. You can imagine a protein is like a long chain with beads on it, and each bead in the protein is an amino acid. And it might be that two... positions which seem like they're opposite ends of the string actually are tightly correlated in the evolution because they're actually next to each other in three-dimensional space. And alpha-fold uses the information from this very, comparison of hundreds of thousands called a deep alignment of sequences to work out which ones most likely be in three, close together in three-dimensional space. And from there, through some very clever artificial intelligence methods to build up this three-dimensional structure. So that's revolutionized structural biology, is more your field, but I'm interested in what it tells us about evolution once we have these structures. Polymath World (33:46.158) Yeah, so how can you just find answers to evolution through using alpha-fold? You sort of explained how alpha-fold works, but at what point can you say, wait a minute, this is an answer, this is something we've been looking for in terms of the evolution of genes or proteins. Zachary Ardern (34:06.674) Yeah, so a lot of evolutionary work since the 1960s, 1970s has been looking at sequences. So either looking at sequences of DNA or sequences of protein. Before that, there was evolutionary work before we had sequences, but then it was looking at morphologies. So it was looking at things like teeth shape or bone shape or skull shape or other kind of feet, like physical features of organisms that you can see in either living organisms or in fossil organisms. From living organisms, we can get really detailed information from sequences. But there's kind of a gap between sequences and functions in this cell. like, we know that this protein, like hemoglobin, has this string of amino acids. But that doesn't really help us, just knowing that string of amino acid doesn't really help us understand that much about how hemoglobin works. A much closer kind of... step to work out how hemoglobin works is to know it's three-dimensional structure and So this can be done experimentally, but it's a very slow process It's you know, know this better than I do but it's at least probably months of work per protein Maybe years for some of difficult proteins if it turns out through this computational method we can infer the structures without doing all that experimental work for many proteins then we have access to Structures for millions of proteins and that means that allows us to understand the functions a lot better. And by comparing structures to each other we can get a sense of how functions are changing across evolutionary time as well. So there's a lot of questions just about how proteins are evolving that we can approach in more detail from different angles when we have the structures rather than just the sequences. And experimentally we didn't have that many structures, we had something like 200,000 structures. But computationally, we have 200 million structures from alpha fold and 600 million structures from a related method called ESM fold. So sticking that together with some other methods, we have something like a billion structures. Whereas just a few years ago, we only had 200,000. So we have this whole new window onto protein evolution, which I think can help us answer many different questions. And we're kind of just starting to get into thinking about what kinds of questions we can answer now that we have all the structures we want. Zachary Ardern (36:28.228) roughly. Polymath World (36:29.25) Yeah, and these models are incredibly accurate. I mean, some of them score amazingly highly. Like CASP, the CASP competition held every two years. The team I'm a part of at Reading builds models that enter CASP. In fact, know, shameless plug, they beat AlphaVold at a couple of things, like some of the loop and coil regions. Zachary Ardern (36:53.286) Mm-hmm. Polymath World (36:55.674) Multifold actually outscores AlphaFold, but not a lot of people know that. know, AlphaFold got the Nobel Prize back in December and deservedly so. In some ways, this is simply a byproduct to the fact that gene sequencing is so cheap and accurate and easy now, isn't it? I mean, people can do it in the field in a handheld device. So we've got this data revolution now. Zachary Ardern (37:15.09) Mm-hmm. Polymath World (37:22.348) in biology that you're really able to reap the benefits from. Zachary Ardern (37:27.312) Yeah, and I think one of the various interesting features of alpha-phode is that it's actually based on evolution. That alpha-phode works because you can compare proteins to each other. And you can see that when this amino acid changes, this one also tends to change, which requires what's called a deep alignment. It requires basically thousands of sequences that are slightly different from each other. And that's only possible because these sequences are from related organisms. So evolutionary biology is often seen as a very theoretical discipline that's not really related to real functional biology. But alpha folders is used by millions, I think probably millions of biologists, at least hundreds of thousands of biologists around the world for very practical purposes like drug design or trying to work out the function of proteins that they're interested in. And that tool only works because the input is the sequences that are quite similar to each other and they're similar because they have a common ancestor. Some of them relatively recently and some of them very ancient, like the ribosome is the ancestor of the ribosome that was shared between bacteria and humans was something like three or four billion years ago. So that's just a plug for evolution. Polymath World (38:48.518) Yeah, yeah, that was extraordinary. Yeah, yeah, definitely. Well, we'll get into that more in a sec. But you mentioned drug discovery. I think drug discovery is usually the first thing I mention when people ask me, well, you know, what is alpha fold or computational biology going to change in the real world? said, well, it could transform drug discovery if it gets to very close to 100 % accuracy every time. But for the man on the street, I think your work is interesting for its own sake, but for the man on the street who doesn't know, what are the real world applications of the kind of work that you do? Zachary Ardern (39:27.867) Yeah, so microbial evolution is really relevant for, in particular, because many microbes are pathogens, so they're bad for human health. And some of these are evolving quite fast, so they're changing quite fast because of, for instance, very high use of antibiotics by humans. So that's exerting a massive selective pressure on very many thousands of different bacterial species in our guts and in the environment. And that means these bacteria are changing in response to the antibacterial drugs that we've produced. And if we want these drugs to work in future, we really need to understand how bacteria are changing in response to them, how we can limit that or how we can respond to that. So that requires detailed genetic work. But if we want to understand the mechanisms of change, then it requires evolutionary genetic work. And I'm doing, not all of my work is evolutionary. Some of it is just trying to work out how much is this gene expressed? What's the expression correlated with? Trying to get clues on the function of some genes that we don't know what they do. A lot of this won't be directly useful, but it might be that we discover the functions of particular genes that are related to antibiotic resistance, for instance, or related to a key aspect of the cell that we could target. in trying to produce new antibiotics. So my work is fundamental science. It's trying to understand how the bacterium works and also some work on viruses, kind of similar. But definitely one of my hopes is that one of the downstream consequences of a better fundamental understanding is that we'll be able to produce better drugs against harmful bacteria or be able to use strategies to limit evolution in directions that's harmful for human health. I'm interested in biotechnology in various ways. There's some potential applications of these overlapping genes that I mentioned. You can actually, if you can force genes which aren't naturally overlapping, but you can design them to force an overlap, then overlapping genes have some interesting properties that they're evolutionarily constrained because you have two genes in one position. If one changes, it limits, it's limited in Zachary Ardern (41:49.328) in its change because change in one could damage the other gene. So overlapping genes are an interesting synthetic biology area that I'm doing some work on as well. And the potential uses of that, could be used in gene therapies. If you're trying to introduce genes into human beings, might not want them to change very much. For instance, they... could be used in what's called gene drives. So this is where you're trying to insert a gene into an organism, and there's some evolutionary pressure against that. But using overlapping genes could be useful. So yeah, there's various kind of synthetic biology applications of some of the work that I'm doing. Polymath World (42:32.11) Yeah, I've said and written before, Symbio is going to change the world and is changing the world in ways that the average man on the street doesn't know, but should be very excited about. What are the big questions in your field going forward and any potential new areas of research for you? Zachary Ardern (42:53.467) Yeah, so I think I'm mostly interested in the fundamental wave of the strain biology. So really understanding like what's the interplay of mutation and selection and what shapes mutation? Like why is it that some effects of mutation are much more common than others? So this is called phenotypic bias or the genotype-phenotype map. So like which changes at the sequence level have which effects at the, what's called the phenotype level, the, or kind of the function level, like the shape of the protein, for instance. And I'm trying to understand some of this kind of fundamental questions that if we understand that better, we'll understand how evolution works better. There's some recent developments in evolutionary theory called the extended evolutionary synthesis. And some aspects of that kind of wider development of evolutionary theory are around questions like what's the... What's the shape of this mapping between the sequences and the structures, the genotype and the phenotype? I think there will be developments in this area for the next few years because of things like AlphaFold. We have these huge data sets that we didn't have access to. We can really understand the kind of space of evolution that evolution explores in much more detail. And that's, most of my research, hopefully in the next few years, will be focused on that. And I'll also be trying to... think through some applications in synthetic biology around overlapping genes and maybe around other protein design questions. Polymath World (44:25.548) This is fascinating because we've spoken to Simon Conway Morris and he's sort of doing the paleobiology ancient life side of evolutionary biology. We've spoken to Matt Wilkinson also at Cambridge who's really worked on the evolution of locomotion and you're doing sort of the molecular level. About a decade ago a professor at Cambridge said to me that evolutionary biology was the fastest growing science at present. It very much seems like this is why. the computational work and the kinds of questions it's answering, the sort of stuff you do seems to me to be a bit of a window into how and why it's growing so fast. It's just a shame that the general public doesn't really know that evolutionary biology is as hot as it is. So I'd love to give you the floor here to finish. Why should students listening to this pick evolutionary genetics as their field? for their future. What's great about it? Why should they get into it? Zachary Ardern (45:30.257) Well, yeah, I'm not sure if they should pick it as their field, but I think they should understand it. If you're interested in biology, you should understand the processes that led to the organisms that we have today. So there's a quote that I like that I wrote down from Richard Owen, who's reasonably well known as an early opponent of Darwin, or least this is how it's framed. So he was a paleontologist. doing work, I guess, somewhat similar to what Simon Connolly Morris is doing, but not from an evolutionary perspective per se. But in response to the publication of The Origin of Species, he said, I'll cut the quote short a bit, but he basically said, the inductive demonstration of the nature and mode of operation of such species producing force will henceforth be the great aim of the philosophical naturalist. So he's saying, after reading The Origin of Species from Darwin, this is what biology is about, basically. The aim of biology is going to be trying to understand what causes the origin of new species. He didn't think that natural selection played as big a role as Darwin did. But regardless of what that process was, even this kind of early skeptic of Darwinism was saying, yeah, this is the next big thing in biology. And I think that's still true. We see it on the small scale with experimental evolution, which seems very theoretical. But we also see it on the small scale with the pandemic. the SARS-CoV-2, the COVID pandemic. We heard about new variants, we heard about genome sequencing, we heard about phylogenetic trees. I think that was actually really useful for evolution biology for people to get a sense that this stuff is real and has real impacts, that you can see sequences changing and that has an effect on how fast the virus reproduces inside the human host or how well it spreads or other factors. So there's like small scale, like the response to pathogens, but also really big scale, like why do creatures have the features that they do? Why is it that mice are quite similar to rats and a bit less similar to rabbits and a bit less similar to dogs? And why is it that we can use mice as model organisms for human biology, even though mice and humans are in some ways quite different? yeah, evolution helps us to understand the similarities and differences between organisms. Zachary Ardern (47:55.016) which is one of the main kind of bases of biology. Polymath World (48:01.338) Yeah, absolutely. That's brilliantly put. If people want to find out more about you or your work or your research, where should they go? Zachary Ardern (48:11.092) Yes, I have a website, www.zacharyardun.com, and that links to some of my research and some of my writing. Yes, I'm trying to develop this area of research. I'm interested in various related questions around philosophy, around theology, trying to piece together a picture of the world that makes sense in light of modern biology. Polymath World (48:37.038) Excellent great. Well, maybe we'll have to get you back on again to talk into that a little bit Maybe dive into some of the philosophy and philosophy of biology too. Dr. Zaccard and thank you so much It's been a pleasure and thank you for explaining your field so well for us today Zachary Ardern (48:52.998) It's been great to be with you, Sam, and good luck to any students who are listening.