31. Reimagining Genomes w/ Jef Boeke

In this episode, we’re joined by Prof. Jef Boeke, a pioneer in synthetic genomics. Jef shares his journey from uncovering retrotransposons in yeast to leading the ambitious Sc2.0 project, an international collaboration to design and build the world’s first fully synthetic Saccharomyces cerevisiae genome. Along the way, we delve into the groundbreaking science, the collaborative spirit of synthetic biology, and what it takes to push the boundaries of genomics.

For more information about EBRC, visit our website at ebrc.org. If you are interested in getting involved with the EBRC Student and Postdoc Association, fill out a membership application for graduate students and postdocs or for undergraduates and join today!

Episode transcripts are the unedited output from Whisper and likely contain errors.

Transcript

0:00

Hello and welcome back to EBSC in Translation. We are a group of graduate students and post-docs working to bring you conversations with members of the engineering biology community. My name is Ice, a postdoctoral fellow in the Collins Lab at MIT. And I'm Will, a postdoctoral fellow in the Baker Lab at the University of Washington. Today we are talking to Jeff Bucca from the Institute for Systems Genetics at New York University at Langan Health. Professor Bucca is the key leader in the synthetic genomics field and we are very excited for our discussion today where we will touch on several recent developments in the field, especially the recent Synthetic Yeast Genome Project or SC2.0, which is an international collaboration across the global scientific community. He's also led several ongoing initiatives in the field, such as the Human Genome Project Right and Dark Matter Project. Yes, and to get started with, we would like to start with a general question. Professor Jeff, can you tell us the story on how you get into this scientific career, like especially in the field of the synthetic genomics and like so far what continues to motivate you in your research direction up until today? Sure, Will, thank you. Great to be here. And of course, it's always fun to reflect back on how I got here. I went to a small college in Maine, Bowdoin College, and you know, it's a liberal arts college and we were encouraged to learn about a broad range of things. And when I went there, my plan was to become an ecologist or an ornithologist or something along those lines. But when I arrived, I realized that the faculty who were working in those areas were reaching the end of their careers. And there were a bunch of young Turks running around doing biochemistry, which I'd never really considered working in. And so I learned a lot about laboratory science, I began working in the lab of a new faculty member, Bill Steinhardt, who had worked both in virology and in plant biology, but more plant molecular biology, but he was an avid botanist as well. And so that was quite inspiring. And then I actually, after graduating from college, I applied for and received a Thomas J. Watson Fellowship to drive a VW bug from Maine to Bolivia and collect plants in the Andes for a year. So I actually have several thousand herbarium specimens scattered around the herbaria of the world. And the principal collection is at the New York Botanical Garden. And then I went on to graduate school, basically in molecular biology and genetics and at Rockefeller. At the time, the model system of choice was bacteriophages. And so I joined a phage lab, Peter Modell and Norton Zinder were my mentors. And I did a lot of work on phage, using phages vectors. And I discovered that I loved fiddling with DNA and building things and designing things. So I sometimes call myself a frustrated engineer trapped in the body of a geneticist, because I really love both of those things. I love the engineering, but I really like the discipline of genetics and the way of thinking of genetics. And so anyway, over my career, I then did a postdoc at MIT in the Whitehead Institute with Gerald Fink. And that's where I discovered retrotransposons, which are DNAs that live inside the genomes of various organisms, from yeast to humans. And we figured out when I was a postdoc, we basically figured out the life cycle of these retrotransposons that they make an RNA copy of themselves that gets turned into DNA and then inserts into a new location in the genome. So that was super exciting. And it was kind of a dream scenario, because after basically two and a half to three years of a postdoc where I really wasn't sure if I was going to get a good publication and a lot of frustration and setbacks, all of a sudden, you know, we had the cover of Cell twice in one year. And I applied for all these jobs. I got a million job interviews and ended up going to Johns Hopkins where I started my lab and worked intensely on yeast transposable elements for the next 10 years. That was my major focus. But I am very much an interdisciplinary type. And so the transposons were kind of the perfect vehicle for me to explore a lot of things about genomes and chromatin and along the way, technology needed to be developed. So I discovered I really enjoyed that we had a very exciting multi team collaboration around the sort of second phase of the yeast genome, not synthesizing it, but knocking out all six thousand genes and making a systematic resource, a collection of knockout mutants together with Mark Johnston, Ron Davis and others. It was, I don't know, probably 20, 20 collaborative teams. But we took on a very large portion of that project. So that was my first taste of real production genomics. And I didn't so much love the turning of the crank part of it. I was really amazed at the outcome and how so many people were able to use the resource and discover a lot of interesting things. I just got really interested in the microarray technology and barcodes and all of that. And yeah, so that was the foundation. So I guess I think of myself as a bit of a serial innovator and constantly on the lookout for new areas to go into. And so I had been promoted to full professor and I'd been at Hopkins for a while. And then I had an opportunity to start an interdisciplinary center there. We hired three assistant professors and one of them was Joel Bader, a biomedical engineer. And I had another colleague, Dr. Chandra, who had been on the floor with me when I was an assistant professor in my original location. He was a postdoc with Hamilton Smith. So Hamilton, Ham, of course, has made phenomenal contributions to actually synthetic genomics, although at the time he wasn't doing that. He was just getting into sequencing and so on himself and making shotgun libraries. So I met Dr. Chandra at that point. And then about 10 years later, I had this profound conversation with him in the coffee shop where I had heard Ron Davis at an earlier meeting say, someone's going to synthesize a chromosome, a yeast chromosome. And I remember thinking, why would anybody want to do that at the time? But I mentioned this to Chandra and he got super excited. He said, oh, my God, we've got to do this. And I said, yeah, but why would we want to just synthesize, resynthesize something that already exists in nature? We need to do something much more important and we need to make changes to it. But the problem is a lot of the changes you want to make, it would be risky and you might make a dead yeast and that wouldn't be that interesting. And so we brought Joel in on it and on the idea and talked and talked about it. And we had many, many discussions. And over a period of about a year, we came up with this game plan for a list of changes we wanted to make to the yeast genome. And I knew I knew from the very beginning that we would be able to do it, that the yeast is so good at doing homologous recombination that we could harness that and we would be able to introduce the synthetic DNA, whether or not it would function, we didn't know. And so there was a certain probability that the whole thing would like crash in the early going. But so we had this very long discussion about balancing the risk of making changes with the sort of the amount we could learn from those changes should they be acceptable. So we wanted to be bold and make really interesting changes, but we didn't want to make dead yeast. And so we came up with this plan. And essentially, 100% of that plan has withstood the test of time. There was there was one time when we kind of had to pivot. And that was regarding one of the things on our list of types of changes we were going to make, which was we had decided we were going to remove all of the introns. So your listeners may or may not know that of the 6000 yeast genes, only about 5% of them contain an intron. And the vast, vast majority of those have just a single intron. So there's no such thing as alternative splicing in yeast, as far as we can tell. So and another lab had done very nice work. Abu Alel's lab in Quebec had done very nice work showing that they could delete the vast majority of introns as long as they deleted it precisely. The yeast had, you know, didn't care. There were some exceptions. There were some essential genes that when they removed the intron, it was lethal. But what they showed was that in those rare cases, if they fiddled with the expression level, they could bring back the viability. And so since they had done a fair amount of work on this, making individual intron deletions and we're still very much guided by that work, we decided to leave in the ones that were tricky and take those out at the end. And then we could make adjustments to expression as needed. Because one of the things we really, questions we really wanted to ask was, can you build the yeast with no introns? And then if you can, can you delete the whole splicing machinery? And if you do that, would you still have yeast or would it be something else? So we still are not at the point where we can ask that question. But we've made huge strides towards getting there. But the pivot part of the story is that after that first paper came out from Abu Lelela's lab, about three or four years later, they did a second paper where they kind of walked back their initial claims and they said, well, it turns out when you delete the introns from the ribosomal protein genes, it's not so simple. So ribosomal proteins are typically encoded by two genes in yeast, and they almost all contain an intron. And what the claim in the second paper was that when you deleted an intron, you often saw effects either on that gene or on the paralogue, the other copy. And they were complex and unpredictable, and they sometimes led to a fitness defect. They weren't lethal because we knew from earlier work that deleting one copy of these duplicated ribosomal protein genes is typically not a lethal event, but the yeast grows slower. And one thing that we were really careful about in all of our planning and so on was we worried about this Mueller's ratchet that if you have a small change that reduces fitness a little bit and then another small change that reduces fitness a little bit and a little bit and a little bit, pretty soon it ratchets down to the fitness to the point where you you're dead. So we did not, you know, that this was a huge concern for us. And so we decided, although we had by that point already designed and built several small chromosomes and taken out all the ribosomal protein introns, we decided to defer removing the introns from the remaining ribosomal protein genes to the end of the project. So now several of our big chromosomes have been made in two versions, one that retains the ribosomal protein introns and one that that removes all of them. And so far, we haven't really seen big fitness effects on those. So we're hopeful that we'll be able to do that. But that's how that's how I came to, you know, lead this international synthetic yeast project. Well, that's perfect. That's such a great overview of your work. It's so interesting that you actually did a really great job covering a lot of the questions that we had lined up. We're kind of curious because you've been continuously involved in this synthetic yeast genome project, like you'd mentioned, talked about some of the project's goals and some of the challenges associated with it. Maybe you can expand a little bit more on some of the potential impacts that you think it might have on our future engineering biology landscape. Sure. Sure. Well, what I didn't do was tell you about all of the other kinds of changes that we engineered into into SC 2.0. And so some of them are somewhat mundane. So we developed a sort of water marking system that we needed to be able to distinguish newly synthesized DNA from native DNA because our our our game plan was not to do a venturesque wholesale replacement of one big giant piece of DNA, you know, replacing the genome in one step because we thought that was too risky. And so our strategy was to replace. Now we do 60 KV, but initially it was 30 kilobases at a time. And that also happens to be a DNA segment size that's quite straightforward to manipulate for us, at least, and really quite routine. So when you do that, so we developed this method we call swap in for switching oxytrophies progressively for integration, whereby we essentially overwrite, say, base pairs one through 60,000 with a synthetic counterpart of that sequence. And when you do that, you harness the power of yeast's endogenous innate ability to do homologous recombination, as well as for the first segment in to grow a new telomere from a little stub of a telomere that we call a universal telomere cap. Sort of a mini telomere that we've now put at all 32 telomere positions in yeast substituting for the native telomere. And speaking of telomeres, sub telomeric regions in yeast have an abundance of repetitive DNAs, not just transposable elements, but also some highly repeated protein coding genes of still unknown function, some medium repetitive genes, some of which have a known function, some of which don't, and some other repeats that don't necessarily encode proteins like the Y prime sequence. And so one of our rules that we came up with was all that stuff's got to go. So one of the largest pieces of these genome that we've thrown out, and again, this is mostly based on work that had been done on a smaller scale by other people, they had shown that you could delete this and there was no impact or minimal impact. So we came up with a strategy for deciding where the sub telomere ended and where the body of the chromosome arm began. And we, in some cases, that was a relatively small amount of DNA, just, you know, 10 KB or less. But in others, it was many tens of kilobases, multiple reading frames deleted. So the incoming DNA has the stub of a telomere, the sub telomeric stuff deleted, and then it has about a thousand base pairs of identity to the native chromosome at the other end and a whole bunch of synthetic DNA in between. Now when you do that, what you hope is going to happen is it's going to grow a new telomere and it's going to undergo one crossover event in the internal part of the chromosome. And most of the time that is what happens, but not 100% of the time. A lot of other kinds of recombination can occur depending on the details of how you do it. And then as you move, once you get that first piece in, then you're ready to put in the second piece and then we put homology at both ends because there's no telomere anymore. And that's essentially how we move from left to right through a chromosome and overwrite one segment at a time. But as I said, at each of those steps, things can go wrong. Maybe the sequences that you chose are not 100% identical and they have some other targets somewhere else in the genome, so you get some kind of weird rearrangement. Maybe the native chromosome duplicates itself and you have a partially synthetic chromosome in there. That's actually pretty common. So you want to have a system that allows you to rapidly screen the colonies for the presence of the new synthetic DNA and the removal of the target DNA that you're discarding. And so we developed this system we call PCR tagging, which it's a PCR-based strategy that's very easy to deploy. You can do it using gels, but you can also do it with more sophisticated automated RT-PCR type procedures if you like. And it's very good and it's a very cheap way to check a whole bunch of colonies and find out which one is correct. Now most of the time, you pick whatever, 12 colonies, 6 to 8 of them are going to be correct, but not always. Some regions, for example, when you get close to the centromere, we found consistently that that segment is harder to get the replacement, but you can get it probably because the centromere has to pick up all the proteins that recognize it, the kinetochore. So the water marking system is this PCR tag and the way it works is we put the PCR tags in every open reading frame, we use recoding, look at the codon usage table, and we wrote a piece of code that looks for the protein sequences where there's the most flexibility in terms of changing codons. So leucine, arginine, and serine all have 6 codons, and therefore it's actually possible in some cases to replace all 3 bases of a codon and still encode the same protein. So the regions that are high in leucine, arginine, and serine codons are identified, and then we look for another one that's within a few hundred base pairs away that has similar properties, and then we can make two sets of primers, one that's perfectly identical to the wild type sequence and one that's perfectly identical to the recoded sequence, and then we use those and get the expected results. And not every primer pair works, but the vast majority of them work, so it's turned out to be an extremely reliable way to check these things. All right, so that's the mundane, right? But very important. We talked about removing the introns. Another thing we did is to remove repeated DNAs from the genome. And now yeast's genome is not that repetitive. There's just a few percent, maybe 8 to 10% of the DNA is considered repetitive. That's accepting the ribosomal DNA, which is very highly repetitive and itself makes up about 10% of the genome, and that we did not change. That is still there in the normal copy number. But all the other types of repeats, there was reason to believe we could remove them with impunity, and so far that's proven to be the case, with one special exception that I'll mention in a minute. But the transposons, of course, which was one of my original interests, was, wow, we could build a yeast that has no transposons, because this was always the discussion in the bar at the transposon meetings. Do you think we could make a eukaryotic organism that has no transposons? And we now think that that question's already been answered by nature, because there are a few organisms that seem not to have any transposable elements, as far as we can tell, but they're a few and far between. Most organisms' genomes are just littered with these things, like our own, 50% basically repeats. So this was a big thing, and then it turned out that actually our own work, but also work of others, years earlier, had shown that these Y elements, as they're called for transposon of yeast, the vast majority of them are associated tightly with transfer RNA genes, TR, okay? And the reason for that is still not 100% sure, but it has been shown that TRNase, which are super highly transcribed, probably the most highly transcribed sequences in the genome, they even have their own special dedicated RNA polymerase, RNA polymerase III, they somehow silence nearby regular open reading frames, you know, Paul II transcription units. So it's not a super well understood mechanism, but it's very clear that there's this sort of negative effect on nearby protein coding genes. And so as a result, particularly on the five prime side of tRNA genes, there's typically this gene free gap, okay? And what you often find in that gap is these transposons. So we think that the transposons know, quote unquote, that this is a safe place to jump where you're not going to break up a gene. And in fact, our earlier work showed, one of my early graduate students, Eric Bolton, showed that when you put a TY into these gene free regions, it doesn't interfere with the expression of the tRNA gene. Sometimes it actually modestly boosts it. So based on all that, we thought it was very likely that we could delete tYs with impunity. And that has totally proven to be true. Now the other type of repeat that is actually a gene is the family of tRNA genes. So as you know, there are 61 codons that specify amino acids, three stop codons, right? And those 61 protein coding codons are recognized and translated into 20 amino acid. It's a degenerate code, but it isn't that there are 61 kinds of tRNA. There are, depending on the organism, somewhere between 61 and 20 tRNAs that have the ability to recognize in some cases more than one codon through Wobble, the third base of the codon, first base the anticodon. And I can never remember what the number is in yeast, it's like 34 or 42 or something like that, different types of tRNA they call iso acceptors. But there are 275 tRNA genes altogether, and they're scattered willy nilly throughout the genome. So one of the things that we came out of our one year debate about what to do about the synthetic genome was, well, gee, you know, God kind of made a mess of the yeast genome, you know, interspersing all these Paul III and Paul II transcription units. Can we maybe consolidate all the tRNAs in one special chromosome and leave, you know, clean up the other chromosomes so they're all protein coding. It's also interesting from a technology point of view, because then it could make it much easier to really engineer the whole cadre of tRNAs if they're all together in one block. And so that's where the whole concept of the neochromosome was born, that we were going to add a 17th chromosome that was going to have just tRNA genes on it. Yeah. So the T2 has come to pass. So that's been built, and Patrick Tsai, a former postdoc, and his lab really took the lead on that. We sort of piloted it and showed how it could be done in such a way that we didn't have any scraps of elements left around. And so there was a whole lot of careful engineering that had to be done, which was a bit idiosyncratic. So we use this software that Joel wrote, and Joel's students and postdocs wrote, called BioStudio, that enabled us to do certain steps in a sort of blanket way. So for example, we changed all the TAG codons to TAA, inspired by George Church, Farron Isaacs, and others, that we were pretty confident that doing that would be okay. So that's very easy as a kind of one-click edit of a genome to do. Similarly, as I'll talk about later, we inserted LOXP sites at strategic positions in the genome. That was also something we could do as a so-called global edit. And then we wrote code to do a global edit to create all the PCR tags. Those are things that can be done as a global edit. All the other editing was done by a human sitting at a terminal and guided by BioStudio. So okay, where's the next tRNA gene? How much DNA should I delete? Because the DNA surrounding it is full of retrotransposons. So typically we go from the nearest gene up to the first transposon boundary and then through either the tRNA or if there was also a tRNA, a tY on the other side, you'd go through there and we'd just chop that whole block out. And that's something that only kind of an expert can do, be very hard to program that because each situation is a little bit different. So it's very much a semi-automated process, the actual design and engineering and some of it we kind of made up as we went along. So anyway, that's how we dealt with the tRNAs, which are of course because there's 275 of them and they come in whatever it is, 40 different flavors, they're repeated, right? And they're perfect copies of each other. The tRNAs also have introns. We also precisely removed all of the introns. So I don't know, but if you've done your homework, you know that Patrick's lab had a great pay paper and cell where they described the design construction of the tRNA neochromosome and basically showed that it functioned. So that's complete now. Yeah. So that was perhaps the boldest thing we did. And then the other major thing was the scramble system. And that was really, that was an important thing because for me, that was the idea that put it over the top, okay, it's worth doing this because with the scramble system, we're going to be able to do things that nobody's ever been able to do before. Yeah. And so just to explain it to your listeners, what we did basically is to insert a LOX P site. This is a 34 base pair sequence that's recognized by a recombinase called Cree. And we put it exactly three base pairs downstream of the stop codon of every non-essential protein coding gene. Yeah. So at the end of the day, there's, I can't remember the exact number, but it's over 3,500 of these LOX P sites scattered throughout the genome. And original idea was we were going to put one on both ends of each gene, but whereas the three prime end was relatively straightforward because it was a very good reason to believe that three base pairs downstream of the stop codon would not interfere with anything. It's very hard to define the five prime end of a promoter. Of course. Moreover, most promoters are bi-directional and therefore, you know, you may not mess up one, but you may mess up the other ones. So we dropped the plan to put the LOX P site in the promoter and just went with the three prime end. We actually did some experiments to show that that didn't interfere with anything in a couple of model genes. And then we went with it and it's also proven to be the case that it really doesn't interfere with anything. And then when we come in with the Cree recombinase, it's kind of like shuffling a deck of cards of 5,000 cards. That's the analogy I often use. And you know, you can get millions of variant genomes in an Eppendorf tube full of yeast. And so that's really powerful. And although many of them may be dead because they've deleted an essential gene, we've come up with some tricks to avoid that particular problem, which is we now often do this in a heterozygous diploid strain where one set of chromosomes has no LOX P sites in it and it's a wild strain. So it can be any strain, it could be a pruing strain, it could be a chocolate producing strain, whatever you like. And the other one is the SC 2.0. And then the worst that can happen is you go from two copies of a gene down to one in terms of deletions, but you still have the benefit of being able to amplify the copy number of certain genes, which could give you a gain of function, which is what this is primarily most interesting for and really allows you to do a whole new kind of genetics where it's all based on the copy number of genes and not based necessarily on base changes. So that was very successful and we're continuing to work on that. We're working closely now with Jose Avalos at Princeton. He has some work going on biofuels and has some interesting results with some of our SC 2.0 strains being able to make better yields and so on. And lots of the team members around the world have been fooling around with different applications of Scramble, but most of it's been done just in the context of strains that have just one synthetic chromosome. And the real power of it is going to reveal itself when we put all the chromosomes together. So we haven't talked about that yet, but Jeremy Hsiao in our lab has shown that he's been able to pull together now, I think eight or nine chromosomes, but in the publication was seven and a half into a single strain and we've developed a series of different methods for doing this. It still takes longer than I would like, but I think we're really in the hunt for the finish line now where we're going to have all the synthetic chromosomes in one strain. And then scrambling is going to be really, I think, you know, going to shine. Yeah, that's very fascinating story, Jeff. It's so great that you mentioned not just about the science of how you are doing this, how it can lead us to, and also like how you go through these challenges. So we have one question that may be not that scientific, but more about like, you know, the management side, about like, most of the time in the engineering biology and synthetic genomics, you have to wait on whether you want to make new tools so that you can study new biology, or you have to focus on like the existing tools and just like answer those question right away. Like, and like, I want to, you know, maybe if you can mention briefly, what would you approach these properly so that you move as fast as we could do it? Maybe like in a short comment would be great. Yeah. Well, yeah, that's a great question. And I think, you know, some of it also, some of it is reflective of what you enjoy doing and what you like to do. And I confess that I'm motivated to find solutions to problems, even when there sometimes isn't an immediate application for it, just because there's a sort of a, you know, because it's their kind of challenge. Can we do this? And so, but that's just me, right? And that doesn't mean that's what everybody should be doing. That would be, you know, not great. So I think, I think it has to be a balance for sure. I mean, I also find it really rewarding when we learn some interesting biology, whether it's using new tools we developed or old tools, we don't really care that much. So I mean, a great example of something that was an unexpected use of a tool we developed that showed us some really cool biology is Weiman Zhang's paper in molecular cell from last year, where he took, he synthesized chromosome four, which is the largest chromosome and it was really difficult. He had to debug eight different problems. And the guy's amazing. He did so much of this work single-handedly. And then we came up with this idea of, let's put the centromere where the telomeres were and let's put the telomeres where the centromere was and see what happens. And so he was able to do that relatively easily. And surprisingly, there was kind of no effect. And that in a way, I sort of anticipated that because a former grad student, Jingchuan Luo, had shown a few years earlier that you can string chromosomes together end to end and they still function pretty normally. And Dr. Chin's lab in Shanghai showed the same thing working independently in a slightly different method. So I wasn't that surprised by the result, but he was very frustrated. And so then we also, we had paper was rejected and appealed and so on. And the reviewers just, they were like, ah, so you synthesize another chromosome. We're not that impressed. And so then Waymans came up with this great idea. He said, well, this chromosome has, I forget, I think 700 LOXP sites on it. What if we could use those to smash the chromosome up against the nuclear envelope? I bet that's going to do something to gene expression. So he developed this whole tool of DNA binding protein that would bind to LOXP and a nuclear envelope protein that would tether the chromosome to the nuclear envelope. And sure enough, when he did that and he turned on the expression of that protein, the genes on SYN4 got shut down. And so that was very rewarding that we finally did something to the architecture of the chromosomes inside the nucleus that actually had an effect on gene expression. Because really a lot of people just couldn't believe that we could make these massive rearrangements to genome structure and have no effect on gene expression because it kind of flies in the face of a lot of, you know, speculation about how important 3D genome structure is. And I'm not saying it's not important, but you can argue with the facts here that gene expression is not affected in a big way by global architecture in either yeast or human cells. So anyway, so that's one little vignette on that front. Yeah. Great. Well, since we've discussed the work of the project so much, we had a couple of questions actually about leading and then executing on these collaborative projects. You've had a really great history of working in things that are interdisciplinary and at times international collaborations between multiple teams. We were wondering if you had any personal techniques or strategies that you kind of used to approach and then execute those collaborations, both with more senior scientists who are, you know, familiar with the field and for younger generations of scientists getting interested and motivated to work on these types of things. Yeah, yeah, that's, that is pretty interesting topic. So two, two areas. So first area kind of led to the second. We decided early on that this would be an incredible training ground for undergrads. And so we started this course called Build a Genome, it's still going even though we're done kind of with the building phase of SC 2.0, but it lives on in many guises. And the whole idea of the course was, of course, to give students, you know, a good background in, you know, genome engineering, PCR, yeast genetics, et cetera. But the main aspect of it was they do projects in the lab, preferably on a kind of production scale where they really learn how to do a particular kind of activity, whether it's PCR and cloning and sequencing, which was very heavy focus in the beginning, whether it's transforming things into yeast and screening to find the right clone or whether it's doing scramble experiments and actually having the students give a lab meeting about it every two weeks. They get up there and, you know, show their slides. And this was before Zoom or any of that. But we figured out how to do a lot of this stuff, you know, with a common server and all the data was stored on a common server. And that was that was super rewarding for me because, I mean, the students, they just loved it. And and they worked, we made them work their tails off. It was it was really successful and I loved it. The students loved it. And it was a teamwork interdisciplinary. Joel Bader had computational track within it. And that was super successful. So so so that was kind of a mini team effort. And and then, of course, you know, we also did iGEM teams, right? So iGEM was inspirational. It was amazing. But I didn't feel like the students got as much out of it or really accomplished anything, at least most of our teams. And so but but it had, you know, so many other positive things that were great. So and during this time, Patrick Tsai was in the lab. He helped with build a genome course. And and he had been very heavily involved in the first iGEM team from China. And he had, of course, a lot of contacts in China because he's from China. And he and at one of the iGEMs that I didn't go to, he met YJ Yuan, who's head of chemical engineering in Tianjin University. And he called me up and he said, YJ is super interested in your build a genome course. And he wants he wants to come and meet with you and talk about it. So so we started talking about build a genome China course. And and he actually flew to Baltimore for like two days to meet with me. And, you know, I still remember going out to a bar and drinking with YJ. He likes to drink. So that was nice. That's very nice. And we yeah. And so he's he's sent multiple people over over a period of years. And that was the beginning. But but then Patrick had lots of other contacts. Jun Biao Dai was just graduating from my lab as a postdoc, hadn't worked on SC 2.0 at all, but got this great job at Tianjin. And he said, I want to do chromosome 12 because it's got the ribosomal DNA on it. And I said, OK, you know, I wanted to do it, too. I couldn't do everything. So I said, great. You know, and then BGI BGI was super interested. So we had all these groups and we we had a big meeting in China. We brought everybody together and we talked about, you know, how we were going to run this and that, you know, we were going to do all the design and we were going to give them the tools to do the building. And these the necessary genetics. And in a way, we kind of knew like the building part that they would be able to do it because our undergrads were killing it. You know, they did pretty easily. And so the tech transfer was relatively straightforward. You know, they all had to sign agreements agreeing to certain things like agree to distribute freely. Nobody was going to patent the strain or the the basic technology. We didn't patent scramble ourselves, try to set a good example. And but then if they came up with a better way to make the DNA, they can patent it, they can, you know, they can do what they want. So that's how we sort of structured the partnerships. And, you know, it just it it really went well, I think, because of the people we chose to be our partners. You know, there were some people who asked to be partners who we didn't decide not to work with. We had some industry partners that worked out and they had to sign the same kinds of agreements. Sometimes they didn't want to. Yeah. But yeah, it worked out well. And then Patrick also played an important role in, you know, communication with various team members and making sure people were sort of playing by the rules. So by and large, it's been very good. I mean, there's definitely been a few times when different teams have have said, oh, being, you know, how about? Yeah. And then, you know, we find out that two groups are doing the same thing. And so, you know, usually we managed to work it out. But I mean, basically, I got scooped on a couple of things by some of our guys, but, you know, I can. It's OK. So, yeah. So anyway, that's that's what I have. I think that that's a great that's a great story and a great example of division, division of labor in terms of, like, you know, doing science, however, triply internationally. Yeah, I think like the the last kind of technical technical question and also management question that we have is about like, how do you approach the ethics part of this big project? And maybe like, you know, not just the SC2.0, but also like the human genome project, right? Or the documented project is going to be big. So like, is there any like concerns on the bioethics? Right. So I think we've been very careful to include that. Like important team member of all of these teams is a bioethicist with a strong focus on bioethics. And sometimes it extends also to social scientists and others who are not, you know, hardcore molecular biologists, but interested in some of the human elements of it. And, well, so we published a whole sort of manifesto of of the on the sort of human practices, elements of it, Deborah Matthews and Anna Sleva and myself. And, you know, I think that really did set the groundwork for all the team members agreed to abide by those principles. So I think for yeast, there's really no major bioethical issue. Maybe if it comes to field release, you know, we need to engage the stakeholders and so on. But nobody's talking about doing anything like that at this point. I think where the biggest issues are, frankly, particularly in the safety arena, is when you start talking about engineering virus or other infectious microorganisms. And then there's a whole separate set of things where you're doing anything that would go towards germline modification of humans. So so we've kind of we've completely stayed away from the latter. Clearly, it is something people talk about and it's good. It's good to discuss it. But, yeah, that's that's that's how we've dealt with it. Great. Well, Jeff, thank you so much for your time today. With the last minute or two, we just wanted to ask if there was anything that you wanted to share with the engineering biology community to fellow researchers, trainees or the general audience or anything that you'd like to promote or direct our attention to to learn more about your work. Well, I think we really need to make DNA synthesis cheaper. We need we need the 10 to 100 fold drop in price that would enable us to do so much more than we can now. I think it's a it's a really major challenge for our field. And it's actually something we're working on just as of the last few weeks. I think we might be on to something. So stay tuned and also everybody else in that game, keep working hard at it because the price drops haven't been too impressive thus far. We need to be more like sequencing. Absolutely. Thanks a lot for emphasizing that. We are looking forward to to to the era where we can synthesize a bit of much like more efficient way. So with that, I think we'll end the podcast here. This has been another episode of EBSC in translation, a production of the Engineering Biology Research Consortium's Student and Postdoc Association. For more information about EBRC, visit our website at EBSC.org. If you are a student or postdoc interested in getting involved with the EBSC Student and Postdoc Association, you can find our membership application link in the episode transcription. A big thank to the entire EBSC SPA podcast team, Andrew Hunt, Ross Jones, David Mai, Heidi Clupe, Rana Saeed, Will Groobie, Matt Williams and Ais Chumpisit Kedisevi. Thanks to the EBRC for their support and of course to you, our listeners, for tuning in. We look forward to sharing our next episode with you soon.