SIO481: Ok What Was the Deal With the Direwolves

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:12 Thomas: Hello, and welcome to Serious Inquiries Only. This episode 482. I know month ago that dire wolves are back. Didn't know they ever existed. I thought they were fake. Just pretend. Other animals that I thought were pretend are extinct? Unicorns? Pegases? Pegasi? Bears? You know? What else is actually real and nobody knows it? What's the deal with this -unquote de -extincted animal, as the scientists responsible for it claim? That real? Is it a hoax? someWhere in between? Might have seen mixed things. Well, I've got an expert in Dr. Eric Jaffe to break it down for us. He's the perfect expert for this. And I love this one because I don't know what to think. Imagine many of you might not know what to think. By the time this is all done, we're going to know exactly what to think about this claim of the de-extincted dire wolf. Is it BS? Is it awesome and real? Where in between? What other implications does this have? What true, what other things might might be de -extincted? How could this technology be used? So all those questions and more. So excited. Such a great science -y episode. We'll take our usual break and please support the show at patreon.com/seriouspod. You can skip the ads. You Can Feel a And you can get that warm, fuzzy feeling. Right, after this break, I'll be on with Dr. Eric Jaffe.

0:01:50 Thomas: I'm joined by Dr. Eric Jaffe. How are you doing today?

0:01:54 Eric: Doing well. It's Teacher Appreciation Week, so I've gotten a bunch of really nice letters from some of my students, you know, expressing said appreciation. So, feeling in a good mood. Summer's right around the corner, ready to kind of relax a bit.

0:02:05 Thomas: Also appreciate you. Want to say that.

0:02:08 Eric: Is podcasting if not audio teaching?

0:02:10 Thomas: Yeah, I don't have like a Starbucks card for you or anything, but I appreciate you, because this story is, it couldn't be more you. I mean, people I'm sure saw headlines about the possible return of a dire wolf, which I, okay. I'll confess. I didn't even know that was a thing. I Didn't know it was a real thing. It was a made up thing. It was really animal. But anyway, this is everything that like you've already talked about trying to bring back dinosaurs and we've done all that. And now we've got some company claiming they brought back a direwolf. This couldn't be more you. I'm so happy to have you to break this down. Do you want to start with, I don't know, what is a dire wolf?

0:02:14 Eric: I think we're getting Chipotle lunch tomorrow, so it's all good. Just want to start on this theme that we'll get into about overselling. It actually works even better than this, right? Because this is actually going to touch on literally everything we've talked about in our, for lack of a better word, series, right, we've got that de-extinction coming about, we've got it in response to a current mass extinction that we're facing now. And it's utilizing those principles of genetics and complexity to use CRISPR to rewrite genome. So it's kind of going to tie a bow on everything. And of course, it's headed by Dr. Brett Weinstein. So we've GOT all the pieces of the puzzle. No, No, I'm kidding. The person who heads this group is actually a very, very prominent and famous scientist. This is not a Elizabeth Holmes situation. This is there's some really exciting science and kind of unpacking what's the science and what's the promotional aspects of business is kind of part of what this podcast is going to do.

0:03:17 Thomas: What? Dear God. Wow. Okay. A direwolf?

0:03:38 Eric: Yeah, I started my outline in exactly the same spot, because if you're going to talk about de -extincting an animal, you should probably understand the contours of what that animal is. And a dire wolf, for any of Game of Thrones fans, is an ancient wolf that lived from about 125 ,000 years ago to about 10 ,000 years ago, And right? It was prominent mainly in kind of that mountain west, southern United States. They had a pretty wide range of North America, but they didn't go above 40 degrees latitude, right, so they were strictly sub, kind like of of that boreal forest line

0:04:11 Thomas: Real sticklers for the rules, I guess. Huh? Okay.

0:04:14 Eric: It's kind of more like the tree line in mountains yeah it's more about like the angle of insulation that the sun gives you the seasons and the climate that that comes across and

0:04:16 Thomas: That makes more sense. Didn't know this was a real animal. How is it different than non -diar wolves, the more optimistic wolves?

0:04:28 Eric: Even that's a slightly more complicated question than you would think because biology doesn't like labels right and so like again this is going to be us kind of putting labels around things that don't necessarily fit our clean definition of stuff and there's actually some con there's controversy but there's some question about where exactly dire wolves fit in this evolutionary tree right in general you can think of dire wolves as like having diverged from wolves roughly six million years ago right so this is an ancient group of canids canids themselves right our dog -like animals split away from cats about 40 million years ago right so carnivores our dogs and our cats for mammals is a pretty old clade dating back to the early Cenozoic, you know, only 10, 15 million years after the asteroid took out the previous occupants of most of their niches. Then really radiated once they got to North America fairly quickly, right? And so we think of arcanids, right, our dog -like animals as in its most ancient grouping, foxes and then our are wolf -like animals, right? And then within that wolf -liked grouping, you have, after the divergence from foxes, our coyotes, our jackals, our deholes, which don't really, aren't that super prominent in North America, but are more Eurasian. Then obviously our wolves, which have that worldwide spread. Dire wolves, depending on which evidence you use, fit somewhere either in the canid line or more likely, as we'll see from both this paper that we're going to talk about and DNA evidence. Fit kind of in an out group with wolves. So they have a common ancestor of wolves that's also similar to a common ancestors that diverge from jackals and coyotes, but they're not quite part of that modern wolf, if that makes sense.

0:06:10 Thomas: Suppose. I don't know. Is it going to look like? You if I saw one of these now, what Would I think it was?

0:06:13 Eric: That's a great question. And if we were to take Colossal Biosciences, and we should probably start there. This is a project that is is brought about by Colossal Biosciences, a company that started in 2021 and is founded by George Church. Now, I joke that Brett Weinstein was the founder of this company. This guy is not Brett Weinstein. I am not going to be debunking his science today. I'm not going to be talking about how he's a grifter. We're not talking about an Elizabeth Holmes at Theranos that's selling snake oil. This guys is kind of one of them ones when it comes to biology, right? Right? He's really known for a couple of things. And I think this will give a scale of just how legitimate this guy is. First of all, he works at Harvard in the Broad Institute as a professor, which is obviously a very distinguished position. He is a National Academy of Sciences fellow. He is as decorated as you can be without actually winning a Nobel Prize. And he helped found the Human Genome Project. He helped induce pluripotent stem cells, right? Right? He helped isolate embryonic stem cells for the first time. He's actually one of the first people to develop a technique to sequence genomes directly, right? Instead of kind of cloning DNA from the organism into like bacterial chromosomes we call plasmids, he was able to take the DNA directly out of the animal and sequence it directly. So we're talking like a truly foundational person. I

0:06:40 Thomas: Well, that's nice. Wow. That's nice, it's nice that it's not just all fraud and hucksters for once. Appreciate that.

0:07:36 Eric: No, I was trying to think of a good comp like if the Beatles were like Charles Darwin right we might not quite be talking like a Pink Floyd level but we are at least at like a Ramones you know someone that has lasting impact is a true pillar in the industry so this is his project that he founded along with a hedge fund entrepreneur person that you know provides the backing but their goal is de -extinction in general we'll get to the dire ones I promise but I think it's important to kind talk about what the goals is, what it is and what it's not. They define de -extinction is, and it's really funny. In their website, they say, everyone is talking about de - Extinction wrong and have the wrong definition for it, which always just makes me think of The Simpsons, where it's like, no, no, it's not the kids that are wrong, right?

0:08:21 Eric: But they define de -extinction slightly differently. They're not thinking of it as truly resurrecting an extinct animal, not to give away the punchline, but they think of it at the process of generating an organism that both resembles and is genetically similar to an extinct species by resurrecting its lost lineage of core genes. Does that sound more like to you? Cloning the dinosaur Jurassic Park style or the how to build a dinosaur that we talked about in the very first podcast.

0:08:21 Thomas: Uh, I guess the, like how to build a dinosaur thing.

0:08:50 Eric: Exactly right. This is more about taking viable DNA that already exists and reprogramming it to give it more ancestral features. Right. That's why I want to give that quick note before we talk about what the difference is between a wolf and a dire wolf, because really that's what they're trying to do. They're trying Tofino copy by that. I mean, They're trying to make a regular gray wolf kind of match the phenotype, match the characteristic and traits of a dire wolf.

0:09:17 Thomas: The difference is you don't have to find a mosquito basically. You

0:09:20 Eric: What they did actually was they used and we'll get into it a little bit later. They used actual dire wolf fossils, pulverized them and then extracted DNA from those pulverised fossils. That becomes important a little later. It is really cool. And that's what I mean is there's going to be a lot of really cool techniques we talk about. Before we get there, we should set the groundwork. A dire wolf is very similar in appearance to a wolf, but they are a little bit stockier. They're on the higher end of the weight range. We're not talking about completely non -overlapping size ranges, but you can think of dire wolves as like the top 20 % of your gray wolves, right? Right? Like they are the bigger, bigger ones.

0:09:56 Thomas: The wolves that can play in the NFL of the wolf NFL.

0:09:59 Eric: Exactly. Kind of more like the way that like the stockiest wide receiver will bleed into the lightest tight end, right? On average, a tight end is going to be heavier than a wide receiver, but there's enough overlap there that you can get that kind mixing of traits, right. What's interesting is prior studies on the the dire wolf actually did not see a lot of what we call admixture, right? Of mixing hybridization with other canid species. So they were thought to be pretty genetically distinct and isolated, which matters because in the time that they were around in that late Pleistocene of North America, there were a lot different canids around, right. We had not quite wolves coming in fully yet, or at least not the ones that we commonly have today, but we had jackals. We had some of those d -holes, we had coyotes, right? We had those different kind of canid species that could mix with the dire wolf theoretically, but we didn't see a ton of evidence of that prior to the preprint that we're going to talk about. So they were kind of their own distinct clade, right, they're not actually, if a wolf is canis lupus, they are not actually of the genus canis.

0:11:05 Eric: Are their own genus. Even that came about with some controversy, And I think it's kind of fun to talk about that controversy because it also, I always like to kind of highlight those basic biological principles. It highlights one of those basic biological principles. Controversy came from when we first found fossils of dire wolves, morphologically, the way they're shaped and everything was very, very similar to our canis wolves, right? Our common current wolves, our Eurasian wolves, their gray wolves, things like that, right. So they thought they were much more related and of the same genus. They had similar bites in dentition. If you remember, dentition in mammals is a very big predictor of relatedness, right? There were all these things that suggested that they were actually pretty closely related, and that's why they were called dire wolves at first, right, they look like an extinct species of arcanist wolves. But DNA evidence, which works slightly differently, showed that they were, or at least suggested that they were going to be more distantly related. And the difference between those is with fossils, you're just looking at the similarity of appearance, right?

0:12:05 Thomas: Yeah.

0:12:05 Eric: But there's this thing called convergent evolution that can drive species that have similar roles in the environment to end up with similar similar morphologies, right? You can think of this as like a pterodactyl and a bat are not very closely related, but their wings have very similar structures, right, and that's convergent evolution finding a similar answer to a similar problem, even if it wasn't a solution that was cracked once and then shared among all the descendants, which is what we mostly think of when we think of traits, right? Placental mammals give birth to live young because, and I'm going to to give a throwback word here, that basal mammal that placental mammals evolved from gave birth to live young and all of the other mammals inherited that trait as part of what we call the ingroup, right? They all share that trait, marking them as mammals.

0:12:54 Thomas: I have to imagine back when the only method they had was collecting old bones and seeing like, well, this probably looks like it goes here, you know, and I'm sure it's more sophisticated than all that. But like, after you imagine with the advent of DNA and all that, that has to have changed a lot where it's like, wow, we thought this was this thing, but it's actually not even closely related to that.

0:13:14 Eric: We've used to classify organisms through what we call Linnaean taxonomy, right? This guy, Carl I believe Swedish biologist back in the day, really naturalist because biology wasn't a field back then, started categorizing animals and plants based on their morphology, right? Their shared characteristics. We've had to reorganize a lot of that. For instance, as we talked about in the first podcast, we had to move birds from their own distinct order into a subgroup of dinosaurs, right.

0:13:41 Thomas: Crazy.

0:13:42 Eric: That's one example of how DNA evidence kind of overturned our previously held thoughts. And I think it's clear from the DNA evidence that our original taxonomy and pardon the pun was barking up the wrong tree, right? The family tree of wolves is actually much more complicated than that shared stuff because of that convergent evolution. If you're hunting the same prey and you're doing the same pack style tactics, you're going to end up with similar features because you have a similar lifestyle. In the same way, California and Florida, even though they're a whole country apart, can often have similar housing materials and stuff based on their shared climate properties. That's where we're at so far, right? Have these dire wolves that are their own genus. That means that they're their own group. But importantly for this, they do have the same number of chromosomes as a dire wolf, as a regular wolf, sorry. And the reason why that's important is a big reason why closely related species can't form hybrids with each other is when they have different numbers of chromosomes. Because your cells have to divide every time they replicate, especially if you're making sex cells to keep the species and the hybrid going past one generation, right? Lot of times the reason why hybrids are sterile is the chromosomes don't line up properly and then you're not making properly formed gametes a lot of the time.

0:15:00 Thomas: I didn't know know that at all I'm I'm

0:15:02 Eric: So that's why I think it depends on whether it's a liger or a tigon whether that animal can have kids, right, it has It has to do with the way that the sex chromosomes house genes in big cats and whether you have the right combination of chromosomes in your gametes when you're a tigon or a liger. Tigon or liger being a hybrid of a lion and a tiger.

0:15:19 Thomas: Gonna say these these are also not made up is that what you're telling me right now

0:15:23 Eric: These are not made up. And a liger or a tiger, and I forget which one, are gigantic. Gigantic they are prone to gigantism and they can be like there's a picture of this one very famous one Hercules that's standing up next to one of those tour like double-decker buses and it's able to put its paws on like the roof of that bus

0:15:40 Thomas: I No, this sounds fake. I'm sorry.

0:15:45 Eric: Glad I can always give you at least one that sounds fake per uh podcast but so really what they're trying to do with the extinction is a lot of both what we talked about and how to build a dinosaur kind talk about out the evolutionary lineage of the direwolf and use the DNA as a tool to kind of reconstruct the molecular clock and not really de -extinct them, as we'll talk about in a second, but to show the series of mutations and changes that can bring them about, right? Kind of recapitulate the evolution.

0:16:14 Thomas: What you've been talking about this whole time with your dinosaur factory theme

0:16:17 Eric: And really, there's a note of caution here, because I think there's really good uses for this that don't involve de - extinction. And they actually flagged this on their website, right? Main reason that they're doing this, their right, actual reason for existing as a company isn't necessarily to bring back the mammoth and the direwolf. Those are kind of the splash pages, right? The thing that draws you in. But the real reason for consisting the moral compass of the the company that I actually agree with is we are in the middle of a giant mass extinction, right? Are losing about 30,000 species a year, 150 a day, six an hour,

0:16:23 Thomas: Parks oh yeah what the hell Well, are most of those bugs and shit I don't care about? Because that's so many.

0:16:58 Eric: Now got that right. Most of them are nameless beetles and frogs in the Amazon rainforest that we will never ever hear about.

0:17:05 Thomas: Eventually we're going to get to the good stuff, you know, and that's not good.

0:17:09 Eric: Because ecosystems eventually collapse as we've talked about in our extinction episode, right? But they think that around 50 % of species will be extinct by 2050, that's probably somewhat accurate, if not underplaying it based on the rapid acceleration of climate change. Right that most of these are beetles and frogs, but bugs have a huge impact on the environment and the ecosystems, right. Eventually, we're going to lose something like 30 % of marine mammals, which is something I think people should should care about. Because to be honest, most of those marine mammals, we need to at least keep them alive long enough to interrogate whether they have human -like intelligence. They really are special, special animals, and it's a shame what we're doing to them. But the real reason for existing as a company is, let's say we start to lose the river dolphins that exist in the Amazon, right? Would be able to sequence the river dolphin DNA and recapitulate those in in a surrogate animal, another species of dolphin, to kind of keep them alive. But furthermore, their big bread and butter as a company is they've started integrating AI programs into their genomic analysis that helps look for functional outputs of proteins. If you remember from our genetics episode, the why DNA matters is because three bases at a time, it codes for an amino acid that builds into protein, and those proteins are what actually gives us the functions and traits that make us who we are, right? Their real goal is to sequence the genome of these endangered animals and then use that CRISPR technology to make changes to make them more resistant to climate change, to acidification of the oceans, to more scarce food supplies and warming weather, right, all of those things. So it really is a very noble goal and I actually think that's what they should be leading with as opposed to the de -extinction because I I think that's actually the much more interesting, pressing, and relevant thing that doesn't also come with some caveats, right? Because if you de -extinct animals that have been gone for 10 ,000 years, you are now introducing things back into the wild

0:19:07 Eric: That don't necessarily have their ecological niche still.

0:19:11 Thomas: The Starks aren't even around anymore, I think, most of them.

0:19:14 Eric: Yeah, winter came for them. And unfortunately, summer is coming for us all instead of winter. You do these de - Extinctions, if you're also equipping them for climate change, don't have a great track record of keeping our animals that we take care of and travel with in human society, right? They very, very, very frequently spill out into the wild and start becoming invasive species. And I would say one worry is all it takes is about 15 wolves that were reintroduced to Yellowstone, and they had such a a profound impact on the ecosystem that they actually changed the shape of the river, so even just a pack of these dire wolves escaping and mixing with other animals could have pretty profound effects on the Ecosystem. That's one thing I want to caution about.

0:20:13 Thomas: Interesting. I was not sure what to expect because I see these headlines and I'm so used to like, actually, this is stupid. They just put a wig on a cat and then they called it a diet. So it's refreshing that we got something really going on here, but I'm also seeing the promotional stuff and like video that I watched is so flashy and it's not what I associate with sound science usually uh there's a lot of marketing there's a lot resources being put into that and uh so I'm skeptical of that um what's the verdict here

0:20:43 Eric: That's the thing is there's kind of two arms to this. There's the flashy part that I think is somewhat oversold and is putting a wig on a mouse, literally, as we'll talk about in a second. But then there's the part where they're doing really groundbreaking science that's related to what George Church is famous for, which is that genomic sequencing and pioneering in genetics, right? That stuff, because he is legitimately one of the, you seminal people in this field is, you are not going to hear me criticize the genetics of this at all. The only thing I would take issue with is what they're doing is overselling what it is. And if they were just talking to the academic community, this would be interesting enough on its own. The only reason why it's being oversold is because it's a private company that's selling to the public. Let's talk about what they are overselling and then we can talk about the really cool stuff that they are doing. Sound why they are overselling it. I've talked obliquely about this, right? Their first project was the de -extinction of the woolly mammoth. And to do this, they hired a lead scientist, Beth Shapiro. Beth Shapuro is another perfectly legitimate scientist that is way, way smarter and accomplished than I am, right? We're talking a rogue scholar. She got a AAAS fellowship.

0:21:52 Eric: That a very good point. So she's only slightly less accomplished than I am, right. In general, we are talking about people that are very, very bright and talented, right, she was also a MacArthur fellowship. So this is someone that knows how to think clearly, carefully, and communicate those ideas. And they want to do is they want to take DNA from fossils and do what's called a sequencing alignment and analysis where they take computer programs in AI and compare it to their closest common relatives. You can think of this is like, let's say I needed to get an old copy of a Bible from ancient Rome reconstructed, would need to do a couple of things here. The first thing I would needs to do is I would need a reference of Latin from ancient room, right. So I know that I'm talking about the same language, right, they take their direwolf DNA, and they used it against prior published direwolf. DNA. Now prior DNA of the direwolf was only very, very partially sequenced, right like like 0 .1 % of the genome was covered. But that gives you a foot in the door in the same way like if you do a Sudoku, just getting two of those numbers on the grid does

0:21:51 Thomas: Podcast is she on though you wow yeah

0:23:00 Eric: Tremendous things for solving the rest of that Sudoku. It gives an anchoring point. They anchor to that old repository and do what's called an iterative analysis. And what that basically is is using those computer programs, those algorithms and AI AI to continually align what they have to that reference genome, right?

0:23:19 Thomas: Checking the sudoku columns and stuff over and over yeah

0:23:24 Eric: Right, and every time I look it over, I spot one other thing that I could change that would clean it up, right. And we're refining that resolution over and over again. Because they only had 0 .1%, they needed another scaffold. And here's the part where not going to say I'm dubious, but it's a little bit over my head genomically because I didn't really do computational biology. So let's say that I more have questions that I think could be answered, but I'm interested to hear what those answers would be. They did was they compared it to a Greenland. The reason they chose Greenland is because there's something that separates Greenland from the rest of America, the right? Ocean. And so those Greenland wolves aren't going to be breeding, mixing their genes with any other canid or any other wolves since they moved on to Greenland, right? That gives them a better approximation of a more ancient form of the wolf because they're not evolving with the common gray wolf. Now, the one problem with that is Greenland has its own host of of environmental challenges. They've also been breeding for generations. So they've done their own evolution.

0:24:29 Thomas: Yeah. Nothing freezes in evolution, really. The

0:24:33 Eric: You can say is it's not touched. It's not biased by the current gray wolf, right? It's at least a different anchoring point. What You Might Be Able To Do, and again, this might be a little above my head, but What You might be able to do is also take an American gray wolf and then do like a triangulation instead of a binary regression to improve your odds. But they basically do that iteratively over and over again with their algorithm to help fill out the genome. They ended up getting a sizable portion of the dire wolf genome, read it over three times coverage. And what that basically means is that they've been able to identify and align the same sequence in overlapping patterns at least three times. What I mean by that is like, let's say I'm trying to figure out a sentence, right? The sentence is, I love science, right? I cut up, I love science into a bunch of fragments. Then I keep getting I L O right. L L I L. Oh, and then I get O V E O V O V. Right. Can align the O in I L O and the O and O V to start to piece it together. But the more times I see L O V I L LO, LO, OVE, or LOV, right, the same patterns overlapping over and over again, it

0:25:46 Eric: Increases my, exactly. And so that's what they mean by coverage depth, right? They have seen it aligned properly enough times that they start to gain confidence that this is actually a real read of the genome, if that makes sense. That brings us to the next step of the project, right? Now have what they feel is a pretty good dire wolf genome, right? So they start aligning it to all the other canids that we have available, because we have full genomes of jackals, d -holes, wolves, foxes, right.

0:25:45 Thomas: More confident you are. What's a D -hole?

0:26:17 Eric: A d -hole is like a Eurasian canid. I don't know a lot about them, but D -H -O -L -E.

0:26:21 Thomas: Do you spell that?

0:26:26 Eric: It's an unfortunate name. I will give you that.

0:26:29 Thomas: I was hearing you correctly. Okay.

0:26:31 Eric: Not D -bags, D -holes. These ancient canids clearly diverged from direwolves about 6 million years ago, which is in agreement with the prior stuff. Now, interestingly, they found a greater degree of relationship to the gray wolves than the prior groups that had much less coverage of their studies did, coverage of Their Genetics did. It, right? Interesting for two reasons. One, it might shift our timeline of dire wolf evolution or introduce the possibility of what we call admixture, right, because before we thought that dire wolves were reproductively isolated. With that, I mean, even if there were other dog -like species in the area, if you're reproductibly isolated for whatever reason, either behaviorally, locationally, or physiologically, you are unable to form hybrids with the other species that are are related to you, right? Grizzly bears and polar bears were geographically isolated in reproduction for a bunch of years. Now with climate change, they're coming back and they're no longer reproductively isolated. So they have admixture again. Dire wolves were thought to be pretty distinct, but they're finding in their better coverage a lot more, or at least a couple of instances of admipture that complicates that picture. Question I would have for them is if if you are referencing against a gray wolf genome, are you not already biasing your results to show more similarities, right? Because if you're using them as a reference to fill in gaps, know, they probably have an answer for this, but I could see a scenario where you're predisposing that sequence to be biased towards gray wolves. They do do corrections against this bias, right? They do their best to try and correct against this bias using algorithms and computer technology.

0:28:09 Thomas: Got a stupid question. I Why thought you said that they were grinding your bones to make the bread. They're getting the fossils and they're just, they're, you know, using that. Why Why isn't that, yeah, why isn't that covering? Is it just not enough? Like, is something too much missing or something?

0:28:17 Eric: Why can't they just use the DNA? Is a great question. Yes, it is the problem of time and fossilization, right. Fossilization, and depending on the preservation style, can be disruptive and destructive to DNA, right, lot of these bones are found in tar pits, which is pretty destructive to the DNA. Then if we're talking about anywhere from 100 ,000

0:28:41 Thomas: They resurrected. Oh, they're covered in tar! Damn it! We didn't know once we resurrected them. They These are completely covered in Tar.

0:28:49 Eric: Tar is actually plant DNA, so that gets back to a whole other thing. But the other reason why it's hard to get a complete genome is is if you are alive from around 100 ,000 years to 10 ,000 years ago, and even longer, right? That's like when they were kind of prominent and where the fossils are being isolated from. Has a half -life that usually gives it a limit of viability at around 10 ,00 years.

0:29:12 Eric: So we're already stretching that limit of where you can extract really good quality DNA from. It's kind of similar to like, even if we found the library Alexandria and it didn't burn down, many of those books would be unreadable or those scrolls would be un readable anyway, because they weren't kept in pristine conditions.

0:29:31 Thomas: But if you, I don't know, find enough copies of them and you get little fragments and you're trying to kind of piece that all together with some other things that you think might be sort of copies of the books that you're looking for and, you know, that kind of thing?

0:29:42 Eric: Exactly. But if you have a Rosetta stone and you have other Greek scrolls and you Have other primary sources that reference the books in the library of Alexander, you can start to put together a better picture that you can have some amount of confidence in. You're never going to get to fully reconstruct the gray wolf because dna is really really or the um dire wolf because DNA is really really complicated right it's as we've talked about in some of our other episodes just an order of magnitude harder than recreating a book from the library of alexandria right we're talking billions upon billions upon millions of base pairs with different functions and regulations depending on their location the 3d structure of the chromosomes at any given moment there's lots of complicated things what they did was they tried to align that to other canid species, right? What DNA they had. And then they used computer programs, NAI, to try and predict the amino acid sequences in the same coding regions that they have in dire wolves that match genes in other dogs that are around today, right. If you have genes in alignment for them, then you can be pretty clear that this is a functional coding sequence If it's matching a functional coding sequence in another. What you do there is you start looking for the differences, right? You compare both mutations that affect the phenotype, right, things that would change the amino acid of the protein, what we call non -synonymous mutations, right. Right. And you can predict the effects of them using computer modeling of how that would change the way the protein folds, which amino acids are now exposed to chemistry of other things. Right. There are ways to predict the function of a protein based on its amino acid sequence. And they use that heavily. But you can also find changes that aren't going to change the function, what we call synonymous mutations. Mutations. And those are mutations where, let's say the three -letter sequence TAA and TAT code for the same amino acid. Doesn't matter that that A turned into a T because you're going to get the same function in that protein because you are building the same amino acid. Take a Lego from a different Lego kit but it's the exact same piece, no one is going to check the serial number underneath to make sure that it's from

0:31:54 Eric: The same box.

0:31:53 Thomas: I even know if they have those.

0:31:56 Eric: Really cool is there is is a rate on which mutations happen just in the background, right? That's just because DNA replication is not an exact thing, right, it's something done by life, and life is prone to error prone to mutations. And you can take the ratio of the changes to functional protein sequences, and things that don't seem to change the function at all, and see whether certain genes are under selective pressure or not. If the changes are more random, that you that they're not really under a lot of selective pressure. They're not really that important for survival because the changes aren't contributing to function. If you have an enrichment in changes that do change the function of the protein, now you're likely looking at something that is changing the features and traits of that animal and is open to being under selective pressure, so you can both get at the rates of evolution and you can get how far away something diverged from other species, which is how you build those phylogenetic trees, right? Those evolutionary trees that show the relationship of things. Why I have a lot more confidence in the genetic evidence than I do in the fossil evidence in terms of reconstructing the timeline of dire wolves, because it's a much much more exact science to get those mutation rates and then do that, what we call DN to DS ratio, right? That's ratio of non -synonymous to synonymous mutations. There is a caveat there because I mentioned that it was thought that these dire wolves were thought to be reproductively isolated, right. They were also going through periods of intense glaciation and deglaciation periods throughout the Pleistocene and late Pleistocyne. What that can introduce is what we call genetic bottlenecks. And you can think of bottlenecks as like, let's say I have a jar of M &Ms, right? They have the same ratio of M & M's that is any bag, right, say roughly 15 % of all the colors, right. If I reach into that jar and pick just 10 random M &M's, there's no guarantee that those 10 M &N's are going to match the overall proportions

0:33:50 Eric: Of the bigger jar, right。 now if I were to reproduce those 10 m &ms they're going to have a much different color pattern than the bigger jar even though they came from that same population so it can throw off your analysis of things like mutation rates divergence and stuff like that when you have these isolated populations that can get small in number because any you know natural disaster that thins out the herd or the pack in this case, right, shift those numbers in really putative ways. So that's one caveat you want to keep in mind is that some of this analysis does have to account for like those genetic bottlenecks, the idea of artifacts from aligning it to that gray wolf reference genome. All of these things can frustrate it, but the AI and sequencing technology they're doing to increase the coverage from 0 .1 % to over 500 times greater and really sequence the genome in in a more meaningful way, is really meaningful and cool science. And that's kind of what I mean by like, I'm not sure the de -extinction is actually the coolest science. I think it's actually the phylogenetic mapping and sequencing that they're doing. And that's why their paper about the extinct dire wolf is actually along the sequencing lines and about reconstructing the phylogenetic tree than it is about reconstructed and de - extincting the dire wolf itself.

0:35:07 Thomas: You're saying the paper they're doing is more, like, academic and, yeah, but the presentation is, okay.

0:35:10 Eric: I'm talking now. Yes, it's more this academic Sequencing. And that's why I'm saying it's not quite what Elizabeth Holmes was doing with Theranos because Therinos, I don't think even had the academic paper, right? They were just selling the fluff. Has a lot of really good hardware and wetware behind the scenes. And they collaborate with a ton of academic labs as well. This is really legitimate and cool science. It's just being used in a slightly hokey way. Here's why it's slightly hokey. And I think it's important to bring up their de -extinction of the woolly mammoth project because the first thing they did, and this is where we're going to get into how they're actually going to go about de - extincting the dire wolf using these genetics, right? So they take these genetics and they found 20 mutations that they think are functionally putative in distincting a dire wolf from a gray wolf, right. Are 20 14 genes that really changed the behavior the translation, the making of the protein, and change the phenotype of the wolf themselves, right? So what they're going to do, and this is similar to what they did with the woolly mammoth, is they're gonna use that CRISPR -Cas9 technology to treat the genetic code of a gray wolf like a Word document. They're gonna go in and they're Gonna make those 20 changes in those 14 genes. And then they're going to grow that embryo inside of that gray wolf, right? Then when that wolf is born, it will have way more features of a dire wolf than a gray wolf. And that's what they are calling a de -extinction of dire wolves. Don't think I

0:36:37 Eric: Have to use fancy science to explain why that's not quite that.

0:36:40 Thomas: I mean, but is it, the like same thing you're trying to do with the growing the dinosaur though? I mean is it like the best you can do anyway really?

0:36:47 Eric: What I wouldn't call it is how to build of Velociraptor, evolution in some ways is kind of like Sean Carroll has this really good analogy for entropy, right, how entropy always increases, where once an egg cracks, you can get all the fragments as best you can, you can find the craziest crazy glue you want, but there really is no true putting the egg back together, right. Once a species goes fully fully extinct, you can find the putative genes. You can really even in some cases fully map the genome. But if you're using a surrogate animal, you're never going to fully put the egg back together. And the reason why is because the mom has a huge effect on the development of that embryo in providing the environment and providing those maternal mRNAs to really get the process started And really helping set up the developing embryo for success in its form and function. And so even though you will have a gray wolf that can serve as that, if you're changing that environment, you're never going to actually get a true dire wolf back, even if you had a true tire wolf genome. Them. What they do is they take out cells from a regular gray wolf, right? Use cells form the hematopoietic system, right, bone marrow stem cells. And these are really useful because they're very easy to reprogram back into embryonic stem cells, and more importantly, they don't hurt the wolf, which means you can cultivate them over and over again and really iterate and refine that process without worrying about harming the animal, then they will take those cells and grow them out on a petri dish. Will then use those CRISPR -based technologies, right, that DNA editing, to introduce those 20 different mutations that they need to do to change the function of the direwolf. Then going to put that back in to a direwolf using this technique called um somatic transplantation right what that basically means is they are taking out the gamete right of the gray wolf they're taking out the ova the genes in the sex cell right they're re -putting in the fully diploid because obviously an egg only has one half of the pairs of chromosomes right not somatic, it's what we call haploid. If the mom had both pairs of chromosomes, you wouldn't need the dad to make a baby,

0:39:02 Thomas: Gotcha.

0:39:02 Eric: Right? They're doing is they're taking body cells from a wolf and re -editing it so that you don't need a mom and a dad. It's already a fully diploid cell, and then placing the DNA into the egg and letting that develop like it was fertilized by a mom or a dad, point is you can already see all the problems with calling this a dire wolf right really a gray wolf with a bigger body plan. They're bulking up a bit. The example they use in their website is they found a transcription factor, one of those proteins that bind onto DNA and tell certain genes, go, go. We need to be turned on. Change one of these proteins to make the wolf bigger.

0:39:19 Thomas: Woke scientists are destroying the family unit.

0:39:45 Eric: They changed the coat color because apparently they were also Game of Thrones fans and they think direwolves should have white coats. I want to note, you can't really tell coat color from fossils. So unless you've seen a direwolf or you have fossilized pigments, you're not going to know their coat color because bones don't have fur, right?

0:39:44 Thomas: Hmm, yeah. More you know.

0:40:05 Eric: That one is really more showing the cosmetic parts of it, but they're basically growing a wolf, kind of like what you said at the beginning, to have a dire wolf wig, right. Thing I want to point out, because this is from their website, they're also lengthening the baculum, which is the penis bone that humans don't have, there's a couple of things that they're doing to make it more direwolf-y. And this is similar to their mammoth project. They're doing the mammoth project in two stages. One is they're trying to find ways to culture stem cells or induce stem cells from cultured elephant cells so that they can do that same somatic nuclear transfer that they're doing with the direwolf. To test out their theories, they're making genome changes in mouse embryos and using what's called electroporation. So basically takes advantage of the fact that DNA has a charge. They're using electricity to make an extra piece of chromosome with the DNA edits they want for the mouse.

0:40:14 Thomas: A little better than that. What oh classic frankenstein method

0:41:11 Eric: Exactly. And it's not to jolt start the mouse, actually, but to bring the DNA and shepherd it into the tissue, right to kind of guide it along. It's the same way. Have you ever seen like the gel electrophoresis that is used to analyze DNA? Well, the way we find out who like our parents, the way we do paternity tests is we take DNA and we use these enzymes that recognize certain sequences. And we cut DNA using those enzymes. And then depending on your sequences, your DNA will be cut in different places. You then match the pattern that you get of those DNA sequences.

0:41:23 Thomas: Yeah don't they grade it like a scantron almost like they kind of it's like the slots and stuff okay

0:41:48 Eric: More like reading a ladder. But yes, right where you just match up the bands, but those bands separate out by running a current with bigger things moving more slowly and smaller things moving more quickly, right. A similar process to electroporation, but it's just a way to introduce those DNA edits at the single cell stage, right. They grow these mice in utero with those DNA edits. And what they found was those mice were bigger on average. They had very, very thick, long hair, like a woolly mammoth. And they're They're using this as proof of concept that they can use their AI technology to find putative changes, putative mutations that control biology, and then introduce those changes to recapitulate the biology, right? All of this is really, really cool stuff. It's just not de-extinction.

0:42:33 Thomas: What would de-extinction mean need to be is just like you need to have the actual complete like genome mapped hmm

0:42:39 Eric: Yeah. You would really need to have something close to the full genome because there's so much stuff we don't know about DNA, as we've talked about before, that I'm a little skeptical that 14 or 14 genes and 20 base pairs is enough to explain 5.7 million years of evolutionary difference, especially when they themselves found really thousands of bases that were different. They were just choosing the 20 that they mapped in their AI program as being the most important, for me, there's really a lot more than just those 20 base pair differences between a dire wolf and a wolf. There was only 20 differences. Would likely be the same species because every human being when they're born picks up around 30 mutations that are completely different from the ones they get from meiosis and their parents. Like 30 base pairs is not, we're not really talking entire genus level differences of species. So that's really really where they're at now is what they're really trying to do now is go into a bunch of different ways to improve that method of cultivating their de-extincted animals, right? One thing they're working on that's really cool, besides like improving the AI so that they can get better reconstruction, finding more samples, because they did all of this just based off two bone samples of direwolves. So like really, really powerful stuff for the limited amount of data they have.

0:44:01 Thomas: Find some

0:44:01 Eric: Exactly. But not only that, they want to find ways to not just use a regular wolf as their parent, but make exogenous wombs, right? Make wombs that you can cultivate because then you can start to find genes in the dire wolf that might match that moms would provide for their kids and get closer and closer to recapturing that ancient environment, for lack of a better word.

0:44:01 Thomas: More bones yeah

0:44:27 Eric: That's really what they're trying to do with de-extinction. I don't think that's the most interesting thing. So if you don't mind, I'd actually like to talk about some of the really cool stuff that they have planned. Because what they are doing is kind of the opposite of the CRISPR mosquito project, right. In that case, people were using CRISPR to get rid of all the mosquitoes so that they don't spread malaria. In this case, let's say we're losing some frogs that we finally found in the Amazon rainforest within the next 10 years, right? They're trying to do is they're tying to sequence the genome of the frogs, find genes that can be functional in helping adapt to the environment, and then use species that are well adapted to map those genes onto those frogs and make them more resistant to climate change. Those ones are really interesting and powerful and don't have the same ethical risks of introducing possibly invasive species into an ecosystem that isn't ready for

0:44:27 Thomas: Oh sort of like trying to jump start evolution in a way right like kind of help it along like all right well can't keep up with these changes that are happening apparently let's do get uh evolution faster on you

0:45:30 Eric: Them. Exactly right. It's kind of like when you have a new tree and you have those scaffolds that help nurse it along before it's able to survive by itself. Right. I would say like that's a much more feasible, possible and also just interesting use of this as we navigate the sixth mass extinction because again those ones you can put the egg back together we have enough there and we still have um animals right these are endangered species not extinct species so you don't need surrogate moms you just need any um animals already alive and you can rapidly both increase their numbers but also give them the resistance that they can start increasing the numbers themselves and that's really where I think the future of this company will be less so in the designer species I

0:46:12 Thomas: Where's the profit in that

0:46:15 Eric: That's to be honest I think that's why they're selling the de-extinction so hard right is because one is a profit motive and one is really really good and important science and

0:46:25 Thomas: If this is great science, I'm sure the federal government would want to. Ah, never mind.

0:46:29 Eric: But right this is a little bit like when steven spielberg makes you know jurassic park so that he can then make schindler's list and not worry about the budget right you do one for the studio and one for you this is kind of the science equivalent of that, it's still really good science. The only thing I'm taking issue with is the way they're selling it, right?

0:46:49 Eric: If they said, we are recapitulating the dire wolf phenotypes in a gray wolf by making functional changes using AI algorithms to align the genomes and predict functionality, but the people that knew what they meant would be legitimately interested in what they're doing, right, that's really, really good and important science. Just in science, were sticklers about definitions and terms and so I think it's important that they got out ahead of that and they said straight up we think the definition of de -extinction is wrong this is our definition sadly I think their definition is wrong but their ideas are right

0:46:49 Thomas: No would know what they meant but you. Yeah yeah okay sure yeah and like I mean the marketing is really funny that one of the videos the direwolf is back from extinction colon restoring a pop culture icon. Like the dire wolf was John Snow back in the day and they're like we're gonna we're bringing it back yeah it's funny yeah

0:47:36 Eric: Pretty sure it's had one moment in the sun and it was being the friend of John Snow

0:47:41 Thomas: Pop culture icon that's a weird way to put it uh but hey you know they gotta sell whatever sells more science units I

0:47:48 Eric: These these are important things right we are losing a lot of those quaternary predators, right? Those top of the food chain predators. And so having things like wolves in Yellowstone was actually super important. I mentioned that it changed the shape of the river, right. Reason why is these animals, those top level predators tend to have really big cascading effects. And we actually just went over this for my kids as we prepare for the Keystone. But the reason why his wolves eat deer, right, they eat things that like to graze on grass ton, right? The wolves weren't there, those populations go out of control, right, they keep growing and growing and grow. So what happens to the grass in that scenario, and more eaten, so it gets less and less, right.

0:48:24 Thomas: Gets more and more eaten yeah

0:48:28 Eric: Grass vegetation is very, very important for things like soil, runoff, erosion, right. So just reintroducing the wolves, reintroduce the grass and help fix the shape of the river and give healthy riverbanks again. So these are really, really important things and we are losing very important animals. Just wish that they would focus on keeping our wolves nice and healthy instead of turning them into dire wolves.

0:48:50 Thomas: Well cool though I mean it's still pretty cool I like that that's real um there's definitely pluses and minuses there but I'm so glad that it's not just a complete hoax bull crap and interesting stuff what's on the horizon you know are they are they looking to is anyone gonna actually do Jurassic Park are we gonna do one of those for real anytime soon

0:49:06 Eric: Jurassic Park would be even harder because the problems we're talking about at 100 to 10,000 years of DNA degradation just gets compounded when we start getting to 60 million.

0:49:16 Thomas: The mosquito doesn't solve that

0:49:18 Eric: Mosquito doesn't I think there actually was found I think someone recently found a like perfectly preserved mosquito in amber but like even that right you'd have to isolate the dna from the blood of that mosquito hope it recently sucked on dinosaur dna have enough of that dna even though a mosquito's proboscis will shear that DNA right like break it up into fragments mix it up with anything else it fed on and then you'd have to isolate that so

0:49:42 Thomas: Man you'd accidentally make an amalgam of all the dinosaurs yeah

0:49:46 Eric: It would be more like trying to reconstruct all the books of the Library of Alexandria at once with all the fragments of the library of Alexandria once instead of one book in one set of fragments.

0:49:57 Thomas: That was a lot of fun thank you so much for taking us through that. But direwolves, are they selling these things? What are they doing with this?

0:50:03 Eric: Think they only have a couple right now, but my guess is they want to take at least one batch through their full life cycle and then they would probably start thinking about commercializing. But also, it might just be a proof of concept, right? I can iterate and refine DNA sequencing of this animal that's been extinct for 10,000 years, think about one that we've seen go extinct in our lifetime. That should be light work, right. This might just be a proof-of-principle to then sell themselves to governments, organizations, right, as we do have the ability to do this if you give us a chance.

0:50:39 Thomas: Interesting. Eric Jaffe, is else we should know about these direwolvs?

0:50:43 Eric: I mean um guess one thing I would be remiss is colossal biosciences wants you to know wolves are given a very bad reputation and a lot of the things that we blame them for as we hunted them to near extinction really overblown wolves are pretty awesome and think it's interesting that wolves were able to come along for the ride with humans and become domesticated whereas dire wolves were not able to make that same ride and the reason why I think that's interesting is the time they went extinct is known for two major things the um of that ice age right and the peeling back of the ice caps which introduced a lot of climate change but also humans coming across on the last vestiges of that land bridge between russia and alaska and so there are these people called the clovis people where you find these mass killing sites of woolly mammoths of other large, what we call megafauna, right? Associated with what we call these Clovis spear points, rock that are shaped in a certain way. That's a hallmark of that culture of people. And so what I find interesting is is this another case of an animal that met its end because of climate change or because they were at the wrong end of humans messing up the ecosystem as an invasive species, right, because if they kill all of the prey of the wolf, then they're no longer available. Now, if these humans came across with gray wolves themselves, it would make sense why dire wolves did not get domesticated, right? That position's already filled. If they didn't come across with grey wolves, what I'd be curious about is why couldn't the dire wolves pick up that same trick when they were also pack hunters? They had similar, right, behaviors to other candidates. So it seems like they would be a prime candidate for domestication as a way to avoid extinction but they clearly didn't pick up that trick that

0:52:26 Thomas: Prime candidate. I get it. I see what you're doing there.

0:52:29 Eric: One I didn't do barking up the wrong tree was on purpose that one was unintentional still

0:52:33 Thomas: Well, that's super interesting. Thank you so much for that breakdown, Dr. Eric Jaffe. Anywhere you want to point listeners to? Anything at all?

0:52:40 Eric: Off the instas um I think I told you a story offline about why I'm never going on social media ever again so I have nothing to plug cheers

0:52:49 Thomas: There you go. Don't find him, everybody. If you want more Dr. Eric Jaffe, don't. No this most of those bugs and shit I don't care about? Because find some more bones yeah