Vincent Racaniello: Viruses and Vaccines | Lex Fridman Podcast #216

Lex Fridman 0:00 The following is a conversation with Vincent Racaniello, professor of microbiology and immunology at Columbia. Vincent is one of the best educators in biology and in general that I've ever had the pleasure of speaking with. I highly recommend you check out his This Week In Virology podcast and watch his introductory lectures on YouTube. In particular, the playlist I recommend is called Virology Lectures 2021. To support this podcast, please check out the sponsors in the description. As a side note, please allow me to say a few words about the COVID vaccines. Some people are scared of a virus hurting or killing somebody they love. Some are scared of their government betraying them, their leaders blinded by power and greed. I have both of these fears, and too, I'm afraid, as FDR said, of fear itself. Fear manifests as anger, and anger leads to division in the hands of charismatic leaders who then manufacture "truth," in quotes, that maximize controversy and a sense of imminent crisis that only they can save us from. And though I'm sometimes mocked for this, I still believe that love, compassion, empathy is the way out from this vicious downward spiral of division. I personally took the vaccine based on my understanding of the data, deciding that, for me, the risk of negative effects from COVID, short-term and long-term, are far worse than the negative effects from the mRNA vaccine. I read, I thought, I decided for me. But I never have and never will talk down to people who don't take the vaccine. I'm humble enough to know just how little I know, how wrong I have been and will be on many of my beliefs and ideas. I think dogmatic certainty and division is more destructive in the long term than any virus. The solution for me, personally, like I said, is to choose empathy and compassion towards all fellow human beings, no matter who they voted for. I hope you do the same. Read, think, and try to imagine that what you currently think is the truth may be totally wrong. This mindset is one that opens you to discovery, innovation, and wisdom. I hope my conversation with Vincent Racaniello is a useful resource for just this kind of exploration. He doesn't talk down to people, and he's the most knowledgeable virologist I've ever spoken to. He has no political agenda, no desire to mock those who disagree with him. He just loves biology and explaining the fundamental mechanisms of how biological systems work. That's a great person to listen to and learn from with an open mind. I hope you join me in doing so, and no matter what, try to put more love out there in the world. This is the Lex Fridman Podcast, and here is my conversation with Vincent Racaniello.

Lex Fridman 3:11 You mentioned in one of your lectures in virology that there are more viruses in a liter of coastal seawater than people on Earth. In the Nature article titled Microbiology By Numbers, it says there are 10 to the 31 viruses on Earth. Also it says-

Vincent Racaniello 3:30 (laughs)

Lex Fridman 3:31 ... that the rate of viral infection in the ocean stands at 10 to the 23 infections per second, and these infections remove 20 to 40% of all bacterial cells-

Vincent Racaniello 3:44 (laughs)

Lex Fridman 3:44 ... each day. There's a war going on. Do you, uh, what do you make of these numbers? Or why are there so many viruses?

Vincent Racaniello 3:52 So the, the numbers you're quoting, they're in my first virology lecture, right?

Lex Fridman 3:57 Yeah.

Vincent Racaniello 3:57 'Cause people don't know these numbers and they get wow, they get wowed by them, so I love to give them. So the-

Lex Fridman 4:04 By the way, sorry to interrupt, but-

Vincent Racaniello 4:05 Yeah.

Lex Fridman 4:06 ... as I was saying, uh, offline, you have one of the best introductory lectures on virology that, uh, I've ever seen. Introductory lectures, period. So I highly recommend people find you on YouTube an- and watch it if you're curious at all about viruses. It, it, um, it... Yeah, there's a lot of times throughout watching it I felt like, "Whoa."

Vincent Racaniello 4:26 Yeah, that's my goal is to whoa-

Lex Fridman 4:28 (laughs)

Vincent Racaniello 4:28 And, and it's what my students tell me. One student once said, uh, "Every day after every lecture, I could go home and tell my roommate something she didn't know-"

Lex Fridman 4:37 Yeah.

Vincent Racaniello 4:37 ... and, and it blew her away. So the number of viruses is really an amazing number. So that number, 10 to the 31, is actually just the bacterial viruses in the ocean. So there are viruses that infect everything on the planet, including bacteria, and there are a lot of bacteria in the ocean. And so 10 to the 31 is from basically particle counts of seawater all over the world. So there are more viruses than 10 to the 31, but just in the ocean, and that number is so big. First of all, the mass exceeds that of elephants on the planet by a thousandfold.

Lex Fridman 5:12 Mm-hmm.

Vincent Racaniello 5:13 And if you lined up those viruses end to end, they would go 200 million light years-

Lex Fridman 5:18 Mm-hmm.

Vincent Racaniello 5:19 ... into space. It's so big a number. (laughs) It's, it's amazing. And then yes, 10 to the 20 some infections per second of these viruses killing bacteria and releasing all this organic matter, and that's part of this what we call the biogeochemical pump, cycling of material in the ocean. Uh, the bacteria die, they start to sink, and then they get metabolized and converted to c- to compounds that, that are needed. A lot of it gets released as carbon dioxide and so forth. So these are actually really important cycles that are catalyzed by the virus.

Lex Fridman 5:54 Well, it's so wild that nature has developed a mechanism for mass murder of bacteria.

Vincent Racaniello 5:58 Well, that's one, that's one way to look at it, but it's just what happened, right? (laughs)

Lex Fridman 6:02 It's interesting. I mean, um, I wonder what the evolutionary advantage of, like, such fast cycling of life.... is. Is it just an accident of evolution that viruses are so numerous, or is it, um, is it a, is it a feature and not a bug?

Vincent Racaniello 6:20 So the, the fast is it rain- it does not all fast. Not all viruses are fast. Some are 20 minutes per cycle, some take weeks per cycle. Um, but that's just per second. There's so many viruses in the ocean that that's what you get per second, no matter how fast-

Lex Fridman 6:39 Right.

Vincent Racaniello 6:39 ... the cycle is.

Lex Fridman 6:40 (laughs)

Vincent Racaniello 6:40 But the, I look at it this way, viruses were probably the first organic entities to evolve on the planet, long ago, billions, billion years ago, just as the earth cooled and organic molecules began to form. I think these self r- we call them self-replicators. They're just short things that today would look like RNA, which is the basis of many viruses, right? They evolved and they were able to replicate. Of course, they were just naked molecules. They had no protection, and it was just RNA based. And that, that's tough because RNA is pretty fragile in the, in the world. And, um, it probably didn't get very big as a consequence. But then proteins evolved, and I'm skipping, like, hundreds of millions of years (laughs) of evolution. Proteins evolved maybe without a cell, maybe with a cell. But then to make a cell, there probably were some RNA-based cells early on, but they were pretty simple. But the cells that we know of today, even bacteria and, and single-celled eukaryotes-

Lex Fridman 7:48 Mm-hmm.

Vincent Racaniello 7:49 ... they have very long DNA genomes. And you need a lot of DNA to make a complicated cell. And so we think at some point the RNA became DNA. And probably one of the earliest enzymes that arose is the enzyme that could copy that RNA into DNA, which we now know today as reverse transcriptase, which my former boss, David Baltimore and Howard Temin co-discovered. And that, that enzyme arose and copied RNA to DNA. And then you could build big cells with D 'cause DNA can be millions and millions of bases in length. And RNA, the, the longest RNA we know of is 40,000, uh, bases, not much bigger than the SARS-CoV-2.

Lex Fridman 8:33 What, what would you say is the magic moment along that line? I saw it was, um, one or two bil- billion, maybe three billion years it took to go from bacteria to, to s- like, complex organism. Like, it seems like Earth had a very long time, like, not a very long time without life.

Vincent Racaniello 8:55 Mm-hmm.

Lex Fridman 8:56 And then a very long time with very primitive life. Um, maybe I'm discriminating calling bacteria (laughs) primitive life, but-

Vincent Racaniello 9:03 Yeah, people would object to you- (laughs)

Lex Fridman 9:04 Yeah. (laughs)

Vincent Racaniello 9:04 ... doing that for sure. (laughs)

Lex Fridman 9:06 But it seems like complex organism, when you s- starts becoming something like, um, I don't know what's a good, uh, not animal-like, but more complexity than just, like, a single cell. Um, what, what do you think is the magic there? What's the hardest thing? If you were trying to engineer Earth and build life and build a simulation, so obviously we're living in a video game, what this is.

Vincent Racaniello 9:27 Mm-hmm.

Lex Fridman 9:28 So if you're trying to build this video, what's the hardest part along-

Vincent Racaniello 9:31 So bacteria-

Lex Fridman 9:31 ... this evolution pathway?

Vincent Racaniello 9:32 ... are mostly single cells. They do make colonies. They get together in biofilms, which are really important, but they're all single bacteria in that. And the key is making an organism where cells do different things, right? We have skin cells and eye cells and brain cells, and bacteria never do that. And the reason is probably energy. Bacteria don't, can't make enough energy to do that. And so there was another cell existing at the time, the archaea.

Lex Fridman 10:04 Mm-hmm.

Vincent Racaniello 10:06 And the idea is that a bacteria went into an archaea and became the modern day mitochondria, the energy factory of the cell. And that now let that cell develop into more and more complicated organisms like we have today. It was all about energy.

Lex Fridman 10:21 So the mitochondria, the energy, uh, the mitochondria is the magic thing.

Vincent Racaniello 10:26 I think so. Uh, it says actually not my idea. It's Nick Jones. Have you heard of Nick Jones? He's a evolutionary biologist in the UK and he's, he's done experimental work on this. And it's his idea that the defining point was the ability to make a lot of energy, which a mitochondria can do. It's basically a whole bacteria inside of a, a bigger cell. And that becomes what we now call eukaryotes. And, uh, uh, that they can get more and more complicated. So let me bring you back to the viruses.

Lex Fridman 10:53 Yeah.

Vincent Racaniello 10:53 I want to finish that story.

Lex Fridman 10:54 Yeah. Which points of viruses come along?

Vincent Racaniello 10:56 So remember we have these pre-cellular, they're called pre-cellular replicons, right? And, um, so we have a pre-cellular stage where we have these self-replicating molecules, then cells arise, uh, and then the self-replicating molecules invade the cells. Bec- why? Because it's a hospitable environment. And I mean, they didn't know that. They just went in and it turned out it was beneficial for them. So it stuck. And they replicate inside the cell now where they have pools of everything they need, they get more and more complicated, and then they steal proteins from the cell to build a protective shell.

Lex Fridman 11:33 Right.

Vincent Racaniello 11:34 And then they can be released as virus particles. They're now protected. They can move from host to host. And because they're at the earliest stages of cellular evolution, they can diversify to, to infect anything that arises. And that is why I think (laughs) there's so many of them, and everything on the planet is infected, because the ancestor of everything was infected many years ago.

Lex Fridman 11:57 So it's easier to steal than to build from scratch. So, like, it's easier to sort of break into somebody else's thing and steal-

Vincent Racaniello 12:06 Yes.

Lex Fridman 12:06 ... their proteins.

Vincent Racaniello 12:06 Yes.

Lex Fridman 12:07 (laughs)

Vincent Racaniello 12:07 My colleague, Dixon DePommier, calls viruses safe crackers.

Lex Fridman 12:11 Safe crackers.

Vincent Racaniello 12:12 (laughs)

Lex Fridman 12:12 So it's just, uh, from an evolutionary perspective, it's-Yeah, it's, it's easier to steal because you can select, but then you have to figure out mechanisms for stealing, for breaking into... for, for cracking the safes.

Vincent Racaniello 12:25 Well, you don't have to figure out, it just happens, right? Because molecules are so diverse that a molecule gets into a cell, and if there's a protein that sticks to it, it's gonna stick, and that gives an advantage. There's no, you know, there's no planning. There's no s- thinking about it, right? It just happens.

Lex Fridman 12:45 Uh, we'll return to that.

Vincent Racaniello 12:47 (laughs)

Lex Fridman 12:47 (laughs) What of... but these numbers are crazy. So what... as these, uh, more complex organisms evolved, uh, let's take us humans as an example, uh, should we be afraid of these high numbers? Should we be worried that there's so many viruses in the world?

Vincent Racaniello 13:05 Uh, to a certain extent. I mean, they have... it's twofold. They're good and bad, right? Viruses are no... there is no question they can be bad. We know that because they've infected and caused disease throughout history. But we're also... you and I are full of viruses that don't hurt us at all, and probably help us. And every organism is the same. So they are clearly beneficial as a consequence. So I think... so every living thing on the planet has multiple viruses infecting everything you can see. And most of them, I think, we don't worry about because they can't infect us. They're unable. In fact, now you could actually... you can actually take your feces and send them to a company and they will sequence your viruses in your feces for you.

Lex Fridman 13:50 Mm-hmm.

Vincent Racaniello 13:50 Your fecal virome, (laughs) right? And the most common virus in human feces is a plant virus that infects peppers. It's called pepper mottle mosaic virus.

Lex Fridman 14:02 (laughs)

Vincent Racaniello 14:03 And that's 'cause people eat a lot of peppers.

Lex Fridman 14:04 Mm-hmm.

Vincent Racaniello 14:05 And it just passes right through you. Cabbage is full of viruses from the insects that walk on the cabbage in the fields. We eat them, they just pass us. So I think most of the viruses we don't need to worry about, except when we're talking about species that are closest to us. Mammals, of course. I, I th-... and I think the most numerous ones are the most concerning, their viruses, like bats. Bats are 20% of mammals, and rodents are 40% of mammals.

Lex Fridman 14:33 Wow.

Vincent Racaniello 14:33 And we humans live nearby, right?

Lex Fridman 14:36 Yeah.

Vincent Racaniello 14:37 And we know throughout history, many viruses have come from bats and from rodents to people, no question about it.

Lex Fridman 14:42 So there's a proximity in terms of just living together and a proximity genetically too. So it's more likely that a virus will jump-

Vincent Racaniello 14:48 Yeah, that's right.

Lex Fridman 14:48 ... from a bat and a rodent.

Vincent Racaniello 14:49 And birds too. Birds can give us their viruses. That's happened ma-... you know, influenza viruses, uh, come from birds mainly. So I think those are the three species I... well, not species. It's higher than species, obviously. But those are the three I would worry about in terms of getting their viruses and we don't really know what's out there, right? We have very little clue about what viruses a- and I've, for years, wanted to capture wild mice in my backyard and see what viruses they have 'cause no one knows.

Lex Fridman 15:20 Well, you... (laughs)

Vincent Racaniello 15:21 And it's an easy-

Lex Fridman 15:21 But you can't ask them. So you mean map, uh, like is there which-

Vincent Racaniello 15:24 I can't ask them. Yeah. No, I have to... would have to sacrifice them and take tissue and then bring it in the lab and do genome sequencing.

Lex Fridman 15:30 So you can do a, a thorough sequencing to determine which viruses?

Vincent Racaniello 15:33 Yeah.

Lex Fridman 15:33 Is there a sufficiently good ca- categorization of viruses where you'd be-

Vincent Racaniello 15:38 Uh, that's a very good question. So whenever you do sequence, right, you get some environmental sample and you extract nucleic acid and you sequence it. What you do is you run it past the database, right? The gold standard is the GenBank database, which is maintained here in the US. And you see if you get any hits, and then you can say, "Ah, look, this, this sequence is similar to this virus," and you can classify all the viruses you see. The problem is 90% of your sequence is dark matter. It doesn't hit with anything. It's... probably a lot of it is unknown viruses, and that's gonna be hard to figure out 'cause someone's gonna have to go after it and sort it through. So yes, you can find a lot of viruses and the numbers you get are astounding. You can find thousands of new viruses just by looking in various life forms. But y- there are many more that we don't pick up because they're not in the database.

Lex Fridman 16:33 Maybe this is, uh, a good time to take a quick tangent. What do you think about AlphaFold2? I don't know if you've been paying attention to that.

Vincent Racaniello 16:39 Mm-hmm. Mm-hmm.

Lex Fridman 16:40 But, uh, them, uh, DeepMind solving the protein folding problem and then also releasing... well, first of all, open sourcing the code, which is, for me as a software person, I, I love.

Vincent Racaniello 16:52 Mm-hmm.

Lex Fridman 16:52 And then second of all, also making, like, 300,000 predictions or something like that for different protein structures and releasing that data.

Vincent Racaniello 17:01 Yeah.

Lex Fridman 17:02 Uh, so on the side of... 'cause you make... you're saying there's dark matter.

Vincent Racaniello 17:06 Right.

Lex Fridman 17:07 Like, is there something, um... what... first, what are your general thoughts, level of excitement about their, uh, their work? And second, how can that be applied to viruses? Do you think we'll be able to explore the dark matter of virology using machine learning?

Vincent Racaniello 17:24 Yeah. Absolutely, because in all this dark sequence, you can translate it and make a protein. You can see what a protein looks like. It has what we call an open reading frame, right?

Lex Fridman 17:33 Mm-hmm.

Vincent Racaniello 17:34 A start and a stop. And right now, it's just a bunch of amino acids, but if we could fold it, maybe the fold would recog-... would be like something we already know.

Lex Fridman 17:43 Mm-hmm.

Vincent Racaniello 17:43 Some protein fold, which gives you a lot of clues, right? 'Cause there are only so many protein folds, uh, (laughs) in biology, and that dark matter is probably one of them. So I think that's very exciting because for years, I followed structural biologists for years, and, you know, in the beginning, we couldn't even solve structures of viruses.

Lex Fridman 18:04 Yeah.

Vincent Racaniello 18:04 They're too big. We could do small molecules like myoglobin. I think that was the first one done. Took years to do that. Then as computational power increased, then they could start to do viruses. But it took a long time. X-ray crystallography be-... dependent on getting crystals of the virus, right? And now we can do cryo-electron microscopy, which is much faster. Uh, you could solve the spike of SARS-CoV-2 was solved in two months by, by Jason McLellan here in, uh, Austin actually at the beginning of the, the pandemic. But you're limited. You can't do huge proteins. You can only do moderately sized ones. So, or I sh- actually, you can do viruses but you can't do small proteins. So that's speeded it up, but it's still too fast to solve. Uh, you get a new protein, you wanna solve its structure. So if we could predict it, and I know from talking to structural biologists, this has been their holy grail from day one.

Lex Fridman 18:58 Mm-hmm.

Vincent Racaniello 18:58 They wanna be able to take a sequence of a protein, put it in a computer, and have the, the structure put out without having to do all the experiment. So that's why this is very exciting that-

Lex Fridman 19:07 Yeah.

Vincent Racaniello 19:07 ... you can predict it. That mean it's not finished obviously and there's more to do. But I think there will be a day where you could take any amino acid sequence and predict what it's gonna look like.

Lex Fridman 19:17 See, but-

Vincent Racaniello 19:17 Yeah.

Lex Fridman 19:18 ... like aren't structural biologists gonna get greedy? So once you have that, don't you wanna go more complicated then? Don't you wanna go ... 'Cause that's, that's just the first step, right?

Vincent Racaniello 19:27 Sure. Sure.

Lex Fridman 19:27 To go from amino acid to the structure.

Vincent Racaniello 19:29 Yeah.

Lex Fridman 19:29 Then there's like multiple protein interactions. Like, how do you get to the virus?

Vincent Racaniello 19:34 Well, so that's what the ultimate goal of getting a structure is, that then you can do experiments and figure out what the structure means, right? So many ... In the old days, structural biology was a career in itself. You worked with people who had a system and just solved proteins for them, and then you moved on to another one. You didn't really do any experiments. The other people got to do all the interesting experiments.

Lex Fridman 19:56 Right.

Vincent Racaniello 19:57 Now, e- e- young structural biologists are multifaceted. They solve the structure and then they say, "What happens if we change this amino acid? Oh, look, it, it blocks binding to the receptor. This must be the receptor binding interface." So that's the exciting stuff, absolutely, is doing the experiments.

Lex Fridman 20:14 I wonder if you can do some kinds of like, um, simulations of like, uh, you know, different proteins or multi-protein systems going to war against each other. (laughs) Like de- to try to figure out, um, you know,

Lex Fridman 20:31 reinforcement learning is used in AlphaZero, for example, to learn chess and Go.

Vincent Racaniello 20:36 Mm-hmm.

Lex Fridman 20:37 And that's using the self-play mechanism where the thing plays against itself-

Vincent Racaniello 20:41 Sure.

Lex Fridman 20:41 ... and learns better and better.

Vincent Racaniello 20:42 Yeah. Yeah.

Lex Fridman 20:42 Whether you can ... I wonder if you can simulate almost evolution in that way for a s- for primitive biological systems-

Vincent Racaniello 20:49 Sure.

Lex Fridman 20:49 ... have them in simulation fight each other and then see what comes out. Like a super dangerous virus comes out or a super like, uh, Chuck Norris type of thing that defends against the super dangerous virus.

Vincent Racaniello 21:00 You could.

Lex Fridman 21:00 And it's all in simulation.

Vincent Racaniello 21:01 Absolutely. So an example would be w- we have all these variants of SARS-CoV-2 arising, right?

Lex Fridman 21:07 Yeah.

Vincent Racaniello 21:07 Which, which look to be selected by, uh, immune responses. But we now, we know what amino acids are changing in the spike and how they block antibody bonding. You could simulate that.

Lex Fridman 21:20 Mm-hmm.

Vincent Racaniello 21:20 You could say, eh, the, what, what is the antibody looking at? You c- these are ... Where antibodies bind on proteins are called epitopes, right?

Lex Fridman 21:27 Mm-hmm.

Vincent Racaniello 21:28 You could map them all and change them in a simulation one by one and s- and go back and forth between the antibody and, and the virus. So all these ... Evolution is a, is what we call an arms race, right? The virus changes and then it evades the host, and then the host can change. The host takes longer to change though, unfortunately. It takes geological (laughs) time. But it can, and then the virus can change and it can go back and forth. And we can see evidence of this in s- in genome sequences of both viruses and their hosts. And so you can take a protein in a host that is a receptor for multiple viruses, and you can see all the impacts of virus pressure on it, and you could simulate that for sure. And that's just one thing that you could do. You could simulate, uh, changes in say a, an enzyme that makes it resistant to a drug and predict all the drug resistance. But, but the problem is people like me, the experimental virologists don't know how to do any of that. So we need to collaborate with people. I guess-

Lex Fridman 22:27 With all, with other humans.

Vincent Racaniello 22:29 (laughs) We do that. We do that.

Lex Fridman 22:30 Okay.

Vincent Racaniello 22:31 But with people from a field that we're not used to, like I suppose people who ... Would it be AI, I suppose?

Lex Fridman 22:39 Yeah, machine learning people.

Vincent Racaniello 22:40 Machine learning people. And you would say, "Look, this is the biological problem. Is there a way we can use your tools to attack it?"

Lex Fridman 22:46 The problem is those people are-

Vincent Racaniello 22:49 (laughs)

Lex Fridman 22:49 ... a- antisocial introverts that, uh, (laughs) -

Vincent Racaniello 22:53 (laughs)

Lex Fridman 22:53 ... that have a place like this and try to hide from other people in the world.

Vincent Racaniello 22:57 Yeah.

Lex Fridman 22:57 Very difficult to find in the wild. Um, okay.

Vincent Racaniello 23:00 (laughs)

Lex Fridman 23:00 So outside of doing amazing brilliant lectures online, you host and produce five, I would say, related podcasts, including my favorite This Week In Virology, also This Week In Parasitism, This Week In Microbiology, and so on. So you're a good person to ask, what are the categories of small things, small biological things, in this world that can kill you, kill us humans? Les- let's look-

Vincent Racaniello 23:31 (laughs)

Lex Fridman 23:31 You said like most viruses are friendly or at least not unfriendly.

Vincent Racaniello 23:36 Right.

Lex Fridman 23:36 But let's look at the unfriendly ones in viruses and bacteria and those kinds of things. When you look at the full spectrum of things that can kill you, can you kind of paint a brief picture?

Vincent Racaniello 23:47 Yeah, I think the, the big picture is that the things that can kill you are a minority of everything that's out there. And we're talking about molecules. So we have in us proteins that can kill us.

Lex Fridman 24:00 Yeah.

Vincent Racaniello 24:01 Prions that are just ... It's a protein in us and if it misfolds, it makes all of its other copies misfold, and then you, you die of a neurological disease.

Lex Fridman 24:11 Yeah.

Vincent Racaniello 24:11 That's pretty rare. Um, so but there are proteins that are viruses. And as I said, only certain ones can kill us.But even though, if we get those from animals, it's not straightforward. If you look at SARS-CoV-2, right, this is probably a once-in-a-hundred-year pandemic, I would say, equivalent to 1918-

Lex Fridman 24:33 Mm-hmm.

Vincent Racaniello 24:33 ... in its devastation, and in between there have been smaller pandemics of other, other viruses, but it doesn't happen all that often. So, we have a lot of viruses. We have a lot of bacteria of various sorts that can cause infections in us, and there's... It's, it's a limited number, right? Your streptococci and staphylococci and clostridia, and we could go on and on, but we know how to handle those as long as we have antimicrobials. It's just that we abuse them and we get resistance, so-

Lex Fridman 25:00 Mm-hmm.

Vincent Racaniello 25:00 ... that can be a problem. Then we have fungi, uh, not mushrooms, but much smaller fungi that m- multiply submicroscopic or just at the microscopic level. They can, you know, in dry climates of the US you can inhale their spores and they can grow in your lung if you're immunosuppressed and so forth. So, those are, uh, the, the tiny guys. And then we have parasites, which we, we do this week in parasitism, where single cells, even worms of various sorts, can invade you and cause all sorts of problems, and-

Lex Fridman 25:33 Uh, how, um... I was, like, kind of terrified to listen to that podcast. What, uh, what's it like? (laughs)

Vincent Racaniello 25:39 Well, I mean-

Lex Fridman 25:40 What... (laughs)

Vincent Racaniello 25:42 What you learn is that-

Lex Fridman 25:44 Yeah.

Vincent Racaniello 25:44 ... you can... You travel somewhere and you can get infected and bring it back home.

Lex Fridman 25:48 Yes.

Vincent Racaniello 25:48 Here in the US we do have certain kinds of parasites, but because of our lifestyle, we more or less have avoided them. For example, there is a parasite called Toxoplasma-

Lex Fridman 26:00 Mm-hmm.

Vincent Racaniello 26:00 ... which has, is infected most of the world actually, because a lot of people like to eat raw meat and you would get it from, from raw meat, and we're not as fond of that here in the US. Uh, we like to cook our meat, but that could be a consequence of eating raw meat.

Lex Fridman 26:16 Is that what leads to, what is it called? Toxoplasmosis?

Vincent Racaniello 26:19 Yeah, so toxoplasmosis, um, m- it's mainly... A big issue is if you're pregnant and you get Toxo, then your fetus is going to be very badly malformed. It's going to have brain defects and so forth. Um, and animals can get it as well. So, there are a lot of parasites of that nature which you often acquire by food.

Lex Fridman 26:43 Mm-hmm.

Vincent Racaniello 26:43 Eating food of different sorts. And it usually happens elsewhere. We just... On, on this week in parasitism we do a, a case. So, Daniel Griffin is our resident physician.

Lex Fridman 26:54 Mm-hmm.

Vincent Racaniello 26:54 He's a doctor, a real doctor, right?

Lex Fridman 26:56 (laughs)

Vincent Racaniello 26:56 And he... Every month he comes up with a case, "Okay, this is a person I saw." And last month this li- young lady had traveled somewhere, and she ate raw fish.

Lex Fridman 27:09 Yeah.

Vincent Racaniello 27:09 It was somewhere in Southeast Asia or something, and she ended up with, uh, with, with red bumps all over her skin, and it turned out it was a parasite from the fish that moved around in her and... And they're very easy to cure. We have... If you have the right doctors and the right drugs, you can cure all these things.

Lex Fridman 27:24 Uh, what about diagnose? Like, connect the red spots to the fact that it's a parasite.

Vincent Racaniello 27:28 Very easy to... if you have the right diagnostics. Now, Daniel often goes to parts of the world where they don't have diagnostics and he has to use other mechanisms. He may have to take a bit and look at it under a microscope, and then he may not be able to get the drug depending on where he is. But if these... But often he sees patients who come back to the US and they get diarrhea or they have a fever and he says, "Where have you been?" And he can put two and two together. And so we let our listeners do that, and they all send in guesses, and it's wonderful to hear them-

Lex Fridman 27:57 That's awesome.

Vincent Racaniello 27:57 ... go through this. So, there are a lot of parasites-

Lex Fridman 27:58 Solve the puzzle and solve the case then.

Vincent Racaniello 27:59 ... that can get you. You have to be careful about eating when you go overseas.

Lex Fridman 28:03 And water too?

Vincent Racaniello 28:04 Water as well. And, you know, in parts of Africa, there are parasites in the lakes, and if you go swimming they can invade you, and in fact they can go into your hair follicles and burrow in and get into your bloodstream.

Lex Fridman 28:15 That's exciting. (laughs)

Vincent Racaniello 28:17 So, um, Daniel is interesting 'cause he's very adventurous and he doesn't... he's not afraid of any of this, uh... So, there's a famous lake in Africa, Lake Malawi, whi- which harbors a lot of these parasites. And he said, "Oh, yeah, yeah. I just make sure I towel off vigorously when I get out." (laughs)

Lex Fridman 28:33 (laughs) Vigorously.

Vincent Racaniello 28:33 "And get, get rid of them." And that was the name of an, uh, an episode. But, you know, food is-

Lex Fridman 28:38 Towel off vigorously?

Vincent Racaniello 28:39 You know sushi? You can, uh, you can get worms from sushi.

Lex Fridman 28:43 Yeah.

Vincent Racaniello 28:43 And the solution is to freeze it. And many sushi restaurants now have liquid nitrogen. They s- they snap-freeze their sushi, and that kills all the parasites. And a study was actually done in Japan showing that freezing does not alter the taste of sushi because it's-

Lex Fridman 28:59 Oh, wow.

Vincent Racaniello 28:59 ... sup... you see a big industry there. (laughs)

Lex Fridman 29:01 Wow. That's brilliant. That's brilliant. Yeah, um, I, I was thinking about, you know, I'm so boring and bland that... Especially when I... I'm now in Texas here and I've been eating quite a bit of barbecue, I realized I really haven't explored the culinary world, and I've been curious to travel and taste different foods. Is, is there something you can say by way of advice, uh, you know, channeling Daniel, I guess? If you were to travel in the world, if eating is the thing that gets you the parasites-

Vincent Racaniello 29:34 Hmm.

Lex Fridman 29:34 ... what's good advice for eating in, uh, strange parts of the world, Mongolia, India, China? Is there something you could say by way of advice?

Vincent Racaniello 29:43 I think, I think Daniel would say, "Make sure your food is cooked." Right?

Lex Fridman 29:47 Cooked. But that's so boring.

Vincent Racaniello 29:49 Yeah, it's, it's unfortunate, and he would agree with you, um, because, you know, ve- many vegetables are, are delicious. Salads even are delicious not cooked, but they can have parasites in them. Meats, fish, people like to have, uh, uncooked fish. So, if you want to be really safe and boring, just make sure everything is cooked. Now, we have a case this week on TWIP-

Lex Fridman 30:10 Mm-hmm.

Vincent Racaniello 30:11 ... of a young man who went... I forgot where he went, but he stayed in a hotel. I think he... Oh, Oaxaca, Mexico.

Lex Fridman 30:17 Mm-hmm.

Vincent Racaniello 30:18 Stayed in a hotel, and he said to Da- he came back with diarrhea and fever.And he said that, "I don't know where... I stayed in the hotel. I just ate hotel food, lots of vegetables and fruits." And probably they weren't washed with clean water. You know, he got something from that. The bottom line is most of these infections with parasites can be diagnosed and you can be treated and you'll be fine. So if you really wanna experience the cuisine, I don't think you should worry about it. That's what Daniel would say.

Lex Fridman 30:50 Let's return to the basics. We're gonna jump around all over the place. What are the basic principles of virology? Maybe a good place to start is what is a virus?

Vincent Racaniello 31:02 That's great. I mean, I talk in my first lecture for 20 minutes before I get to that.

Lex Fridman 31:06 (laughs)

Vincent Racaniello 31:07 But, and I wonder if I should put it up front, but it's kind of a boring definition. So if you do that first, people will turn off. So first you tell 'em about all the millions and billions of viruses around.

Lex Fridman 31:18 (laughs)

Vincent Racaniello 31:19 So a virus, we have a very specific definition 'cause it's different from everything else on the planet. Um, 'cause first of all, it's a parasite. It takes... A parasite means you take something from someone else. You know, we have human parasites who take money from others, right? But in biological terms, uh, a parasite takes something from the host that would, the host would otherwise use energy or some building block or something like that.

Lex Fridman 31:46 There's never really a symbiotic relationship between a virus and a host.

Vincent Racaniello 31:50 Well, there, there can be.

Lex Fridman 31:52 But-

Vincent Racaniello 31:52 So that's the dichotomy, I think, is that we define them as parasites, yet I just told you 20 minutes ago that many viruses are probably beneficial. So I think what it means is we, at some point, we're gonna have to change our definition, right? Because, uh, after all definitions we make are just constructs that make it easier for us to study, that not necessarily represent what's right.

Lex Fridman 32:16 Yeah.

Vincent Racaniello 32:17 Right?

Lex Fridman 32:17 Like, uh, like Pluto was a planet at first, and now it's not a planet anymore.

Vincent Racaniello 32:21 A planet.

Lex Fridman 32:21 And a lot of people are very upset.

Vincent Racaniello 32:23 But it's only according to us. There may be another race living somewhere else who thinks it's a planet, right?

Lex Fridman 32:28 Well, maybe that's why viruses are attacking humans. They're very angry that we're, uh, calling 'em parasites.

Vincent Racaniello 32:34 (laughs) So right now, our definition includes parasite because a virus cannot do anything without a cell. If, if I, if this mug were full of viruses, it would not do anything for years. It would eventually probably lose its infectivity, but it's not gonna reproduce here. It needs cells and, you know, to the first people who discovered viruses, that was astounding that they didn't just reproduce, divide on their own like bacteria. So a virus needs to get inside of a cell, inside the cell. It can't just hang around on the surface. It needs to get in, in order to make more of itself. And so we call it an obligate intracellular parasite because it needs to get in a cell and then it takes things from the cell in the form of all kinds of molecules and processes and energy and so forth to make new viruses.

Lex Fridman 33:23 Uh, obligate means it's obligated to be inside the cell. Interesting.

Vincent Racaniello 33:26 Absolutely.

Lex Fridman 33:27 Okay.

Vincent Racaniello 33:27 It will not reproduce outside of the cell. So this mug of viruses is, can in no way be living, in my opinion. However, once it gets inside of a cell, now the cell is a virus-infected cell. It's alive.

Lex Fridman 33:42 Mm-hmm.

Vincent Racaniello 33:42 So a virus, in my view, has two phases, right? It's this non-living particulate phase that everyone is used to. Uh, I, I'll send you, you need a virus for your table. I, I'll send you a nice model. I think it would look good here.

Lex Fridman 33:55 Which, yes, definitely.

Vincent Racaniello 33:56 You know, to go with all this other stuff.

Lex Fridman 33:57 Yeah. Well, these are all mechanical. There's no biology here.

Vincent Racaniello 34:01 So you wouldn't want a virus here?

Lex Fridman 34:02 No, I'd want a vi- of course.

Vincent Racaniello 34:04 I'll send you one and then-

Lex Fridman 34:05 I would love that.

Vincent Racaniello 34:05 ... you could, you could look at it because now that we have the three-dimensional structures solved by structural biologists, we take the coordinates and we put it in a 3D printer, and you can make amazing models, right?

Lex Fridman 34:17 Mm. And, and, and-

Vincent Racaniello 34:18 Of any virus.

Lex Fridman 34:18 And so there's a huge variety of viruses in terms of-

Vincent Racaniello 34:21 Huge. That we know of.

Lex Fridman 34:23 Yeah.

Vincent Racaniello 34:23 Which is only a fraction of what's out there.

Lex Fridman 34:24 What's the category? So there's RNA, there's DNA viruses. What, what are those?

Vincent Racaniello 34:29 So there are-

Lex Fridman 34:29 What's DNA and RNA? (laughs)

Vincent Racaniello 34:30 ... two, two, two broad categorizations. RNA and D, these are genetic material.

Lex Fridman 34:36 Mm-hmm.

Vincent Racaniello 34:37 Can be two different chemicals. So RNA... Well, everything else on the planet besides viruses is all DNA based. You and I are DNA based. Everything on the planet today is DNA based, except some viruses are RNA based. And that's because, as I mentioned earlier, the first life that arose on the planet was RNA based.

Lex Fridman 34:57 Yeah, so these are like old school viruses.

Vincent Racaniello 34:59 These are old school. These are, we call relics, yeah. (laughs)

Lex Fridman 35:02 Yeah.

Vincent Racaniello 35:02 Relics. And this has got a name. This is called the RNA world, which I think is great.

Lex Fridman 35:06 Is it big still? Or are they, are the relics dying out?

Vincent Racaniello 35:09 Oh, no. The relics, in my opinion, are the most successful viruses, the RNA viruses, and SARS-CoV-2 is an RNA virus.

Lex Fridman 35:16 Mm.

Vincent Racaniello 35:16 We can talk about why they're so successful, but you have, broadly speaking, viruses with RNA genetic information, which are relics. Of course, they're contemporary. They have adapted to the modern world (laughs) and the modern organisms living in it. And then we have DNA-based viruses, which are extremely conservative and slow. They're very successful. You know, everyone has a herpes virus infection, but they are, they don't get the news like the RNA viruses, uh, do. The HIVs and the influenza viruses and the, um, SARS coronaviruses, they get all the press and they're RNA based 'cause RNA lets you change more so than DNA.

Lex Fridman 35:54 So they, they evolve much faster, the RNA viruses.

Vincent Racaniello 35:57 Much faster. And in fact, when I, I have an electron evolution, I don't know if you've listened to that one. You should. It's really, I think it's really interesting. RNA viruses exist at their error threshold, which means they can't make any more mutations when they reproduce. Otherwise, they're dead. They would-

Lex Fridman 36:19 Wow.

Vincent Racaniello 36:19 ... go extinct. They're, they're evolving at their error threshold. DNA viruses are...... hundreds of times lower than their error thresholds.

Lex Fridman 36:27 Wow.

Vincent Racaniello 36:28 And we know this. We can do an experiment to find that out. Now, why that is, is a good question. Uh, but, uh, that's the, that's the reason why RNA viruses are far more successful.

Lex Fridman 36:39 Yeah.

Vincent Racaniello 36:40 They infect many more hosts, and they're very, I would say, slippery. They can change hosts really quickly because in any animal harboring an RNA virus, like, let's say a bat in some cave somewhere, it's not just one genome. S- it's millions of different genomes of all, all kinds, with, all within the framework of, say, coronavirus, but they're all different. And one genome in there might just be right for infecting a person if it ever encountered that person. I mean, that's the thing that-

Lex Fridman 37:09 Or there could be a large number, just a tiny fraction, but a large number of, uh, of them, and they're all operating-

Vincent Racaniello 37:15 That's right.

Lex Fridman 37:16 ... at the, at the threshold of, uh-

Vincent Racaniello 37:18 That's right.

Lex Fridman 37:18 ... error.

Vincent Racaniello 37:18 That's right.

Lex Fridman 37:18 That's fascinating. It's like little, like, uh, it's like startups, little entrepreneurs, like a startup world.

Vincent Racaniello 37:23 Yes. And many of them fail.

Lex Fridman 37:25 Yeah. Many of them fail.

Vincent Racaniello 37:25 Many of them change and fail.

Lex Fridman 37:26 And then there's the DNA viruses that are like the IBM and the Google.

Vincent Racaniello 37:29 Exactly. Exactly.

Lex Fridman 37:30 (laughs) The big corporations-

Vincent Racaniello 37:31 It's very good. I like that.

Lex Fridman 37:32 ... that become conservative with the bureaucracies and all that kinda stuff, so they-

Vincent Racaniello 37:35 They have a lot of baggage.

Lex Fridman 37:36 Yeah.

Vincent Racaniello 37:36 Yeah, yeah. It's expensive for them to reproduce. Yeah.

Lex Fridman 37:39 (coughs)

Vincent Racaniello 37:39 And they don't move quickly. Yes, the RNA viruses are the fast-moving members. So, that's what a virus is. We call them o- oblit intracellular parasites. And then I told you, there's DNA and RNA, but then let's go further. The, the nucleic acid's not naked because naked nucleic acid in the world isn't good. I mean, it, it existed in the, in the pre-cellular world, but there probably weren't a lot of threats to it then. Naked nucleic acid doesn't last long in the environment, so they're, they're covered. The nucleic acid is c- is covered. It can be covered with a protein shell, a pure protein shell-

Lex Fridman 38:15 Mm-hmm.

Vincent Racaniello 38:15 ... or it can have a membrane around it, which would be, uh, lipids from the host cell.

Lex Fridman 38:22 So, uh, lipids, so it's, it's a fatty membrane.

Vincent Racaniello 38:26 Fatty membrane. Yeah, so our cells are coated with fatty membranes, right? Our cells-

Lex Fridman 38:30 Mm-hmm.

Vincent Racaniello 38:30 The outer plasma membrane, right?

Lex Fridman 38:32 Yeah, but-

Vincent Racaniello 38:32 That's the same.

Lex Fridman 38:32 But viruses can be too? So, they're-

Vincent Racaniello 38:34 Viruses actually

Narrator 38:34 They can take it.

Lex Fridman 38:35 ... cells, but without the ability to do the mitochondria stuff.

Vincent Racaniello 38:38 Some, some are. Some are... uh, they don't have nuclei. They don't have mitochondria.

Lex Fridman 38:42 Yeah.

Vincent Racaniello 38:42 But they do have a nucleic acid. They ag- they have a membrane, and then of course there're spikes in the membrane-

Lex Fridman 38:48 Mm-hmm.

Vincent Racaniello 38:48 ... that allow them to attach to cells. And so that completes our, our two different kinds of viruses.

Lex Fridman 38:53 So, they have, they all have, like, attachment mechanisms-

Vincent Racaniello 38:55 Yeah.

Lex Fridman 38:56 ... like ways to, like keys into the, into the safe.

Vincent Racaniello 38:59 Yep. They all have to get into cells. There are, there are a couple of exceptions, though. Uh, there are viruses of fungi, uh, and, uh, plants. So, let's do the fungi. Fr- fungi would be like yeast. Uh, the virus of yeast cell wall is pretty hard to get through, so viruses typically don't attach to a yeast and get inside. Rather, they, they just live in the yeast forever.

Lex Fridman 39:25 Yeah.

Vincent Racaniello 39:25 And they multiply as mostly nucleic acids, and as the yeast divide, they go into the daughter cells, and that's how they exist. Plant viruses, also the plant cell wall would be very hard to get across with a, by binding a protein. So, plant viruses get into plants either by pests that inject them in. Th- they're sucking sap out, and they inject virus at the same time. Or farmers, they have contaminated farm equipment, and they roll over the plants and introduces viruses. So, those fungi and plant viruses, they don't have this specific receptor binding to get them into the cell, but everything else, yeah, the virus binds to something on the surface very specific. It's taken into the cell 'cause that's what cells do. When things bind their exterior-

Lex Fridman 40:10 Mm-hmm.

Vincent Racaniello 40:11 ... they take it in 'cause in most cases, it's good. It's something they need. And so the virus slips in. I guess you'd call that a Trojan horse, right?

Lex Fridman 40:19 Trojan horse. It's so hard-

Vincent Racaniello 40:20 (laughs)

Lex Fridman 40:20 ... to not anthropomorphize this whole thing.

Vincent Racaniello 40:23 It is hard. Yeah.

Lex Fridman 40:24 So obviously, they don't know any of this. It's not an actual Trojan horse, so th- they, they're not getting actually tricked in the way humans trick each other.

Vincent Racaniello 40:34 No, it's all passive, and it's just through so many years of evolution, it's, you select something that works, and it continues. And what survives then goes on with a, uh, perhaps a slightly-

Lex Fridman 40:47 Passive.

Vincent Racaniello 40:47 ... different approach.

Lex Fridman 40:48 I, I love this idea of pass-... of course, according to Sam Harris, uh, so from a sufficiently intelligent alien civilization observing humans, our behavior might seem passive too-

Vincent Racaniello 40:58 (laughs)

Lex Fridman 40:58 ... 'cause they understand fully exactly what we're doing, and then there's no free will. We're all just operating in the same way-

Vincent Racaniello 41:04 Could be.

Lex Fridman 41:04 ... a cell does, but just at a much higher level of complexity. Yeah. (laughs) So, I love the d- the distinction between active and passive because-

Vincent Racaniello 41:11 I mean, the point is that I think anthropomorphizing to a certain extent is fine 'cause it helps people understand. But when you start to say, "I think that the virus is doing that because," then you're putting a human lens on it, and you may be wrong-

Lex Fridman 41:26 Yeah.

Vincent Racaniello 41:26 ... because you don't know why things happen for a virus. Um, so right now, we have variants emerging, and people say, "Well, I think it's because the antibodies are selecting for variants." Eh, that's a good idea, but it may not be the only thing that's going on.

Lex Fridman 41:41 Why do you start imagining them coming to the table negotiating? (laughs)

Vincent Racaniello 41:45 (laughs)

Lex Fridman 41:45 Yeah. It, you get into trouble with that.

Vincent Racaniello 41:48 That's why, uh, that's why I tell my students, "Be careful about the anthropomorphizing because you're gonna apply your values to a virus, and you have different value. You're a human, and you have, wh- what you think is the reason for this outcome may not be right. That's all. Just be open-minded-

Lex Fridman 42:03 Yeah.

Vincent Racaniello 42:03 ... about it."

Lex Fridman 42:04 In both directions. I actually, one of the things I push back on is in, in the space of robotics, people, most people in robotics try to n- not anthropomorphize.

Vincent Racaniello 42:14 Mm-hmm.

Lex Fridman 42:14 For example, they don't give a gender or a name to robots. They really try to see it as a machine. And to me, that makes sense in one s- in, in one way, but it totally doesn't make sense in another. If that robot is to interact, operate in the human world and interact with humans, we have to-... we have to anthropomorphize it in order to understand as an engineering problem-

Vincent Racaniello 42:38 Hmm.

Lex Fridman 42:39 ... h- how should it operate in a human world? Now, the difference with viruses, the scale of operation, it doesn't make sense to treat them as human-like 'cause the scale of operation is much smaller. But with robots, you're in the same-

Vincent Racaniello 42:53 Sure.

Lex Fridman 42:53 ... timescale and the same spatial scale.

Vincent Racaniello 42:55 And of course, in the movies, they always give them names and personalities, right?

Lex Fridman 42:58 Yeah. Well, yeah, that's the mo- uh, but that's my argument is we should do the same when you're trying to solve the engineering problem-

Vincent Racaniello 43:05 Hmm.

Lex Fridman 43:05 ... of robotics too. It's not just for the movies. Well, let me ask you this-

Vincent Racaniello 43:08 Hmm.

Lex Fridman 43:08 ... because you said controversially, not really-

Vincent Racaniello 43:12 (laughs)

Lex Fridman 43:12 ... that, uh, viruses are not living. Um, uh, defend yourself.

Vincent Racaniello 43:19 (laughs) So-

Lex Fridman 43:19 Are viruses alive or not?

Vincent Racaniello 43:20 So I've seen many people say, "Oh, they have to be. They, they have nucleic acids. They evolve. They mutate." That's all true, but they don't do it on their own. The particles in my mug are just not doing any of that unless they get into a cell.

Lex Fridman 43:33 Mm-hmm.

Vincent Racaniello 43:34 So a virus-infected cell is alive. I totally agree with that because in fact when a virus gets in a cell, it converts it into a virus-making factory, if you will. It's no longer a cell. It's a, some people call it a viral cell. I don't really like that, but it's fine. So, that's what I'm talking about. The particle is not alive. You can have your virus-infected cell as, as alive, but the particle, it just would not do anything forever without getting inside of a cell. Well, once it's in a cell, it's, it, it, it is alive then, but it's no longer a particle. It's taken apart and nucleic acid is moving around the cell. It's making proteins. Eventually, it makes new particles. And then those particles released from the cell, they're not living anymore. So, it's kinda, I think it's kind of like a spore, a spore of a, uh, or a seed. Although it's, the seed just doesn't work because the seeds, the cells in the seed have the ability to make their own energy and so forth. But a bacterial spore, and it's the same thing, doesn't do anything unless you add water and nutrients, and then it starts to divide. But it doesn't need to get into a cell. It's very different from a virus. So that's why the particle... And when people think of virus, they're always thinking of the particle. And that's why I say it can't be alive 'cause the particle can't do anything on its own. But if you think of a virus as an organism with a particle phase in a, in a part, in a cell, then it makes sense to be alive.

Lex Fridman 44:58 Mm-hmm. And by the way, when you say particle, you're referring to that structure that you've been mentioning, some-

Vincent Racaniello 45:02 Right.

Lex Fridman 45:02 ... kind with membrane and not, that, that's-

Vincent Racaniello 45:04 Right.

Lex Fridman 45:04 ... that's been called, what is that? A virion particle or something like that?

Vincent Racaniello 45:06 Virion.

Lex Fridman 45:07 Virion.

Vincent Racaniello 45:07 So it's, what you should have here, I'll send you one, and then you can refer to it there.

Lex Fridman 45:11 What's the sexiest one to have? Like, what, what, in terms of particles to have on the table? (laughs)

Vincent Racaniello 45:18 Well, unfortunately, the ones that you can 3D print, uh-

Lex Fridman 45:21 Oh, they're not going to be super 'umph'.

Vincent Racaniello 45:23 They're, they're only, they're the ones w- that we know the structures of, right? So someone sent me, last year, a 3D model of SARS-CoV-2, and it's beautiful. It's actually cracked open so you can see the RNA-

Lex Fridman 45:34 Oh, wow.

Vincent Racaniello 45:34 ... and the spikes are sticking out, and they even put some antibody sticking onto the spikes. And I mean-

Lex Fridman 45:41 That's super cool.

Vincent Racaniello 45:41 ... when I show this on a livestream, people love this. They go, "Oh my God, that's beautiful." And it is.

Lex Fridman 45:45 Mm-hmm.

Vincent Racaniello 45:45 It's absolutely gorgeous. I have that. I have my virus that I worked on most of my career, poliovirus. I have a 3D model of that, which I actually just had made. It's gorgeous. And you can have it made in any color you want,

Vincent Racaniello 45:57 right?

Lex Fridman 45:57 What would you say is the most fascinating, terrifying, surprising, beautiful virus to you? So, of all the viruses you looked at, sometimes when you just sit late at night with a glass of wine looking over the sunset, which virus do you think about?

Vincent Racaniello 46:16 So, uh, uh, fulfilling all of those adjectives is hard, right? Fascinating, exciting, terrifying.

Lex Fridman 46:25 Well, the terrifying is an optional one, I think, 'cause may- maybe that puts a lot of pressure.

Vincent Racaniello 46:30 Uh, see, terrifying is, to me, it, it... I'm not terrified because I think we can handle, uh, most viruses, uh, as you see with this brand new one that emerged a year ago, we, we can handle it.

Lex Fridman 46:42 From a virology perspective.

Vincent Racaniello 46:44 Yeah, I mean, the, the human perspective (laughs) is a different story, right? That's always an issue, but... Um, so I, I think there, there are a couple of different categories of virus. So we could do the, the terrifying, and I think rabies is a terrifying virus because unless you're vaccinated, 100% certainty you're gonna die.

Lex Fridman 47:05 (laughs)

Vincent Racaniello 47:06 So you get bitten by a rabid raccoon or bat or dog, whatever, and, you know, and there's still 70,000 deaths a year of rabies throughout the world because there are a lot of feral dogs running around that are infected. Unless you're vaccinated, you're gonna die. There's nothing we can do. But we do have a vaccine which we can actually give you even after you've been bitten, which is the only vaccine that works that way. It's a therapeutic, right? It will treat your illness 'cause the disease takes so long to develop, d- you know, eventually you get all kinds of neurological issues and paralysis and so forth. But it takes time, and you can be vaccinated, and it will prevent that in the meanwhile. So people always say, "What's the most lethal virus? Is it Ebola?" I said, "No, it's actually rabies." It, unless you're vaccinated, it will kill you.

Lex Fridman 47:55 But it's-

Vincent Racaniello 47:56 (laughs)

Lex Fridman 47:56 Maybe it's, uh, is good to linger 'cause we'll talk about vaccines, uh, a few times today.

Vincent Racaniello 48:01 Sure.

Lex Fridman 48:01 It's good to linger on cases where vaccines have clearly, um, undoubtably helped human civilization.

Vincent Racaniello 48:11 Mm-hmm.

Lex Fridman 48:12 And it seems like, uh, rabies is a good example.

Vincent Racaniello 48:15 Oh, rabies is great because everyone knows what happens when somebody gets rabies, right? They have fear of water, f- hydrophobia. Uh, your, your body becomes spastic and stiff and jerks around and-... you lose consciousness, you can't, uh, no, no more, uh-

Lex Fridman 48:35 It's, it's not a fun ride to death.

Vincent Racaniello 48:36 It's horrible. It's a horrible way to die. So I think most people know that. It's been popularized enough in, in media, right? So that nobody would probably object to getting... "Oh, I, I was just bit by this raccoon and it ran off." "Okay, well, we should assume it's rabid. We should immunize you." And most people are okay with that-

Lex Fridman 48:55 Yeah.

Vincent Racaniello 48:56 ... 'cause they know the consequences. And it's also pretty rare, right? It's not like something that you're trying to get into the arms of, you know, 3- 250, 300 million Americans. That's hard, but-

Lex Fridman 49:08 Yeah.

Vincent Racaniello 49:08 ... the, the, the, the few thousand every year is, it's easy.

Lex Fridman 49:11 So the transmissibility is difficult, right? It has to... Oh, it's not, it's not airborne, so-

Vincent Racaniello 49:16 It's not airborne. It's just... You has to be, you have to be bitten. Although some, some people claim you could walk into a cave and the bats, you know, breathing out rabies virus could infect you. But I don't really think that's well s- that's well substantiated.

Lex Fridman 49:32 Yeah.

Vincent Racaniello 49:32 I think it's a bite.

Lex Fridman 49:33 How would you do a study on that? (laughs)

Vincent Racaniello 49:34 Yeah. It's, it's very hard to do. You'd have to collect the vapors in the cave and show that they're infectious, which... And by the way, (laughs) someone emailed me the other day, you'll like this. They say, "Why can't we just immunize all the bats in the world (laughs) against these viruses?" And I said, "Well, how would you do that? They're caves everywhere, right?"

Lex Fridman 49:52 Yeah.

Vincent Racaniello 49:53 A- a- and he said, "Well, maybe you could just go and aerosolize them." (laughs)

Lex Fridman 49:56 Yeah.

Vincent Racaniello 49:57 That's pretty dangerous.

Lex Fridman 49:58 And then, and then all the bats should have vaccine passports to make sure that they're all v-

Vincent Racaniello 50:02 Yeah, and so you have to get their consent before you do it. But you, we do immunize wildlife against rabies.

Lex Fridman 50:11 I see.

Vincent Racaniello 50:12 We have rabies vaccines for wild animals. There are a whole bunch of them that get rabies. And we put it, uh, we put it in bait and drop it from helicopters in the woods, and it drops down-

Lex Fridman 50:22 Wow.

Vincent Racaniello 50:22 ... the incidents of, uh, rabies in people.

Lex Fridman 50:24 Wow.

Vincent Racaniello 50:25 You know, people hiking get bitten and so forth, it drops the incidents. So we can do that.

Lex Fridman 50:29 I didn't know that. I always wondered how much medical care we're doing for animals in the wild because... And I've recently become more and more aware that animals are living in extreme poverty.

Vincent Racaniello 50:41 Mm-hmm.

Lex Fridman 50:41 Right? Like, you don't know... You think, like, natural, it's great. Um, you know, like, uh, like, when animals are living on a farm, it's terrible. But then you, you also have to compare to, like, what life is like in the... Or like the zoo. You have to compare what life is like in, in the wild. And, and-

Vincent Racaniello 50:58 Sure. Well, uh, life in the wild is very tough, I think. I mean, most animals have to... Well, carnivores anyway, they have to catch their food every day.

Lex Fridman 51:05 Yeah.

Vincent Racaniello 51:05 Right?

Lex Fridman 51:06 And then there's vi- the viruses there.

Vincent Racaniello 51:08 They have viruses as well.

Lex Fridman 51:09 All of that.

Vincent Racaniello 51:09 So the rabies immunization is the only one I'm aware of for wild animals. Um, we do immunize lots of other animals. Uh, we immunize chickens and pigs and cows. Even fish, farmed fish. We, uh, we pick each fish up and give it an injection, you know, when it's a small fish. Um, but that's mostly so that the farmers get a good yield. We don't really care about the animals, right? We want a good yield for market. And then there's some examples where we immunize animals to prevent spillovers into people.

Lex Fridman 51:49 Mm-hmm.

Vincent Racaniello 51:50 So there's a disease called Hendra in Australia, which, um, w- w- was discovered in the '90s. And it turns out there's a... There are bats, fruit bats, that have this virus, and the bats are fine, but sometimes they fly into horse stalls and the horses get infected. These are... In Australia it was initially race horses, which are very expensive, right? The horses got infected and they died, and the humans who would take care of them would die also. So now they immunize the horses to prevent the h- well, to save the horses. Probably that's the motivation-

Lex Fridman 52:21 Mm-hmm.

Vincent Racaniello 52:21 ... 'cause these horses are hundreds of thousands of dollars, right? And then the people don't get sick because the horses don't get sick. So you don't wanna immunize all the people because it's too rare, but-

Lex Fridman 52:30 Mm-hmm.

Vincent Racaniello 52:31 ... that approach is called the One World Health approach, which means everything's connected on the planet and we have to think of everything in the grander scheme, not just us.

Lex Fridman 52:41 Yeah, so you can immunize some things along the trajectory that a virus would take.

Vincent Racaniello 52:45 Exactly. So it's-

Lex Fridman 52:45 So it's not something... Some living beings, yeah.

Vincent Racaniello 52:48 Um, in the Arabian Peninsula, they have a MERS coronavirus issue. Every month there are a couple of cases where a camel will infect a human, and the human can get very sick. It's respiratory disease, very, very much like COVID. And so camels are very common there. They're used... They're raced, they're used as pets, they're eaten, so there's a lot of human-camel contact. But the number of cases are rare, two, two a month, so you don't wanna immunize all the humans, so the idea would be to immunize the camels.

Lex Fridman 53:19 Mm-hmm.

Vincent Racaniello 53:20 So.

Lex Fridman 53:20 (laughs) I like it. So, okay, so you put rabies.

Vincent Racaniello 53:23 (laughs) .

Lex Fridman 53:23 But Ebola, um, also is, is a, a famously deadly one.

Vincent Racaniello 53:29 Right.

Lex Fridman 53:30 What is it? It kills like, I don't know, 50, 60% of its-

Vincent Racaniello 53:33 Could be 50 to 90, but that's in Africa where the healthcare isn't great. What you saw when they... When the cases of Ebola came to the US, we, we could take care of it. We knew how to take care of it. We had fancy hospitals and so forth, and now we have a vaccine. So we can... And the vaccine is really good, but there are many governments in Africa who, that are suspicious of, of us and they don't want to use our vaccines, so they-

Lex Fridman 53:59 So there's a vaccine for Ebola?

Vincent Racaniello 54:00 There is, yeah.

Lex Fridman 54:01 And, uh, the effectiveness and safety of it? How much is understood?

Vincent Racaniello 54:06 So this is difficult because there's not a lot of Ebola, right?

Lex Fridman 54:11 Yeah.

Vincent Racaniello 54:12 It's not a continuous ongoing thing. There are sporadic outbreaks here and there.

Lex Fridman 54:16 Of a few thousand people.

Vincent Racaniello 54:17 At most, at most. Usually a few hundred. And the biggest ever... In fact, this is why we didn't for years have an Ebola vaccine. The US military together with Canada developed an Ebola vaccine for service people, right? They wanted to say, "Well, we're sending people into these Ebola areas, we want a vaccine for them." So they had developed it through all the preclinical, which means before it goes into people.... and that stopped because there was no money to do a phase one, and a phase two, and a phase three. In fact, for a phase two and three, you need to have infections going on 'cause you're looking at how well the vaccine prevents infections, right? So, then there was a West African outbreak in 2015. 2013, 2015, the most cases ever, 25,000, so they got to test the vaccine.

Lex Fridman 55:02 Mm-hmm.

Vincent Racaniello 55:03 Um, but they only put it in a few thousand people. It's not like it's been in hundreds of thousands of people (laughs) like the COVID vaccines has been. So, it's, it looks like it's, it has high efficacy, um, but we'd like to have more data. Side effects may be, are not so great. Then there are a couple of different available vaccines, some have been tested more than others. Uh, I would say this would probably not be widely accepted in the US-

Lex Fridman 55:30 (laughs)

Vincent Racaniello 55:31 Um...

Lex Fridman 55:31 ... but then (laughs) neither would be (laughs) something over 50%, uh, deadliness, uh, of a virus.

Vincent Racaniello 55:38 No, no. I think if you were...

Lex Fridman 55:39 (laughs)

Vincent Racaniello 55:40 Uh, in fact, many physicians work in countries that have Ebola, so they get vaccinated because they understand the choice.

Lex Fridman 55:46 Yeah.

Vincent Racaniello 55:46 Right?

Lex Fridman 55:46 Right. It's always about the choice. Um, so...

Vincent Racaniello 55:49 So, then one more thing. To answer the-

Lex Fridman 55:51 Yeah.

Vincent Racaniello 55:52 ... interesting, what, what are some of the viruses you really are fascinated by? There are a number of viruses that have clearly been shown to alter host behavior, and that's how they spread.

Lex Fridman 56:04 Mm-hmm.

Vincent Racaniello 56:04 I think those are fascinating. For example, there's some viruses of plants that are spread by aphids, and the aphid bites the plant, the virus reproduces in the plant, then it somehow engineers the plant to give off volatile organics to attract more aphids-

Lex Fridman 56:26 Mm-hmm.

Vincent Racaniello 56:26 ... which will spread the virus.

Lex Fridman 56:29 Whoa.

Vincent Racaniello 56:30 Isn't that amazing? (laughs)

Lex Fridman 56:31 Yeah.

Vincent Racaniello 56:32 So, that's altering the behavior.

Lex Fridman 56:33 Reminds me of Twitter. (laughs)

Vincent Racaniello 56:35 Altering because somehow the virus, uh, infecting the plant cells gives off these organics-

Lex Fridman 56:39 So interesting.

Vincent Racaniello 56:40 ... and it attracts the aphids. And then, furthermore, somehow the, when the aphid bites, it tastes horrible, so they immediately leave with the virus they've just picked up and go to another plant to spread it. So, they're attracted and then repulsed at the same time.

Lex Fridman 56:53 And obviously you don't want to anthropomorphize this like a strategy they're taking on. Somehow this worked out.

Vincent Racaniello 56:58 It worked out this way. It just evolved. And, you know, evolution is sometimes hard to trace, right?

Lex Fridman 57:03 Mm-hmm.

Vincent Racaniello 57:04 Like Da- Darwin famously said he could never figure out how an eye evolved from a single cell, right? But-

Lex Fridman 57:10 Yeah.

Vincent Racaniello 57:10 ... it did.

Lex Fridman 57:11 The more complicated, complex the, the holistic organism is that the virus invades, the, the less able it is to control that organism, right? So, I wonder if there's viruses that can control human behavior, um, you know, to induce more spread of the virus.

Vincent Racaniello 57:32 (sighs) Well, I don't-

Lex Fridman 57:33 What-

Vincent Racaniello 57:33 ... see why not.

Lex Fridman 57:34 There's not enough humans, I suppose, to, like, evolve through?

Vincent Racaniello 57:37 Well, we can't do the experiment to test it, right? We have to observe.

Lex Fridman 57:40 (laughs)

Vincent Racaniello 57:40 And that's always hard when you're observing 'cause there's so many things that can confound what you're looking at.

Lex Fridman 57:45 Yeah, change human behavior, yeah.

Vincent Racaniello 57:47 I mean, there's so many things that impinge on our behavior. But, um, yeah. It's, I think it's possible. Um, I think it's highly possible. If it does it in a plant, why not change some other organism's behavior? I think it's fine. Anyway, those fascinate me. There are lots of examples of those that are fascinating, and how they work, people are trying to figure out. But there's not a lot of money to work on, you know, insect and plant viruses unless you're going through the USDA. So, they don't get a lot of, uh, work moving forward.

Lex Fridman 58:17 Well, is there, if you understand some of those viruses, is that transferrable to human viruses, that understanding?

Vincent Racaniello 58:24 I think some of it could be, sure. I think the general principles, for example, how the, how does the virus cause volatile organics to be made? It must be turning on some genes, and you could learn principles from that, how, how the virus might do that. Sure, I think everything is broadly applicable. To s- so to say it's not useful to study viruses of insects and plants is just wrong, because in science, you probably know this, maybe in your field it's the same, if you're curious, you're gonna run into interesting things-

Lex Fridman 58:55 Okay.

Vincent Racaniello 58:55 ... that you never f- planned on, right?

Lex Fridman 58:57 That's why people, like, you can criticize, uh, why do we wanna go on Mars? Why do we wanna colonize Mars? Well, it, it's like, why do you want to go to the moon? The, the reality is when you do really difficult things-

Vincent Racaniello 59:10 Yeah.

Lex Fridman 59:10 ... and engineering-

Vincent Racaniello 59:11 Yeah.

Lex Fridman 59:11 ... things, like, all these inventions along the way-

Vincent Racaniello 59:13 Sure.

Lex Fridman 59:13 ... are created. It's kind of fascinating how basically just d- pick a, pick a thing that everyone can agree is kind of cool and is really hard and do that. And then you'll have, like, thousands of inventions that have nothing to do with the thing.

Vincent Racaniello 59:28 That's right.

Lex Fridman 59:28 Yeah.

Vincent Racaniello 59:28 And I think you should let curious scientists just follow what they're interested in-

Lex Fridman 59:34 Yeah.

Vincent Racaniello 59:34 ... to a certain extent. You can't, you know, in science, we say we have translational research where we say, "Okay, here's some money, go cure cancer, or diabetes, or heart disease, whatever," right? And that's fine, but that often doesn't work out very well. What works better is to say, "Here, you're, you have a good lab, you have a good track record. Here's some money-"

Lex Fridman 59:52 Yeah.

Vincent Racaniello 59:52 "... do something." And that's where PCR, CRISPR, recombinant DNA, all that stuff, which has made the field explode, that's all it came from, not from people saying, "I want to cure genetic diseases by gene editing," but by saying, "What are these repeated things in this bacterium doing?" And-

Lex Fridman 1:00:10 Yep. Can I ask you a big philosophical question? So, there's these deadly viruses that are not very transmissible.

Vincent Racaniello 1:00:19 Mm-hmm.

Lex Fridman 1:00:20 Ebola, rabies. And then there's these less deadly viruses that are very transmissible, um, like, uh, like COVID is, I guess kind of borderline. But, uh, why isn't there super transmissible, super deadly viruses?

Vincent Racaniello 1:00:38 I think if you compare SARS-1 and 2, you get, uh, somewhat of an answer, right? SARS-1 was more deadly, in fact-... over half of the time when people were infected, they ended up in the hospital 'cause they were that sick. And then th- the peak of virus shedding from them happened long after they went in the hospital. So, it's easy to contain, uh, the infection when you're in a hospital, right? There was not much pre-symptomatic or asymptomatic shedding with SARS-1.

Lex Fridman 1:01:11 And shedding means you, you're inf- you become infectious.

Vincent Racaniello 1:01:14 So, in, in a respiratory virus, you're, you inhale the droplets with the virus and they, they reproduce in your upper respiratory tract, what we call the nasopharynx, right?

Lex Fridman 1:01:22 Mm-hmm.

Vincent Racaniello 1:01:23 The nose and going back to that little cavity just above y- your mouth. So, the virus reproduces really well. And then as you talk and sneeze and cough, you expel droplets, and then those are inhaled by other people, and then they reproduce. And for SARS-2, we now know there's a lot of reproduction just before you feel anything, if at all.

Lex Fridman 1:01:42 Mm-hmm.

Vincent Racaniello 1:01:42 So, there's a lot of shedding and transmission before you get symptomatic.

Lex Fridman 1:01:47 So-

Vincent Racaniello 1:01:47 And then many people don't ever get symptomatic, right? So, they-

Lex Fridman 1:01:50 Okay.

Vincent Racaniello 1:01:50 ... spread really easily. So, that explains why some viruses can transmit a lot better than others. And if one h- happens to knock you out, then you're never gonna transmit 'cause you're in the hospital, like SARS-1.

Lex Fridman 1:02:03 But why can't you have both? Why can't you just wait a while before it knocks you out, but when it knocks you out, it really kills you?

Vincent Racaniello 1:02:10 That's, that is a philosophical question, right? Because we could talk about why we haven't observed it. I mean, one, one issue is that if you, uh, if you're killed too quickly by a highly lethal virus, you're not gonna transmit it very well, right? So, Ebola can kill you quite rapidly and most of the transmission occurs when people are being cared for at home or in hospitals. The doctors and nurses get the virus, but people walking around, you're not walking around when you have Ebola. You're too sick. You know? You have black bloody diarrhea, you're vomiting, you're, you're ble- bleeding from your skin and mucus membranes. You're not walking around, you're not going to parties. So, I think that's part of it, that if the, if the infection is too lethal, you're simply not a good transmitter. And I think transmission is probably one of the most powerful selection forces for viruses, because a virus always has to have, find a new host. If it doesn't, it's a startup that fails, right? If it doesn't find a new host, it's gone.

Lex Fridman 1:03:18 Uh-huh.

Vincent Racaniello 1:03:18 And so anything that makes the virus transmit better is gonna help it. And if killing you, being less lethal is part of that, that works too.

Lex Fridman 1:03:28 So, there's a strong selection pressure against being lethal?

Vincent Racaniello 1:03:32 I think there's a strong selection prof- uh, pressure against being lethal and being more transmissible. Those two seem to work in opposite ways. And now we don't have a lot of data to support this. This is kind of a, a thought experiment, but there is one experiment done in Australia many years ago. I don't know if you know this, but in the 1800s, the hunters in Australia imported a rabbit from Europe so they could hunt it because the native rabbit in Australia was too fast for them. They couldn't shoot them.

Lex Fridman 1:04:03 Hm.

Vincent Racaniello 1:04:03 So, they brought in this European rabbit, and they, they reproduced out of control. Within a couple of years, they were everywhere, millions of rabbits in all the watering holes, and now they had a problem. So, they decided to use a virus to get rid of these excess rabbits. And they used a, a virus, a pox virus called myxoma virus, which is a natural virus of a different kind of rabbit. But for these European rabbits, it was quite lethal and it spread by mosquitoes. So, they said, "Okay, let's, let's, uh, release this virus." And in the first year, 99.2% of the rabbits were killed.

Lex Fridman 1:04:39 Oh.

Vincent Racaniello 1:04:40 But that .8% that were left had some form of resistance. They were variants, you know? Every organism, not just viruses, makes mutants, and there were some variants of the rabbits that could survive infection. And then in subsequent years, the virus became less lethal, and then the mosquitoes had a better shot of transmitting it from one rabbit to another if the rabbit lived longer. That's the selection probably. And so in the end, the, the rabbits lived on, the, the virus was there, it evolved to be more transmissible and less, uh, lethal. So, that's the only-

Lex Fridman 1:05:12 Life is amazing.

Vincent Racaniello 1:05:13 ... that's the only data.

Lex Fridman 1:05:13 Life on Earth is amazing.

Vincent Racaniello 1:05:15 It is, it is. If you take the time to look at it and see what's happened, it is amazing.

Lex Fridman 1:05:20 It's also humbling that it just makes you realize humans are just a small part of the picture.

Vincent Racaniello 1:05:25 Of course.

Lex Fridman 1:05:26 Well-

Vincent Racaniello 1:05:26 And we're wrecking it, aren't we? (laughs)

Lex Fridman 1:05:28 Well, I mean, that's, that, it's, we're not really... I mean, viruses are wrecking it some ways. Part of this, we're not really wrecking anything. It's all part of it. (laughs)

Vincent Racaniello 1:05:38 But, you know, when... The ways that human exists encourages viruses to infect us, right? When we were hunter-gatherers living in bands of 100 people, very few viruses because it was hard f- for the virus to go from one band to another, and perhaps a hunter would, one of these humans would get an animal and bring a virus into camp and some people would die, but it would never spread-

Lex Fridman 1:06:01 Mm-hmm.

Vincent Racaniello 1:06:01 ... to another. And then when we started to congregate in cities, we figured out agriculture and so forth and how to harvest animals, then we could get bigger and bigger populations, and the viruses went crazy. And they went from animals to us. So, measles went from cows to humans when humans learned to domesticate cows and, and, uh, started gathering in big cities.

Lex Fridman 1:06:20 Yeah, but now that humans are able to communicate and travel globally, the viruses become more and more dangerous, transmissible. Uh, thereby, if you look at Earth as an organism, thereby pushing humans to be more innovative, create AlphaFold 2 and 3 and 4 and 5, create better systems, and eventually there's rockets that keep flying from Earth. And eventually, uh, the virus is becoming super dangerous and threatening all of human civilization, will force it to become a multi-planetary species-

Vincent Racaniello 1:06:52 (laughs)

Lex Fridman 1:06:52 ... and this organism starts expanding. So, I think it's a feature, not a bug. I, I don't know. Um-

Vincent Racaniello 1:06:57 Well, I think that we have...... our early, probably the, most of the, what we're studying viruses since 1900, right? Most of that time was because of diseases they caused. The first virus is discovered, yellow fever virus, smallpox, uh, poliovirus, influenza virus, those were all because people got sick and they said, "Oh, look, this is a virus that's associated with it." And so we got good at learning how to take care of these infections, making vaccines and so forth over the years. And it's only in the last 20 years that we recognize that there are more viruses out there that are far more interesting perhaps, but we've learned how to deal with the bad ones, for sure.

Lex Fridman 1:07:41 So we talked about what is a virus, we talked about some of the most dangerous and deadly viruses.

Vincent Racaniello 1:07:46 Mm-hmm.

Lex Fridman 1:07:47 Can we zoom in and talk about COVID-19 virus?

Vincent Racaniello 1:07:51 Sure.

Lex Fridman 1:07:51 I don't know what your preferred name is, but-

Vincent Racaniello 1:07:53 Well, there're two names.

Lex Fridman 1:07:54 ... maybe for this-

Vincent Racaniello 1:07:54 Right, the virus is SARS-CoV-2.

Lex Fridman 1:07:56 Yeah.

Vincent Racaniello 1:07:56 Which is hard, it's long, right? And then COVID-19 is the disease. So you could say the virus of COVID-19, that's fine. Yeah.

Lex Fridman 1:08:03 The virus of COVID-19. But for the purpose of this conversation, we'll, every once in a while just say COVID.

Vincent Racaniello 1:08:08 It's fine.

Lex Fridman 1:08:08 (laughs)

Vincent Racaniello 1:08:09 No problem. (laughs)

Lex Fridman 1:08:10 What, what is this virus from a... I don't know how many ways we can talk about it. I think from a basic structural, like the varian structure, biological structure perspective-

Vincent Racaniello 1:08:25 Mm-hmm.

Lex Fridman 1:08:25 ... what is it? What are its variants? Can you describe the basics-

Vincent Racaniello 1:08:30 Okay.

Lex Fridman 1:08:30 ... the important characteristics of the virus?

Vincent Racaniello 1:08:32 So, viruses are classified by humans just to make it easier to keep track of them.

Lex Fridman 1:08:37 Yeah.

Vincent Racaniello 1:08:37 Right? So this is a coronavirus, which is because when they were first discovered, I think the first ones were animal coronaviruses. They, they looked at them under the electron microscope and it looked like the solar corona. And that's all there is to it. And I have to say that early in the outbreak, the, the place with the highest seropositivity in the US for a while, 68%, was a working class neighborhood in New York City called Corona.

Lex Fridman 1:09:05 Mm-hmm.

Vincent Racaniello 1:09:06 Can you, can you beat that? Right?

Lex Fridman 1:09:08 That's crazy. Yeah.

Vincent Racaniello 1:09:09 So coronaviruses, they have membranes, right? We talked about membranes, they have spiked proteins in the membranes so they can attach to cells. And inside they have RNA. And they are the viruses with the longest RNA that we know of. No, none other comes close. For some reason, they're able to maintain 30,000. So SARS-CoV-2 RNA is 30,000 bases of RNA, and some of the other coronas are even longer, 40,000.

Lex Fridman 1:09:37 This is a f- uh, coronas are a family of viruses-

Vincent Racaniello 1:09:40 Yes.

Lex Fridman 1:09:40 ... that included the w- what, the, the one you mentioned before, version one? (laughs)

Vincent Racaniello 1:09:46 So SARS-CoV-1, yeah.

Lex Fridman 1:09:47 CoV-1, and I guess other ones as well.

Vincent Racaniello 1:09:49 So the first, we first learned of them in animals, a lot of animals, pigs and, um, cows and horses have coronaviruses. And then, uh, in the '60s we discovered a couple of human coronaviruses that just cause colds, very mild colds that you wouldn't even think twice about, right? And then suddenly in 20... 2003, there's this outbreak of severe respiratory disease in, in China. And, you know, they, i- it started in November and they didn't tell the world until February. And that was really bad because it was already spreading by the time they told people about it. But this went to many, 29 different countries, only 8,000 people were infected, and then it stopped. And that was the first time we saw an epidemic coronavirus, and it... What they did afterwards is they said, "Okay, i- it looks like it came from the meat markets." They have live meat markets in Guangzhou, in the south of China, where you can go and pick out an animal and the guy will slaughter it for you and give it to you. And then of course there's blood everywhere, and that's how they got infected. And they figured out that there's this animal called a palm civet that was the source of virus. The palm civets are shipped in from the countryside, and they... The palm civets somehow in the countryside got it from a bat. So they went looking in caves in the countryside and they found in one cave all the viruses that could make up SARS-1. And that was 2000 and, uh, I would say took about five, eight years a- after that outbreak. So that was the first hint that bats have coronaviruses that can infect people and cause problems, right? And after that, we should have been ready.

Lex Fridman 1:11:35 So didn't they already start developing vaccines-

Vincent Racaniello 1:11:38 Yes.

Lex Fridman 1:11:38 ... after then?

Vincent Racaniello 1:11:38 So some people started making vaccines, they tested them in, in mice, um, but they never got into people. And some people started working on antiviral drugs. Nothing ever came of them because, you know, industry, there's no, there's no disease, it's gone. Why should we make vaccines and drugs? And NIH in the US you submit a grant and they say, "Ah, it's too risky. There's no, none of this virus around." So people were really short-sighted. Because I always say we could have had antivirals for this. Absolutely.

Lex Fridman 1:12:12 Yeah.

Vincent Racaniello 1:12:12 For sure. No question. In fact, one, the one antiviral that's in phase three, it's called molnupiravir, it's the only one that you can take orally. It's a pill.

Lex Fridman 1:12:23 Mm-hmm.

Vincent Racaniello 1:12:23 It looks really good. That was developed five years ago, but never taken into humans. It could have been ready. So we dropped the ball, and then the next decade, 2012 MERS coronavirus comes, uh, comes up in the Arabian Peninsula. This comes from camels and infects people, but probably the camels got it from bats originally-

Lex Fridman 1:12:44 Mm-hmm.

Vincent Racaniello 1:12:44 ... some time ago. But that never transmits from person to person, very rarely.

Lex Fridman 1:12:50 Mm-hmm.

Vincent Racaniello 1:12:50 Every new little outbreak is a new infection from a camel. So that was 2012, and now here we are 2019, a new outbreak of respiratory disease in China, and this one really...... goes all over the world, w- where SARS-1 could not. And it's a coronavirus, you know, it's different enough from SARS-1 that it has very different properties, causes a lot-

Lex Fridman 1:13:13 But it still has a membrane, it still has a very long RNA in the middle-

Vincent Racaniello 1:13:17 Right.

Lex Fridman 1:13:17 ... and then it still has the spike proteins.

Vincent Racaniello 1:13:19 That's right.

Lex Fridman 1:13:20 What are the things that are... What are the little unique things that make it that much more effective?

Vincent Racaniello 1:13:27 That make it cause a pandemic-

Lex Fridman 1:13:29 Pandemic.

Vincent Racaniello 1:13:30 ... of millions of people, as opposed to SARS-1?

Lex Fridman 1:13:32 Yeah.

Vincent Racaniello 1:13:32 Well, the genome is 20% different from SARS-1, say, and in those bases there's some, there are things that, uh, make it different from SARS-1. It binds the same receptor, ACE2, on the cell surface, though, that's remarkable. Um, it has a lot of the same proteins. They look similar. Like, if you look at the structure of the spikes, they look similar, but there's enough amino acid differences to, to make the biolo- And what it is, we don't know, because, uh, how do you figure that out? You need to study animals, 'cause you can't infect people. And the, the animal models aren't great, you know, for, for, for sp-

Lex Fridman 1:14:10 So, the way you figure that out is you figure out how those differences, what functional... Like, how the difference in the amino acids lead to functional difference of the virus?

Vincent Racaniello 1:14:20 That's right.

Lex Fridman 1:14:21 Like, how it attaches-

Vincent Racaniello 1:14:22 That's right.

Lex Fridman 1:14:22 ... how it breaks the cell wall?

Vincent Racaniello 1:14:23 Exactly. Sure.

Lex Fridman 1:14:23 And how the hell do you figure that out? Like, I guess there's models of interaction.

Vincent Racaniello 1:14:29 So, you, you n- You need a... First, you need an animal of some kind to infect, right? You can use mice, people have used ferrets, guinea pigs, non-human primates, all of the above. The non-human primates are very expensive so not many people do that. Um, and then you can put the virus in their respiratory tract. But in fact, none of them get sick like people do. You know, many people with, with COVID get a mild disease, but 20% get a very severe, longer-lasting disease and they can die from it, right? No animal does that yet. So, we have no insight into what's controlling that. But if you just wanna look at the very first part of infection and the shedding and the transmission, you can do it in any one of several animal models. Ferrets are really good for transmission. They tend... They have nasal structures like humans and they, they... You could put them in cages next to each other and they'll transmit the virus really nicely, so you can study that.

Lex Fridman 1:15:26 Cool.

Vincent Racaniello 1:15:26 But the other thing that's important that we should mention is, how do you manipulate these viruses? So, these are RNA viruses. You can't manipulate RNA. We don't know how to do it. But with DNA, because of the recombinant DNA revolution that occurred in the '70s, we can change DNA any way we want.

Lex Fridman 1:15:51 Mm-hmm.

Vincent Racaniello 1:15:51 We could change a single base, we can cut out bases, we can put other things in really easily. And, if, if I may give it a personal aspect, when I went to MIT as a postdoc in 1979, David Baltimore said, "Here's what I want you to do. The, the moratorium on recombinant DNA experiments on viruses has just been lifted. I want you to make a DNA copy of polio and see if you put that in a cell whether it will start an infection."

Lex Fridman 1:16:22 Mm-hmm.

Vincent Racaniello 1:16:23 I said, "Okay." So, I made a DNA copy of poliovirus. It's only 7,500 bases, it's much smaller than corona. And I took that DNA and I put it in a piece of DNA from a bacteria called a plasmid.

Lex Fridman 1:16:37 Mm-hmm.

Vincent Racaniello 1:16:38 And you can grow plasmids in many, many bacteria, make lots of them and purify the DNA really easily. And I took that DNA and I, I sequenced it because we wanted... we didn't know the genome sequence of polio at the time. And that took me a year, by the way, 'cause the techniques we had were really archaic. And nowadays you could do it in 15 minutes, right? It's amazing. And I took the DNA, I put it into cells and out came polio. So, that's the start. Now, since then, everybody has taken that technique and used it for their virus. You can now do it with SARS-CoV-2. You make a DNA copy of any RNA virus, you can modify it, and you put it back into cells and you'll get vi- your modified virus out. So, that's an important part of understanding the properties of the virus, let's say, in an animal. By changing the virus, you're changing a DNA copy, you're making the virus then and putting it into the animal.

Lex Fridman 1:17:31 C- c- uh, can you clarify? So, e- even an RNA virus you can take and turn it into DNA?

Vincent Racaniello 1:17:38 Yes.

Lex Fridman 1:17:38 And then that allows you to modify it?

Vincent Racaniello 1:17:41 Yes. So that you take-

Lex Fridman 1:17:41 Uh, what's that, what's that, uh, mapping?

Vincent Racaniello 1:17:44 Enzyme.

Lex Fridman 1:17:44 Well, no, no, what's the process of going from RNA to DNA?

Vincent Racaniello 1:17:48 Reverse transcription.

Lex Fridman 1:17:49 That's reverse transcription?

Vincent Racaniello 1:17:50 Right.

Lex Fridman 1:17:51 Oh, so you actually go into the wh- the process of reverse transcription to do this?

Vincent Racaniello 1:17:54 Yes. So, the-

Lex Fridman 1:17:55 Oh.

Vincent Racaniello 1:17:55 ... remember, David Baltimore and Howard Schlag-

Lex Fridman 1:17:57 Yes.

Vincent Racaniello 1:17:57 ... had discovered this enzyme in the '70s. They got the Nobel Prize for that. And when I went to David's lab at MIT, he had the enzyme in the freezer. He said, "Here, take this and make a DNA copy of polio."

Lex Fridman 1:18:08 Yeah, I didn't ma- make the connection that you can use that kind of thing-

Vincent Racaniello 1:18:11 Yeah.

Lex Fridman 1:18:12 ... for an RNA virus.

Vincent Racaniello 1:18:14 And so that's-

Lex Fridman 1:18:14 And then modify it. That's fascinating.

Vincent Racaniello 1:18:16 See, any DNA virus already exists as DNA, so you can modify it.

Lex Fridman 1:18:19 Yeah.

Vincent Racaniello 1:18:19 That's no... But for RNA viruses it was difficult and... So then from that point on for influenza, every other RNA virus and coronaviruses people made DNA copies and that's what they used to modify and, and ask questions about what things are doing, right?

Lex Fridman 1:18:34 Right.

Vincent Racaniello 1:18:34 What's this gene doing? What if we take it out, what happens?

Lex Fridman 1:18:36 So, can you do the same thing with, uh, COVID, uh, is- is-

Vincent Racaniello 1:18:40 Of course.

Lex Fridman 1:18:40 ... take the RNA and then-

Vincent Racaniello 1:18:41 Of course. And in fact in January 2020, as soon as the genome sequence was released from China, the labs all over were synthesizing this 30,000 base DNA and-

Lex Fridman 1:18:52 Mm-hmm.

Vincent Racaniello 1:18:52 ... getting virus.

Lex Fridman 1:18:54 What, what can you figure out without infecting anything, just, uh, turning into, with reverse transcription, turning into DNA, modifying stuff and then putting it into a cell? What can you figure out?... from that?

Vincent Racaniello 1:19:08 Oh, well you could, uh, say you can cut out a gene. You see some genes in the sequence, "I don't know what these genes do."

Lex Fridman 1:19:14 I see.

Vincent Racaniello 1:19:14 "Let's cut them out." And then you could cut them out of the DNA. You put the DNA in cells and maybe you get virus out, and you go, "Oh, stop... Clearly that gene's not needed-"

Lex Fridman 1:19:25 Mm-hmm.

Vincent Racaniello 1:19:25 "... for the virus to reproduce, at least in cells." Right? Or, maybe you take the gene out and the, you never get any virus. So it's lethal to-

Lex Fridman 1:19:31 Is there nice systematic ways of doing this? Do people kind of-

Vincent Racaniello 1:19:34 Yes.

Lex Fridman 1:19:34 ... automate it?

Vincent Racaniello 1:19:35 Absolutely. And we, I mean, the problem with SARS, the COVID virus, is it's 30,000 bases.

Lex Fridman 1:19:42 It's long.

Vincent Racaniello 1:19:42 There's a lot of stuff there.

Lex Fridman 1:19:44 Yeah.

Vincent Racaniello 1:19:44 And what makes it more difficult is that you have to w- It's been classified as a BSL-3 agent, Biosafety Level 3. And so not everyone has a lab that's capable of doing that, so it limits the number of people who can do experiments. You know, some, we ha- We're, we're lucky to have a few in New York City, but not every place has them. So, you cannot work with the virus just out on the bench like we do with many other viruses. You have to wear a suit, and then you have to have special procedures and containment and so forth. So it makes it difficult to do basic experiments on the virus.

Lex Fridman 1:20:20 But, when it's a pandemic, there's a lot of money, there's a lot of incentive to work on it harder. (laughs) So...

Vincent Racaniello 1:20:25 Right, and also you don't need to work on the virus. You can then take bits of it and work. You could take, say, just the spike, right, and say, "Can we make a vaccine with just the spike?"

Lex Fridman 1:20:33 Right.

Vincent Racaniello 1:20:33 'Cause that doesn't require BSL-3. So, yes.

Lex Fridman 1:20:36 So f- uh, like building a vaccine requires you to figure out how, or antiviral drugs, how to attack various structural parts of the virus and the functional parts of the virus.

Vincent Racaniello 1:20:47 Right. You have to decide on a target.

Lex Fridman 1:20:50 Yeah.

Vincent Racaniello 1:20:50 Like, "I'm gonna make an antiviral. What am I gonna target in the virus?" And there are a few things that make more sense than others. Usually, we like to target enzymes. I don't know if you remember any, your biochemistry, but, you know, enzymes are catalytic. You don't need a lot of them to do a lot of things, so they're typically in low concentrations in, in a virus-infected cell, so it's easier to inhibit them with a drug. And then this coronas have a couple of enzymes that we can target. So, it's e- You've to figure that out ahead of time. You decide what to go after and then you can look for drugs that inhibit what, what you're interested in. It's not that hard to do.

Lex Fridman 1:21:31 It, there's just something beautiful about biology, about the mechanisms of biology. And I kind of regret, um, falling in love with computer science so much that I, um, left that biology textbook on the shelf-

Vincent Racaniello 1:21:46 Hmm.

Lex Fridman 1:21:46 ... and left it behind, but, uh, hopefully we'll return to it now, 'cause I think one