Betül Kaçar: Origin of Life, Ancient DNA, Panspermia, and Aliens | Lex Fridman Podcast #350

You can study chemistry, you can study physics, you can study geology anywhere in the universe, but this is the only place you can study biology. This is the only place to be a biologist.

Earth.

That's it, yeah. So definitely something very fundamental happened here, and you cannot take biology out of the equation. If you want to understand how that vast chemistry space, how that general sequence space got narrowed down to what is available or what is used by life, you need to understand the rules of selection. And that's when evolution and biology comes into play.

The following is a conversation with Batu Kochar, an astrobiologist at University of Wisconsin, studying the essential biological attributes of life. This is the Lex Friedman podcast to support it. Please check out our sponsors in the description. And now, dear friends, here's Batul Kachar. What is the phylogenetic tree or the evolutionary tree of life? And what can we learn by running it back and studying ancient gene sequences as you have?

I think phylogenetic trees could be one of the most romantic and beautiful notions that can come out of biology. It shows us a way to depict the connectedness of life and all living beings with one another. It itself is an ever evolving notion. Biologists like visualizations, they like these graphics, these diagrams, and tree of life is one of them.

So the tree starts at a common ancestor.

It's actually the other way around. It starts from at the end, it starts from the branches, it starts from the tip of the branch, actually, and then further depending on what you collected to build the tree. So depending on the branches, depending on what's on the tip of the branch, and I will explain what I mean, the root will be determined by what is really sitting on the tip of the branch of the tree.

So we could study the leaves of the tree by looking at what we have today and then start to reverse engineer, start to move back in time to try to understand what the rest of the tree, what the roots of the tree look like.

Exactly. So the tree itself, by just taking a few steps back and looking at the entire tree itself, can give you an idea about the connectedness, the relatedness of the organisms or whatever. Again, you use to create your tree. There are different ways, but in this case, I'm imagining entire diversity of life today is sitting on the tips of the branches of this tree. And we look at, biologists look at the tree itself. We like to think of it as the topology of the tree to understand when certain organisms or their ancestry may have merged over time. Depending on the tools you use, you might use this tree to then reconstruct the ancestors as well.

And so what are the different ways to do the reconstruction? So you can do that at the gene level, or you could do it at the higher, complex biology level. Right. So in which way have you approached this fascinating problem?

We approached it in every way we can. So it's the gene could be protein, the product of the gene, or species, or could be even groups of species. It totally depends on what you want to do with your tree. If you want to understand certain past events, whether an organism exchanged a certain DNA with another one along the course of evolution, you can build your tree accordingly. If you rather use the tree to reconstruct or resurrect ancient DNA, which is what we do, then in our case, for instance, we do both gene, protein and species because we want to compare the tree that we create using these different information.

Okay, well, let me ask you the ridiculous question then. So how realistic is Jurassic Park? Can we study the genes of ancient organisms and can we bring those ancient organisms back? So the reason I asked that kind of ridiculous sounding question is maybe gives us context of what we can and can't do by looking back in time.

Yeah. So dinosaurs or all these mammals, at least for us, is the exciting thing already happened by the time we hit to the larger organisms or to eukaryotes.

Oh, to you, the fun stuff is before we got to the mammal.

The fun stuff is what people think is boring. I think the phase that. Well, at least two different times in the geologic history. One is the first life, past origin of life. How did first life look like? And the second is why do we think that over certain periods of geologic time no significant innovation happened to the degree of leaving no record behind.

So what do we not have a record of? Which part? So you say the fun stuff to you is after the origin of life, which we'll talk about after the origin of life, there's single cell organisms, the whole thing with the photosynthesis, the whole thing with the eukaryotes and multicell organisms. And what else is the fun stuff? The whole oxygen thing, which mixes in with the origin of life. There's a bunch of different inventions. All they have to do with this primitive kind of life looking organisms that we don't have a good record of.

So I will tell you the more interesting things for us. One is the origin of life, or what happened right following the emergence of life. How did the first cells look like? And then pretty much anything that we think shaped the environments and was shaped by the environment in a way that impacted the entire planet that enabled you and I to have this conversation. We have very little understanding of the biological innovations that took place in the past of this planet. We work with a very limited set of, I don't want to even say data because they are fossil records. So let's say imprints, either that comes from the rock and the rock record itself or what I just described these trees that we create and whatever we can infer about the past. So we have two distinct ways that comes from geology and biology and they each have their limitations.

Okay, so, right, so there's an interplay. The geology gives you that little bit of data and then the biology gives you that little bit of kind of constraints in the materials you get to work with to infer how does this result in the kind of data that we're seeing. And now we can have this through the fog. We can see, we can look back hundreds of millions of years, a couple of billion years, and try to infer even further.

And I like that you said fog. It is pretty foggy. And it gets foggier and foggier the further you try to see into the past. Biology is you basically study with, study the survivors, broadly speaking, and you're trying to pitch the sort of put together their history based on whatever you can recover today. What makes biology fascinating also erased its own history in a way. Right. So you work with this 4 billion year product, that's genome, that's the DNA. It's great. It's a very dynamic, ever evolving technology, chemical thing. And so you will get some information, but you're not going to get much unless you know where to look because it is responding to the environment.

Yeah. So what we have, that's fascinating, what we have is the survivors, the successful, the successful organisms, even the primitive ones, even the bacteria we have today.

So bacteria is not primitive and we.

Sorry, sorry to offend the bacteria.

We should be very grateful to bacteria. First of all, they are our great, great ancestors. I like this quote by Douglas Adams. Humans don't like their ancestors. They rarely invite them over for dinner. Right, but bacteria is in your dinner, bacteria is in your gut. Bacteria is helping you.

So we do invite them for dinner.

We might. Well, they get themselves in midas in a way. Yes. And they're definitely older and definitely very sophisticated, very resilient than anything else. As someone working as a bacteriologist, I feel like I need to defend them in this case because they don't get much shout out when we think about life.

So you do study bacteria. So which organisms Gives you hints that are alive today that give you hints about what ancient organisms were like. Is it bacteria, is it viruses? What do you study in the lab?

We study variety of different bacteria depending on the questions that we ask. We engineer bacteria. So ideally we want to work with bacteria that we can engineer. Seldom we developed the tools to engineer them. And it depends on the question that we are interested in. If, if we are interested in connecting the biology and geology to understand the early life and fundamental innovations across billions of years. There are really good candidates like cyanobacteria. So we use cyanobacteria very frequently in the lab. We can engineer its genome, we can perturb its function by poking its own DNA with the foreign DNA that we engineer in the lab. We work with E. Coli. It's the most simple in terms of model systems goes organism that one can study. Well established, sort of a pet lab pet that we use it a lot for cloning and for understanding basic functions of the cell. Given that it's really well studied and.

What you do with that. E. Coli you said that you injected with foreign DNA.

We inject pretty much all the bacteria that we work with with foreign DNA. We also work with diazotrophs. These are Azodobacteria. They're one of the prime nitrogen fixers, nitrogen fixing bacteria.

Can you explain what that is? Nitrogen fixing is the source of its energy.

So nitrogen is a triple bond gas that's pretty abundant in the atmosphere. But nitrogen itself cannot be directly utilized by cells. Given it is triple bond, it needs to be converted to ammonia that is then used for the downstream cellular functions.

And that's what counts as nitrogen fixing.

Yes. So nitrogen is fixed before our cells can make use of it.

No offense to nitrogen either.

Well, it's actually a very important element. It's one of the most abundant elements on our planet that is used by biology. It's in ATP, it's in chlorophyll, that relies on nitrogen. So it's a very important enzyme for a lot of cell functions.

And there's just one mechanism that evolution invented to convert it, fix it.

So far we know there's only one nitrogen fixation pathway as opposed to, say, carbon. You can find up to seven or eight different carbon based microbes invented to fix carbon. That's not the case for nitrogen. It's a singularity across geologic time. We think it evolved around 2.7, maybe roughly 3, probably less than 3 billion years ago. And that's the only way that nature invented to fix the nitrogen in the Atmosphere for the subsequent use.

Would we still have life as we know it today if we didn't invent that nitrogen fixing step?

I cannot think of it, no. It's essential to life as we know. You and I are having this conversation because life found a way to fix nitrogen.

Is that one of the tougher ones? If you put it sort of oxygen, nitrogen, carbon. What are in terms of being able to work with these elements? What is the hardest thing? What is the most essential for life? Just to give context.

Well, we, we think of this as the cocktail. You may hear what's in the cocktail. It's the Schnapps. Right. Carbon, hydrogen, oxygen, nitrogen, sulfur. So there are five elements that life relies on. We don't quite know whether that's the only out of many options that life necessarily needs to operate on, but that's just how it happened on our own planet. And there are many abiotic ways to fix nitrogen. And like lightning. Right. Lightning can accumulate ammonia. Humans found A way about 100 years ago, I think, around World War I, the Haber Bosch process, that we can abiotically convert nitrogen into ammonia. Actually, 50% of the nitrogen in our bodies comes from the human conversion of nitrogen to ammonia. The fertilizer that we use, urea, comes from that process. It's in our food. So we found a way to fix our own nitrogen for ourselves.

Yeah, but that's way after the original invention of hydrogen.

Oh, absolutely, absolutely.

And without that, we wouldn't have. We wouldn't have all the steps of evolution along the way.

Oh, absolutely. It's very. We tried to replicate in the most simplest way what nature has come up with. Right. We do this by taking nitrogen, using a lot of pressure and then generating ammonia. Life does this in a more sophisticated way, relying on one single enzyme called nitrogenase. It's the nitrogen that is used together with eight electron donor and ATP together with a lot of hydrogen. Life pushes this metabolism down to create fixed nitrogen. It's quite remarkable.

So the lab pet E. Coli, inject them with DNA. So E. Coli does nitrogen fixing in part, or is that a different one?

So some biological engineers engineered E. Coli to fix nitrogen, I believe. Not us. We use the nature's nitrogen fixing bug and engineer it with the nitrogen fixing metabolism that we resurrected using our computational and phylogenetic tools.

How complicated are these little organisms? What are we talking about?

It depends on how we define complication.

Okay, so I could tell that you appreciate and respect the full complexity of even the most seemingly primitive organisms, because none of them are primitive. Okay. That said, what are we talking about? What kind of machineries do they have that you're working with when you're injecting them with DNA?

So I will start with one of the most fascinating machineries that we target, which is the translation machinery. It is a very unique subsystem of cellular life in comparison to, I would say, metabolism. And we used to, you know, when we are thinking about cellular life, we think of cell as the basic unit or the building block. But from a key perspective that's not the case, that one may argue that everything that happens inside the cell serves the translation and the translation machinery. There is a nice paper that called this. That entire cell is hopelessly addicted to this main informatic computing biological chemical system that is operating at the heart of.

The cell, which is the translation.

It is the translation.

Translation from what to what? So RNA to enzymes.

It converts a linear sequence of MRNA into a folded, later folded protein. That's, that's when the, that's the core processing center for information for life. It has multiple steps. It initiates, it elongates, it terminates and it recycles. It operates discrete bits of information. Itself is like a chemical decoding device. And that is incredibly unique for translation that I don't think you will find anywhere else in the cell that does this.

So even though it's called translation, it's really like a factory that reads the schematic and builds a three dimensional object. It's like a printer.

I would divide it into actually even four more additional steps or disciplines than what would it take to study it, by the way you described it. It's a chemical system. It's the compounds that make it up are chemicals. It's physical, it tracks the energy to make its job, to do its job. It's informatic. What is processed are the bits. It's computational. The discrete states that the system is placed when the information is being processed. That itself is computational and it's biological. There's variability and inheritance that come from imperfect replication even and imperfect computation.

And that's so good. So from the biology comes the like when you mess up the bugs are the features. That's the biology. Informatics is obvious in the rna. That's a set of information there. The different steps along the way is actually kind of what the computer does with bits. It's their own computation. Physical, there's a, I guess almost like a mechanical process to the whole thing that requires energy. And actually it's manipulating actual physical objects and chemicals because you have to ultimately it's all chemistry.

Yeah. And it tracks this information. So it is almost a mini computer device inside ourselves. And that's the oldest computational device of life. It's likely the key operation system that had to evolve for life to emerge.

It's more interesting or it's more complicated in interesting ways than the computers we have today. I mean, everything you said, which is really, really nice. I mean, I guess our computers have the informatic and they have the computational, but they don't have the chemical, the physical or the biology.

Exactly. And the computers don't link information to function. Right. They are not tightly coupled. Nowhere close to what translation or the way translation does it. So that's the number one, I think difference between the two. And yes, it's informatic. And we can discuss this further too.

100%. Let's please discuss this further. Which part are we discussing? For each one of those are fascinating worlds. Each of the five.

Yeah. So. Well, we can start with the more, I guess the ones that are more established, which is the chemical aspect of the translation machinery. The specific compounds make up the assembly of rna. Chemists showed this in many different ways. We can rip apart the entire machinery. We know that at the core of it, there's an RNA that operates not only as an information system itself or information itself, but also as an enzyme. And origin of life chemists make these molecules easily. Now we know we can manipulate rna. We can make. Even with single pot chemistries, we can create compounds.

What's a single pot chemistry?

I would say when you add all the recipes that you know that will lead you to the final product.

This is what original life chemists do is they, they come up with this pod, they throw a bunch of chemicals in and they try to try to. They're basically chefs of a certain kind.

I'm not sure if that's what they call it, but that's how I think of it because it is all combined in a test tube and you know the outcome and, and it's, it's very mathematical. Once you know the right environment and the right chemistry that needs to get into this container or this path, you know what the outcome is. There is no luck there anymore. It's a pretty rigid, established input output system and it's all chemistry.

So you actually wear a lot of hats. Is one of them origin of Life Chemist?

My PhD is in chemistry, but I don't do origin of life chemistry.

But you're interested in origin of life.

Yes, absolutely.

So some of your, some of your best friends are origin of life chemists.

Just make sure that you have good chemist friends. If you're interested in origin of Life. Yeah, that's 100% requirement, should be mandatory.

Okay, so chemistry. So what else about this machinery that we need to know chemically?

Well, chemically, I think that's it. You have enzymes, you have proteins. Enzymes are doing their thing. They know how to chew energy using ATP or gtp. They know what to do in their own way. They do their enzymatic thing. So it's not just the ribosome that is at the heart of the transition, but there are a lot of different proteins. You're looking about 100 different components that compose this machinery.

Well, let me ask kind of maybe it's a ridiculous question, but did the chemistry make this machine or did the machine use chemistry to achieve a purpose? So, like, I guess there's a lot of different chemical possibilities on Earth. Is is this translation machinery just like picking and choosing different chemical reactions that it can use to achieve a purpose, or did the chemistry basically, like, there's like a momentum, like a constraint to the thing that can only build a certain kind of machinery, basically. Is chemistry fundamental or is it just emergent? Like, how important is chemistry to this whole process?

You cannot have life without chemistry. You cannot have any cellular process without chemistry. What makes life interesting is that even if the chemistry is imperfect, even if there are accidents along the way, if something binds to another chemical in a way it shouldn't, there is resilience within the system that it can maybe not necessarily repair itself, but it moves on. However, in perfect, mistakes can be handled.

That's where the biology comes in.

That's where the biology comes in. But in terms of chemistry, you absolutely cannot have a translation machinery without chemistry. And so you're. As I said, there are four main steps. These are the core steps that are conserved in all translation machinery. And I should say, all life has this machine. Right. Every cell on Earth. On Earth, yeah. Yes.

When you think of this machine, do you think very specifically about the kind of machinery that we're talking about, or do you think more philosophically, a machine that converts information into function?

I cannot separate the two. I think what makes this machinery fascinating is that those five components that I listed are they coexist. So, for instance, if we. Let's just talking about the chemistry part, we know the certain rate constant all these proteins that operate in this machinery needs to harbor in order to get the mechanism going. If you are bringing the information to the translation machinery and you are the initiator of this computation system, you need to have. You can only afford a certain range of mistakes. If you're too fast, then the next message cannot be delivered fast. If you're too slow, you may stall the process. So there is definitely a chemistry constant going on within the machinery. Again, it's not perfect, far from it. But they all have their own margin of error that they can tolerate versus they cannot. Otherwise they collapse, the system collapses.

So it's like jazz ensemble, the notes of the chemistry. But you can be.

I love that you said jazz. It's definitely true. It's a party and it's like everybody's invited and they need to operate together. Right. What's really cool about it, I think there are many things that are very interesting about this thing, but if you remove it from the cell and put it in a cell free environment, it works just fine. Right. So you can get cell free translation systems. Put this translation in a test tube and it is doing its thing. It doesn't need the rest of the cell to translate information. Of course you need to feed the information, at least so far. Because we are far from evolving a translation. Maybe not so far evolving a translation in the lab or machinery that can process information as it generates it. We have not done that yet.

It's a pretty complicated machinery. So it's hard for those origin of life chemists to find a pot that.

Generates because it's far more than chemistry. You need biology, obviously, you need biochemistry. You need to think as I think network systems folk. You need to think about computation, you need to think about information. And that is not happening yet. Except we are trying to bring this perspective. But the more you understand how information systems work, once you see it, you cannot unsee it. It's one of those things.

But you can still rip it out. And the chemistry happens.

Yes, and chemistry can happen. Even if you strip some of the parts out. You can get very minimal level of information processing. That does not look anything like the translation that cells relies on. But chemists showed from linear, you can generate information that arrives to a processing center in the form of a linear polymer. The informatic part of this system that I think sets it apart from computation and from metabolism comes in if you think about the information itself. Right. So we have four nucleotide letters that compose DNA and they are processed in the translation in triplets. So you have in triplet codon fragments. So you have four times four times four. So you have 64 possible states that can be encoded by four letters in three positions. All right, so.

So amazing. Yeah, it's so amazing.

There is only one code that says start. There's only one and then there's two, if not three, that says stop. So that's what you work with. But you can have 64 possible states. But life only uses 20amino acids. So life uses 64 possible states minus 4 of the starts and stops to code for 20amino acids in different combinations. That is really amazing. If you think about there are 500 different amino acids life can choose. Right. It's narrowed it down to 20. We don't know why A lot of people think about this genetic code is quite fascinating. I mean, it didn't do it for 4 billion years. I don't know. We may wait for another 4 billion years.

But you didn't have those amino acids in the very beginning, right?

You don't know. So we would be fooling ourselves if we said we know exactly how many amino acids existed early on. But there's no reason to think that it wasn't the same or similar. Yeah, we don't have a good reason. But because roughly 20 out of 60 states are used, you're using 1/3 of your possible states in your information system. So this may seem like a waste, but informatically it's important because it is abundant and it is redundant. Right? So this code degeneracy, you see this in that's implemented by this translation machinery inside the cell. So it means you can make errors, right? You can make errors, but the message will still get through. You can speak missing some letters to the information can miss some parts, but the message will still get through. So that's two thirds of the not used states gives you that robustness and resilience within the system.

So at the informatic level, there's room for error. There's probably room for error, probably in all five categories that we're talking about. There's probably room for error in the computation. There's probably room for error.

Yes, exactly. Everywhere there's room for error because informatic capacity is made possible together with the other components. And not only that, but also the product yields a function null, in this case enzyme or protein. Right. So that's really amazing.

For me it is. I mean, in my head, just so you know, because I'm a computer science AI person, The parallels between even like language models that encode language, or now they're able to encode basically any kind of thing, including images and actions, all in this kind of way. The parallel in terms of informatic and computation is just incredible.

Actually, I have a image maybe I can send you.

Can we pull it up now?

If you just do genetic codon charts, we can pull that up. Yeah. It's a very standard table. So I can explain why this is so amazing. So you're looking at. This is life's Alphabet, right? And so I also want to make a very quick link now to your first question, the tree of life. When we try to understand ancient languages or the cultures or the cultures that use these extinct languages, we start with the modern languages, right? So we look at Indo European languages and try to understand certain words and make trees to understand. This is what Slavic word is for snow, Something like snig.

Now we jumped to languages that humans spoke throughout humans.

Exactly. So we make trees to understand what is the original ancestor. What did they use to say? Snow. And if you have a lot of cultures who use the word snow, you can imagine that it was snowy. That's why they needed that word. It's the same thing for biology, right? If we understand some function about that enzyme, we can understand the environment that they lived in. It's similar in that sense. So now you're looking at the Alphabet of life. In this case, it's not 20 or 25 letters. It's. You have four letters. So what is really interesting, that stands out to me when I look at this, on the outer shell, you're looking at the 20amino acids that compose life, right? The one, the methionine that you see, that's the start. So the start is always the same to me. That is fascinating that all life starts with the same starts. There's no other start code. So you sent the ag. You know, aug to the cell that when that information arrives, the translation knows. All right, I gotta start function is coming the following. This is a chain of information until the stop code arrives, which are highlighted in black squares.

So for people just listening, we're looking at a standard RNA colored table organized in a wheel. There's an outer shell and there's an inner shell, all using the four letters that we're talking about. With that, we can. Can compose all of the amino acids. And there's a start and there's a stop. And presumably you put together with these letters, you walk around the wheel to put together the words, the sentences that.

Make words, the sentences. And again, you get one start. There are three different ways to stop this. One way to start it. And for each letter you have multiple options. So say you have a code, A A. The second code can be another A. And even if you mess that up, you still can't rescue yourself. So you can get, for instance, I'm looking at the lysine K. You get an A and you get an A and then you get an A. That gives you the lysine. Right. But if you get an A and if you get an A and I get a G, you still get the lysine. So there are different combinations. So even if there's an error, we don't know if these are selected because they were erroneous and somehow they got locked down. We don't know if there is a mechanism behind this or. We certainly don't know this definitively, but this is informatic part of this. And notice that the colors in some tables too, the colors will be coded in a way that the type of the nucleotides can be similar chemically. But the point is that you will still end up with the same amino acids or something similar to it, even if you mess up the code.

Do we understand the mechanism? How natural selection interplays with this resilience to error? Which errors result in the same output, like the same function and which don't, which actually results in a dysfunction which are.

We understand to some degree how translation and the rest of the cell work together, how an error at the translation level, this is the really core level, can impact entire cell. But we understand very little about the evolutionary mechanisms behind the selection of the system. It's thought to be as one of the hardest problems in biology and it is still the dark side of biology, even though it is so essential. So this is. Yeah, you're looking at the language of life, so to speak, and how it can found ways rather to tolerate its own mistakes.

So the entire phylogenetic tree can be like deconstructed with this wheel of language.

Because all the final Letters, that's the 20amino acids, that's our Alphabet, they are all brought together with these bits of information. Right. So when you look at the genes, you're looking at those four letters. When you look at the proteins, you're looking at the 20amino acids, which may be a little easier way to track the information when we create the tree.

So using this language, who can describe all life that's lived on Earth?

I wish. We are not that good at it yet. Right.

So in theory, this is one way to look at life on Earth.

If you're a biologist and you want to understand how life evolved from a molecular perspective, this would be the way to do it. And this is what nature narrowed its code down to. So when we think of nitrogen, when we Think of carbon. When we think of sulfur, it's all in this. That all these nucleotides are built based on those elements.

And this is fundamentally the informatic perspective.

Exactly. That's the informatic perspective. And it's important to emphasize that this is not engineered by humans. This evolved by itself.

Right. Humans didn't invent this just because we're just describing. We're trying to describe the language of life.

It appears to be a highly optimized chemical and information code. It may indicate that a great deal of chemical evolution and this may indicate that a lot of selection pressure and Darwinian evolution happened prior to the rise of last universal common ancestor. Because this is almost a bridge that connects the earliest cells to the last universal common ancestor.

Okay, can you describe what the heck you just said? So this mechanism evolved before the what common ancestry. So there's the last universal common ancestor.

Yes. When we talk about the tree, when we think about the root, if you ideally included all the living information or all the available information that comes from living organisms on your tree, then it. On the root of your tree lies the last universal common ancestor. Luca. Right.

Why last? Last universal. Because the earlier universe, it also had trees, but they all died off.

We call it the last because it is sort of the first one that we can track, because we cannot. We don't know what we cannot track. Right.

There was one organism that started the whole thing.

It's more like a. I would think of it as more like a population, a group of organisms than it's.

Hang on a second. I tweeted this, so I want to know the accuracy of my tweet. All right. Sometimes early in the morning, I tweet very pothead like things. I said that we all evolved from one common ancestor that was a single cell organism 3.5 billion years ago. Something like this. How. How true is that tweet? Do I need to delete it? No, there's not actually.

Correct.

I mean, I think, of course there's a lot to say, which is like we don't know exactly, but to what degrees. The single organism aspect. Is that true versus multiple organisms.

Do you want me to be brutally honest?

Yes, please. There's still time. This is how we get caveats and tweets.

So first of all, it's not. 3.5 is still very conservative estimate. That's what.

In which direction?

I would say it's 3.8 is probably safer to say at this point.

A bunch of people said it probably way before.

If you put an approximately I'll take that.

I didn't. I just love the idea that I was once, first of all, as a single organism, I was once a cell.

Well, you're still is. You're a group of cells.

No, but I started from a single cell. Me, Lex.

You mean like you versus luca? Are you relating to LUCA right now.

Or related to my own development? My own development. I started from a single cell. It's like it built up and stuff. Okay, that and then so that's for single biological organism. And then from an evolutionary perspective, the luca, like I start, like my ancestor, is a single cell. And then here I am, sitting half asleep, tweeting like, I started from a single cell, evolved a ton of murder along the way to this, like, brutal search for adaptation through the 3.5.8 billion years.

So you defy the code of Douglas Adams, you are proud of your ancestors, and you invite them over to dinner, and you invite them over to your tw. Yeah.

And it's amazing that this intelligence, to the degree you can call it intelligence, emerged to be able to tweet whatever the heck I want.

That means it's almost intelligence at the chemical level. And this is also probably one of the first chemically intelligent system that evolved by itself in nature.

Yeah. So you see, translation is fundamentally like intelligent mechanism in its own way.

And again, if we manage to figure out how to drive life's evolution, if it can evolve a sophisticated sort of informatic processing system like this, you may ask yourself, what might chemical systems be capable of independently doing under different circumstances?

Yeah, so like locally, they're intelligent. Locally, they don't need the rest of the shebang. Like, they don't need the big picture.

So that, that's. That's a great segue into what makes us biological. Right. The. The heart of the cellular activities are translation. You kill translation, you kill the cell.

Yes, You.

Not only the transition itself, you kill the component that initiates it. You kill the cell, you kill. You remove the component that elongates it, you kill the cell. So there are many different ways to disrupt this machinery. All the parts are important. Now, it can vary across different organisms. We see variation between bacteria versus eukaryotes versus archaea. Right. So it is not the same exact steps, but variation. It can get more crowded as we get closer to eukaryotes, for instance, but you are still computing about 20amino acids per second. Right. This is what you're generating every second.

That single machinery is doing. 20 a second.

21 for bacteria, I believe, 8 for eukaryotes, or 9, 21 a second.

I mean that's super inefficient or super efficient, depending on how you think about it.

I think it's great.

Yeah, but it's way slower than a computer could generate through simulation.

I think if you can show me a computer that does this, we are done here.

Well, this is the big. This includes the five things, not just. But I could show you a computer that's doing the informatic, right?

Yes, you can show me that, but you cannot show me the one that has all.

For now. For now I will ask you about probably alphafold, Right.

I think the more we learn about. And this is why early life and origin is also very fascinating and applicable to many different disciplines. There is no way you see this the way we just described it. Unless you think about early life and early geochemistry and earliest emergent systems. But going back to the biological component, all of these attributes that we think about life or that we associate with biology stems from translation and as well as metabolism. But I see metabolism as a way to keep translation going and translation keeps metabolism going. But translation is arguably a bit more sophisticated process for the reasons that I just described.

So metabolism is a source of energy for this translation process.

It's a way to process materials and it is inherently dynamic and it is flexible, but it is not focused on repetition as translation does. So that's the main difference. Translation is kind of in a way just it repeats. Right. So you have the metabolism that can synthesize materials, creates or benefits from available energy and again it's a dynamic system and then you have computation that that is inherently repetitive, needs to carry out repetitive processes and it does the tasks and it implements an algorithm, but it is not dynamic. So you see both of those attributes in translation combined. It is repetitive and it is dynamic and it also processes this information. So they are fundamentally different. I don't know if you can get life if you don't find a way to process the information around you in.

A repetitive dynamic way.

Yeah. And somehow that's what got selected. Maybe not selected. I don't know if it was accidental, but that's what it seems to be conserved for 4 billion years. That that's what life established.

What's the connection between translation and the self replication which seems to be a. Another weird thing that life just started doing, wanting to just replicate it.

I think when we truly understand the answer to that question, we may have just made ourselves life. Right. I don't think we know quite how translation machinery as a whole fits into equation because so we try to understand ribosomes, rna, how the linear information is processed, or the genetic code, why this codons, not others, why 20, not more, not less. And we are sort of moving towards transition. That's what we are working on anyway, to finally look at the patterns in which this system operates itself. And if you understand that you're really unlocking a very emergent behavior.

One of the things you didn't mention is physical. Is there something to mention about that component that's interesting?

There's actually a paper published in 2013, I want to say the first author, Zernoff. So they surveyed computational engineered systems level computation, energy consumption, and they tried to understand whether the universe is using its own or life is using its full capacity of energy consumption and whether if different planets in the universe had life, would the capacity would increase or decrease. Does life operate at its energy maximum? And they think that it does, that it actually operates at an efficiency that is far more above and beyond any computational system.

How is that possible to determine at all that?

You tell me. That's why I dropped the citation. I found the citation. It's quite an interesting paper. It's a bit, you know, it's a, it's a. Obviously you can only calculate and infer these things, but it's a good question to ask.

Is the life that we see here on Earth and life elsewhere in the universe, is it using the energy most efficiently? Yeah, yeah, it seems to be very efficient. Again, if we compare to computers, it seems to be incredibly efficient at using.

I think they look at the, like the theoretical optimum for electronic devices and then try to understand where life falls on, on this. And life is certainly more efficient.

And that's ultimately the physical side. How well are you using for this entire mechanism the energy available to you? And so given, given all the resilience to errors and all that kind of stuff, it seems that it's close to its maximum.

Yep.

And this, this paper aside, it does seem that life, obviously that's the constraint we have on Earth. Right. Is the amount of energy.

Yeah.

So that's one way to define life. Well, the input is energy and the output is what? I don't know, self replicating. Wait, how. Okay, let's go there. How do you, how do you personally define life? Do you have a favorite definition that you try to sneak up on? Is it possible to define life on Earth?

I don't know. It depends on what you are defining it for. If you're defining it for finding different life forms, then it probably needs to have some quantification in it so that you can use it in whatever the mission that you're operating to.

You mean, like, it's not binary. It's. This is like a 7 out of 10 on lifelike.

I don't know. I don't think that defining is that essential. I think it's a good exercise, but I'm not sure if we need to agree a universally defined way of understanding life because the definition itself seems to be ever evolving anyway. Right? We have the NASA's definition. It's. It is. It has its minuses and pluses, but it seems to be doing its job.

But what are the different. If there is a line and it's impossible or unproductive to define that line, nevertheless, we know it when we see it is one definition that the Supreme Court likes. And that's a kind of an important thing to think about when we look about when we look at life on other planets. So how do we try to identify if a thing is living when we go to Mars, when we go to the different moons in our solar system, when we go outside our solar system to look for life on other planets?

It's unlikely to be a sort of smoking gun event. Right? It's not going to be, hey, I found this.

You don't think so?

I don't think so. Unless you find an elephant on some exoplanet, then I can say, yeah, there's life here.

No, but is there a dynamic nature to the thing? Like, it moves. It has a membrane that looks like there's stuff inside it doesn't need to move.

Right. I mean, like, look at plants. I mean, they, they grow. But there are plants that can be also pretty dormant. And arguably they are most. They do everything that. One of my favorite professors once said that the plant does everything that a giraffe does without moving. So movement is not Zen statement necessarily.

But on a certain timescale, the plant does move. It just moves slower. Yes, it moves pretty.

I would say that it's hard to quantify this or even measure it, but it is a life is definitely chemistry finding solutions. Right? So it is chemistry exploring itself and maintaining this exploration for billions of years.

Okay, so a planet is a bunch of chemistry. And then you run it and say, all right, figure out what cool stuff you can come up with. That's essentially what life is. Given a chemistry, what is the cool stuff I can come up with if.

That chemistry or the solutions that it embarks upon are maintained in a form of memory. Right. So you don't just need to have the exploring chemical space, but you need to also maintain a memory of some of those solutions for over long periods of time. So that's the memory component makes it more living to me. Because chemistry can always sample, right? So chemistry is chemistry, but are you just constantly sampling or are you building on your former solutions and then maintaining a memory of those solutions over billions of years? Or at least that's what happened here.

Chemistry can't build life if it's always living in the moment. The physicist will be very upset with you. Okay, so memory could be a fundamental.

I mean, life is not just life. I mean, life is obviously chemistry and physics leading to biology. So this is not a disciplinary problem of one discipline triumphing other discipline. It's that. But what, what you need to have is definitely chemistry is everywhere, right? I tend to think you can be a chemist, you can study chemistry, you can study physics, you can study geology anywhere in the universe, but this is the only place you can study biology. This is the only place to be a biologist. That's it. So definitely something very fundamental happened here. And you cannot take biology out of the equation. If you want to understand how that vast chemistry space, how that general sequence space got narrowed down to what is available or what is used by life, you need to understand the rules of selection. And that's when evolution and biology comes into play.

So the rules of natural selection operate to you on the level of biology rules.

I don't know if there are any rules like that would be fascinating to find in terms of the biology's rules. That's a very interesting and it's very fascinating area of study now, and probably we will hear more about that decades to come. But if you want to go from the broad to specific, you need to understand the rules of selection. And that is going to come from understanding biology. Yes.

Well, actually, let me ask you about selection. You have a paper on evolutionary stalling, where you describe that evolution is not good at multitasking or like in populations that evolve quickly. It's a very specific thing. But there could be a generalizable fundamental thing to this, that evolution is not able to improve multiple modules simultaneously. I guess the question is what part of the organism does evolution focus on to improve?

Yeah, that was the driving question. We meddled with the part where you shouldn't be messing up with translation.

This is the should or should not, you shouldn't.

As I said, there are many ways to break it and all life needs it.

That's one of the your favorite things to do is to break life. See what happens.

It's. Yeah, because that's how kids learn. Right. So you have to break something and then see how it will. Then you do it over and over again to see if it will fix itself in the same ways. Yeah. So that's, it's our. I don't know, it's the most fundamental properties of our, ourselves as human beings. So if we shouldn't break translation, then we should try to break it. Yes, to see how it will repair.

So which part did you break?

I broke elongation.

So what's the role of elongation in this process?

So we have four steps of the translations. Initiate. Elongate. So elongate the chain of the information chain that you're now creating the peptide chain, or let's say broadly polymer chain. And there's the termination step and there's the recycling. So all of these steps are carried out by proteins that are also named after these steps. Initiation is the initiation factor. Protein elongation is the elongated protein. We broke elongation. So the starting codon could still arrive to where it's supposed to go. But the following information couldn't get carried out because we replaced elongation with its own ancestral version. So we inserted roughly a 700 million year old elongation factor protein after removing the modern gene. So we made this ancient, modern hybrid organism.

And that essentially creates in some way the ancient version of that organism.

I wouldn't say. So it's the, it's a, it's a hybrid organism. It's not necessary because you. The rest of this cell, the rest of the genome is still modern. And that goes back to the difference between Jurassic Park. There are many differences. Obviously, given that this is not fiction, we're doing it. But also we are not necessarily. I think in Jurassic park they are taking an ancient organ, they find an ancient organism and then put a modern gene inside the ancient organism. In our case, we are still working with what we got, but putting an ancestral DNA inside the modern organism.

So you're like taking a new car and putting an old engine into it?

In a way, yeah. Yes.

Seeing what happens.

Yes. But in our case it's more like transformer than just a regular car. It is doing things.

Yeah. So it's a more complicated organism than just a car. I got it. So what does that teach you?

We wanted to understand multiple things. One is how does the cell respond to perturbation? And we didn't just put the ancient DNA, we inserted, we sampled DNA from currently existing organisms. So the cousins of this microbe and collected DNA sequences from the cousins as well. So both ancestor and the current cousin DNA, so to speak, and engineered all of these things to the modern bacteria and generated a collection of microbes that either have the ancient component or the variant, elongated component that still alive today, but coming from a different part of the tree.

So you broke elongation. Was that something you did as part of the paper on evolutionary stalling, to try to figure out how evolution figures out what to try to improve? Did that help?

Yes, because we were not supposed to mess with the translation. That's exactly what we did. And we altered elongation by changing it with different versions of elongation that are either coming from species that still are around today. You can imagine them as sitting on the tips of the tree, near branch, far branch, compared to the organism that we're working with, cousins, distant cousins, as well as the ancestors of the bacteria that we are now modifying.

How much different variation is there in that elongation step? Like what are the different flavors of elongation?

That's a very good question. So mechanistically or mechanically, it's the same, it's very conserved. So all life elongates the same way. It's nothing but a shuttle. You just carry the chemical with you. The bit to the heart of the machine.

Is it essentially doing like a copy paste operation?

It has its tail that's attached to the code, which is then carried biochemically to the linear chain to the core of ribosome. And it sits on there, it's released, and the peptides click the codes, rather, click. Once that chemistry that is at the tail end occurs, the protein leaves the center. So you can imagine it's like it hops in there and hops out. And when it hops and hops out, it leaves the information behind. That's all it does, is bring the information get out of there. And it's all triggered by biophysics, biochemistry, because of the way the enzyme chews energy, in this case gtp, how the phosphor leaves the center, that gives the additional kick to the enzyme to leave the center.

So which parts are different then? Where's the flavors of different flavors?

Usually the parts that matter don't change over time. Nature conserves the sites of these proteins that are important for its job. If there's a difference, then we want to know, especially if there's a difference between two cousins. And we look at the sites that interact with the most important parts of this machinery. If we see any Difference, we tend to mutate or we revert, we engineer that part, we alter that part, because it gives us a clue that there must be something interesting going on here. Or not.

Okay, so that's not the fundamental part of the machinery, but it's some flavorful characteristic that you can play with.

So now you stripped the machinery down to its parts and now you are looking at the parts of the parts. And it depends where you're looking and how you're looking and what you're looking at. But usually we see up to 70% level conserved identity across all modern versions. When you travel back in time, the identity decreases. So elongation likely existed. We have good reason to think that it existed at the dawn of life. So you're looking at a 3.8 billion year old mechanism. And when we look at the ancestors that we resurrect, we see about 40% identity. So the identity definitely decreases as you go back in time, but still, 60% shared information over 4 billion year, it's pretty good.

Is that just for elongation or for the entire translation?

Depends on what you do. So for initiation, we've also recently published this, it's a different story, but overall you see high level of identity that is kept intact, especially if the component is essential for Life.

Okay, so 40% and 60, 70% you said, but like from generation to generation, how does evolution, and presumably that's what that paper is looking at is the parts of the parts. So how does it able to say, like mess with the parts and try to come up with a cooler, improved version of the organism?

Yeah. So let me describe to you what we did in that experiment. We took, we took bacteria, we perturbed the elongation in all of these with different variants. So we had an initial set of group of bacteria that we had. We then subjected these bacteria to evolution in the lab. Right. So first of all, we knew we broke it because upon engineering, we measured what's going on with the cell. It's not growing as well, they're not healthy. We can see it with our eyes, we can measure it, that if they were generating an offspring every 20 minutes, now it is 40 minutes. So we really mess them up. They don't want to work with this thing, they don't want each other, but they need each other. So we created that situation for them, which is good because we wanted to see how they will cooperate with each other to fix this problem, because we know that that's not the condition that they want to live in, especially when they know what they can do. So with that, we subjected these.