Audio Player
Starting at:
Michael Levin: Introducing Anthrobots and Hyper-Embryos [WORLD EXCLUSIVE]
January 17, 2024
•
1:39:38
•
undefined
Audio:
Download MP3
ℹ️ Timestamps visible: Timestamps may be inaccurate if the MP3 has dynamically injected ads. Hide timestamps.
Transcript
Enhanced with Timestamps
250 sentences
15,257 words
Method: api-polled
Transcription time: 96m 30s
The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze.
Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates.
Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a total listener, you get a special discount. Head over to economist.com slash to subscribe. That's economist.com slash to for your discount.
We have a world exclusive today. By the time this airs, it will have just been announced that two of your papers with you, Michael Levin, have been published with your co-authors, which are here, Ghazem Gumuskaya and Angela Tang. So Michael, I'd like you to explain what these papers are. And then Ghazem and Angela, I'd like you to explain respectively the significance of those papers. So Michael, please. Yeah.
Sure. Yeah, it's really interesting to me that these are all coming out at the same time because there's a kind of a fundamental similarity here. One, so Gizems has to do with something we call Anthrobots, which are these, it's a new biorobotics platform. They're made of
Human human cells and other kind of a self motile construct that has all kinds of possibilities for for medicine and for telling us about evolution and development and so on and so we'll talk about that.
Angela's paper is about the ability of embryos to communicate with each other and to help each other resist various teratogenic influences. So things that would normally cause developmental defects, it turns out that in a group, embryos are able to work together to better overcome those kinds of influences. And so what these things have in common is really our attempts to understand where biological information comes from.
Because in the one case, so in the anthropoid case, we have this coherent construct that is using a completely wild type human genome to make a form and function that are not the typical things you see in a normal human target morphology. In the case of Angela's paper,
What you see is that the robustness of development is not just the property of a single embryo with its own genome, but actually a group phenomenon where a number of, in fact, the larger the group, the better, large collections of embryos are able to together solve the problem of morphogenesis better than individuals or small groups. And so in both cases, it's this really interesting dynamic of where biological control information comes from.
All right, great. Gazam, if you don't mind, please, what is the significance of that?
Sure, so like at a high level what is really exciting about this work I think is this paradigm shift in our thinking about biology and nature. We tend to think about nature as this thing sitting outside waiting to be investigated by scientists and written books about but in reality what I came to realize as a designer sort of eight years ago that nature can actually be a design medium
through this new field called synthetic biology, which recognizes that biological structures have embodied computational frameworks that determine their architecture and their function. So I became really interested personally in trying to get new architectures to build themselves with
sort of composed of biological tissues. And why biology? Because biology has a lot of properties that the sort of traditional sort of construction methods don't. Well, properties like carbon negativity, it can take carbon from the environment in building itself. This idea of self-construction, you know, starting from a single seed and becoming a fully fleshed structure, healing,
Ability to take information from the environment, process that and give a output. Regeneration, so a lot of different properties that we don't see in human designed and built frameworks. So specifically for Anthrobots, why it's really exciting.
Traditionally, what's been done in the field of synthetic morphogenesis, which is this idea of creating new biological structures, is to try to edit the DNA, the genetic code, to create new patterns with cells. But that has been really limited in the type of complexity and the scale that the structures can be generated.
Antrobots is looking into biology and the morphogenetic code as a sort of more layered enterprise. So it's not only genetics, but also epigenetics, which consists of bioelectricity, as we'll discuss in the episode, as well as other sort of like methylation type of changes in the genome.
So with Entrobots, we wanted to leverage that. Can we start with one structure that is known to build something in the body? In this case, human trachea.
But then can be by giving it environmental inputs, can we get it to create a completely novel structure without touching the genome? So we still don't know what epigenetic factors and of which there are a lot of different possibilities that exactly give rise to the answer about. So the process is something we're investigating. But what we've seen is that we were able to get these cells to create a new architecture that we had designed a priori.
Um, so yeah, like in summary, there are the, the, the first sort of fully cellular living self-constructing biological robots and they build themselves from single cells and, um, they don't have sort of electrical wiring or mechanical parts as a traditional sort of robot would have. Um, but they are still sort of programmed anatomies in that sense we call them.
biobots. That's where the sort of robotics aspect is coming from. And also like the bots we know of, they can do useful work. So we showed that they are able to traverse wounded human neuronal tissues and induce repair in those wounds in the course of like three days. So in a nutshell, that's the aspiration and the summary.
Okay, yeah, I mean that's actually that's a really important point because when we made the Zenobots, some people thought that this might be some sort of result that's specific to amphibians and we know amphibians are kind of plastic with this embryonic tissues or tend to be plastic. And so there's this temptation to think of this as a
Angela, your work on embryos assisting one another, can you please explain why is that consequential?
I think the significance of this work is that it's taking a step in addressing this knowledge gap that exists within the field with respect to these instructive lateral interactions that can be used to correct any developmental defects at a level that occurs above just the cells, tissues, DNA, or organs.
So we're looking at whole organisms and how they can interact with each other and aid each other. And all of this, I guess, is important because if we're able to understand and harness these instructive cues, then we can use cells and tell them what to build and when to stop building and use them to fix defects and disease. So there's kind of this application for medicine potentially in the future.
Yeah, and you said that when you had first brought it up to Michael, you thought it was crazy. And you weren't sure what he was going to say about it. So why is it so ludicrous? I guess in the field like biology, everyone studies the genome. So the idea that our genome is going to predict what we're going to look like in our future. So our genome leads to our phenotype. And that's kind of the underlying one of the underlying principles of biology.
My work is kind of looking at, okay, well, what about things outside of the genome and kind of contradicting that a little bit because it's looking at now our genomes are not being changed, but we're able to change all these other things without affecting the genome in any way. So I thought that this is crazy. Like I'm not doing anything to the genome, but all of these different defects are forming. And the only thing that's different is the number of embryos in a cohort. Um, so I think to me that was very shocking.
Okay, so Gazem, I would like you to explain this simply. Suppose we were to isolate the cell from a human lung and we put it in a petri dish. Okay, does this cell exhibit the characteristics that are consistent with an anthropobot? Now, I imagine not. So what modifications are necessary for this cell to attain the anthropobotic status?
Sure, so our starting question was how do we get to this target morphology, which is this multicellular spheroid with cilia on the surface and where that target morphology comes from is the xenobats, except that we wanted to change how we get there.
So we talked about self-construction. We wanted a single cell to give rise to that target architecture. And then we also talked about necessity to use human cells as well as necessity to use adult cells. So our starting material for that reason could have been a
Progenitor cell from any one of the saliated epithelial tissues in the body. So not just the airway. You also have the saliated epithelia in the oviductal region or in the brain. We started with human long epithelium because the accessibility of the cells due to the research that's out there. There is a lot of sort of material available for long research due to cancer and other
other diseases. So for that reason, our starting material was the human airway epithelium progenitor cells. So these cells are already committed to becoming any one of the cell types in the human airway. So they could be
secretory cells or ciliated cells. And what was already known in the literature is that if you culture these cells in a rich matrix, extracellular material rich matrix, they will form the spheroid and a single cell will give rise to that spheroid. The problem was that in that configuration, the cilia would be looking inside to the lumen because the goal of those types of
culture methods which touches the field of organoids is to recapitulate the native tissue architecture. So when you look at the human lung to trachea you have cilia inside and you have these sort of basal cells outside. So the protocol that was already developed in the literature was mimicking that.
But for us, for our purposes, to get the cilia to look outside so they can be motile like the xenobats, we had to come up with a way to get them to flip inside out. So that's what we've accomplished and we've tried a lot of different approaches for this.
and what ended up working for us was to sort of trick those ciliated cells that develop and look inside to migrate outward. So how we've accomplished that was by changing two things, by changing the growth phase. So these spheroids are again embedded in a matrix and that's what's making the cilia to look inside. So if you remove that matrix and instead supply a low phase liquid based environment,
So human cells are already known for being plastic in a certain manner?
So, all tissues like cells that have the progenitor status are known to be plastic into a degree.
I mean the plasticity drops as you move in. Mike and Angela can also speak to this. What's known is that it drops from the embryonic state to the sort of adult and elderly state, but it's never sort of a binary thing. It is for some organisms, but for humans.
It's not a binary thing because there are a lot of protocols out there that use the progenitor cells, which are semi-stem-like and semi-committed cells that form different structures in vitro. It's just a question of how do you get it to form the structure of interest for you.
And the method that's used in the literature a lot is to use genetic circuits to do this, to get that cell to execute certain morphogenic functions. But using that method, getting to something as complex as a ciliated spheroid would not be possible. And that's why we wanted to play into the native plasticity of these cells. Now, before we get to Angela,
Michael, I would like you to talk about the correlation between the structure and the motility. Yeah, so when you look at any kind of animal, you always wonder what the relationship is between the structure and the function. In other words, typically in a standard kind of creature or model system that we study, there's a long history of evolutionary selection that makes sure that the structure that it has is actually
Perfectly suited for the functionality that that you wanted to have and so this means moving around in the three dimensional world but also physiological kinds of actions and so on and so in this case what we're dealing with here is something super interesting it's a it's a creature that although it has a perfectly standard human genome.
is not something that's ever been specifically under selection before to be a good anthropobot. There's never been any anthropobots. And so what we're looking at is a new functionality, a new set of behaviors and a new structure that underlies it. But we didn't know ahead of time what that relationship was going to be.
I'm in the team did this really amazing job characterizing two things first of all characterizing all the different shapes and that they come in that these enter what's coming and there are a few discrete forms and so this is there they're actually I'm called a morphe types because they're actually discrete types of shape that you have in terms of the distribution of the Celia and and you know what the overall shape is and so on
And then we characterize the behaviors and they have a variety of different paths that they can take through the medium. And that looks very much like there's a there's a notion in behavior science of
of an ethogram, which is basically just a diagram of the different behaviors that an animal can do and which of those behaviors can statistically likely to follow which other behaviors, you know, so you'll like for some sort of fish, you'll do like there's this particular kind of mating behavior that does and so if it if it darts a certain way, then next it'll do something else and that's a 50% chance of doing something else.
So you can you can start to build up this diagram and so what we did was to study these these anthro was as if they were a new kind of creature because in an important sense they were we didn't know what all the behaviors were going to be.
We didn't know what the relationship with the with the shape was going to be. And so they because then worked out how the different shapes produce the different motions and really what what the behaviors are and how they how they tend to follow each other, you know, one after the other. And I really think that this is also just the beginning, because, you know, this is the first paper on anthropos. This is a very basic characterization. We need to do all kinds of interesting things now to see how they react to their environment, how they
Were you surprised that they split into these distinct forms?
There were many possibilities. So one possibility is that they would have no behavior at all. Another possibility would be that they would only have one type of behavior. Another possibility would be that there would be multiple behaviors, but any given bot could only do one of those behaviors. So you really don't know what to expect.
and the fact that these behaviors are discrete that you can characterize and this is an unbiased statistical analysis showed us that there were in fact classes of behavior. The fact that they fall into these classes is super interesting and I look forward to discovering what other classes of behavior there are and then really fleshing out the ethology of this new form to understand why it changes behaviors when it does.
Angela, there's a term in your paper called cross embryo morphogenetic assistant. So it's this mouthful, which is abbreviated to SEMA. So did you point SEMA? And what is it? What does it mean? Yeah, so
Mike and I wanted something that was catchy and very obvious. So the long form of it, what you just said, is kind of very obvious. It's how embryos can communicate with each other and assist each other. And then just to make it not so much of a mouthful, we decided that we would just call it SEMA for short. And then the whole phenomenon is this idea that
these larger groups
this is how I handled it, and now you too can survive the stressor. So what we see is that as we increase our cohort size, we see an increase in survival depending on the tradigen that we're using, but also as our group size increase, we also see a decrease in the types of certain types of defects that we know. So this is both increasing survival and decreasing the frequency of defects.
Okay, so I'd like to attempt to explain this extremely elementary. So people have embryos. Embryos are like the seeded babies, single cell babies or extremely small babies. Okay, when you have just one of them and you give it something called a teratogen, which is something that disrupts the development of one of these little guys, so they grow up abnormal. It turns out that when you have multiple embryos next to one another,
that if you give it a teratogen, it's more resilient to these perturbations. It's more stable. So it's as if they're communicating somehow these embryos are communicating and saying, Hey, don't mess up your arm like this. Here's how to make a correct arm. Is that approximately correct? Yeah, that's, um, I think the, how we're envisioning this happening kind of
Michael, there's this concept called group wisdom. Is there some variation of that?
Yeah, you know, we know there's this phenomenon called the wisdom of the crowds. And it's there used to be this old thing where they'd be in certain in stores, it'd be this giant jar of jelly beans, right. And people would try to guess how many jelly beans. And so the result there is that every individual person is quite often nobody has any idea how many jelly beans are there. But if you just average all the guesses, it turns out that the crowd is spot on.
And so this phenomenon has been known for a long time. And our initial hypothesis was that something like that was happening, that basically every embryo was contributing some amount of information, and together they were able to have a much more steady view of what a correct embryo was supposed to be. And remember, these are frog embryos, so they start out as frog eggs, and they're about a millimeter in diameter, and there's many of them developing in a petri dish.
So originally that's the thought we had, but then Angela found a very interesting piece of data which suggests that the real story is more complex and more interesting, which is that if you were to compose a group of some embryos that were exposed to teratogens and some embryos that were never exposed,
The simple model would suggest that it would actually work quite well because the animals that were never exposed to the teratogens, well, they have a very good idea of how to make their body and their organs and they should do a great job of informing the others of any information that they might be missing in terms of that teratogen. But it turns out that actually that doesn't work. The optimal effect is when everybody was affected by the teratogen and it's the
It's the result of having confronted this, this influence and having overcome it. That is actually what's necessary. Everyone has to have seen it in order for the group to know what to do, which is, which is super interesting. And it means that it's, it's a much more active process. It isn't just that, you know, I know what I know, and I'm just going to spread that information and everybody else can make use of it. No, it's actually that most likely what's happening is that that information is being derived on the fly as embryos are encountering
these external threats, they are deriving signals that are important for them to reinforce the normal path that they take through this space of possible configurations and they're sharing that information with each other. Now Michael, is it less harmful because you're giving the same amount of teratogen to two embryos that you would to one or are you titrating proportionately? So you'd give 3x if there are three embryos?
That's a great question. Of course, we titrated everything. We scale up the amount of teratogen based on how many embryos there are. Everybody's still getting the same amount of influence. Angela, in the paper, there's the term inter embryonic communication and inter embryonic interaction. What would be the difference between those two?
Yeah, so I'm thinking inter-embryo communication is, in our paper, we talk about like a potential molecule calcium and ATP that can be used as a communicator. So it's almost like a little message that's being sent between embryos.
Inter embryo interactions is when these embryos are able to grow together. So now if you have an N of one, you may still be sending out messages, but you have no other embryos to interact with. Versus in these larger groups, we have both the message and people to receive the message. You also mentioned that it doesn't rely on genetic homogeneity. So how much of a variation is tolerable?
Yeah, so within an individual cohort, we have, I don't know, like, they're not all the same, they're not all homogeneous. And none of that's not really important to us either. So there's a part of our paper in which we look at embryos or embryos from two different lineages. So we have wild type and albino embryos. So they have different phenotypes and different genomes.
And what we see is when we treat them both with stressor, there's no difference between the two. So that indicates that, you know, you don't need to have the same genome to be impacted the same way. Yeah. And can you explain what a wild type is? Because I know before reading this paper, I never encountered that term outside of Pokemon. And in Pokemon, wild type just means you find it in the wild. It doesn't belong to a trainer. I'm like, OK, what does a wild type cell mean?
That's pretty spot on. So wild type genotype or wild type anything really just means how this organism exists in nature. So it's unperturbed, there's no changes that we've done to it, no genetic modification.
I see, but you could still have wild type. So it has to be some relative term because every cell is mutated. So it can't just be like a non mutated cell. It has to be with respect to something like its relevance in a scientific experiment. You can't just find some cells and say these are wild type. You have to say these are wild type relative to something else or no.
Yeah, no, you're correct. The wild type of a given species just means that that is the standard genome that you will find out in nature. It has not been modified by the experiment. So it's just a genetics term. It means that it's not a mutant of some sort. It's just the natural. Now, having said that, of course, you're absolutely right. Even within the natural population, you're obviously going to get variability.
so it's not and in fact with these frogs these are not an inbred population the way that you might have with with certain model systems these are this is an outbred population and so of course there are genetic differences even between individual frogs in our colony but wild type just means that there's no there's not been any specific mutation performed on them. Like in the lab by a scientist. Great great. Gazem
How do you think different cell types would affect the formation and the behavior of the biobots? And I just noticed I used the word biobot, okay, not anthropobot, because that's in the titles, but then in the paper. So I also want to know, why did you all title it biobots, but then also coined the term anthropobot? But that's a separate question. So how do you think different cells would affect the formation of these anthropobots slash biobots?
I mean I guess like just kind of taking a step back and establishing the terminology there. Biobot is not a term we've coined, it's something that has already been discussed in the literature. What that essentially refers to is programmable anatomy.
I mean, that's sort of using biological cells, either completely biological, like in the case of entrobots or xenobots, or some sort of a hybrid between biological and a mechanical or chemical scaffold carrying those cells and providing mechanical support. So there could be a lot of different types of biobots and there have been.
I guess the field really started flourishing around 2013. What we have been trying to do is to create bio-bats that are fully cellular. The first example of that is xenobats. In creating xenobats, no external gel or mechanical or electrical substance was used.
I'm so and robots are another example of that again fully biological and the name is coming from the human origin so by about the more general term and and robot are examples of them.
Exactly. All fully cellular bio-bots, yes. And there could be more in the future that could be named differently. And so your original question was like how different types of cells could create different types of bio-bots. So it all depends on what you're trying to accomplish. No one bio-bot is better reversed than the other one. It just depends on what your goals are.
Um, for Xenobots, for example, like, uh, the goal was to create something that could, um, or rather I should say if your goal is to create something that could survive outside, like in the wild, um, Xenobots is a better way to go than Anthrobots because Anthrobots are, um, they, because they're derived from human cells, they have sterility requirements. Um, and they are better for say medicine. Uh, whereas like Anthrobots,
I'm sorry to interrupt. They have what kind of requirement? So sterility, which means that they cannot come across, they cannot interface with pathogens of any kind, bacteria or viruses or fungi. This is a general requirement for mammalian cell culture. When we do cell culture for any purpose, making biobots or investigating diseases,
There are strict sterility requirements. We work inside these sort of hoods with laminar flow that push air out to prevent anything kind of coming and landing on the cells. So with these strict sterility requirements, for example, if you're trying to build something that will go into the
reverse and try to like detect the presence of a certain toxin and report back you would not want to use a mammalian cell or an anthropobite and like xenobot would be perfect for that. Or like another example if you're trying to build something where you want to mass fabricate then you would want to have a property like self-construction as I mentioned because then you have you know each
Biobot building itself, which means you can build like thousands of them in parallel without having to, you know, do anything like otherwise you would have to individually sculpt thousands. So it all depends on what your goals are. And that also impacts the decision of what cell type to use. So yeah, I mean, it's a lot of different cell types could be used for a lot of different Biobots. It all depends on what the design specifications are.
We're just trying to bring the idea of design into biology, bring the two together and harnessing some of the properties that only biology has, such as healing after damage or self-replication or self-construction, and bring that into this process of fabrication of new structures.
Could you please speak to the potential medical applications?
Yeah, so again, one of our design specifications was that we want to use this in medicine. And like Angela was talking about, having them be vile type was really important for us. So very not sort of inserting any foreign DNA that could when in turn deployed in the human body could have off target effects.
So for that reason it was important for us to keep the human DNA wild type so we've tried from more than twenty different human donors and in every single time we're able to create an answer about and that's across a lot of different ages and you know genders and races and we've seen that this works with a lot you know diverse like human genomes.
So what sort of more specifically in the medical field that we are hoping to accomplish is can we take a cell from a human donor? So currently, as we mentioned, we're taking these from the human airway epithelium, but down the line, you can envision taking a cell from a human patient's skin.
And then turning it into a induced pluripotent stem cell, which in turn would basically revert the clock and have the ability to then be differentiated into all these different kinds of tissues in the human body, including the human lung.
So in terms of application, that's what we're envisioning. That's not something we've shown in the paper, but that protocol has already been worked out in the literature. So starting with a human skin cell and then turning it into an into an anthropo that is geared towards a specific application based on what that patient might need. And then when we put that into the body, it is a synthetic construct. It's something that doesn't
Yeah, I was going to ask about if you all have tested biocompatibility or immunogenicity if you're already envisioning it for medical applications.
Not yet our preliminary experiments have been with human cells but in vitro only. So next step for us would be ex vivo tissue so human cells extracted from humans and then after that it would be in vivo or proxies for in vivo experiments so that would be step three.
Yeah, but the but the I mean, these cells are already coming from the they've already been inside the patient. So while while we haven't specifically tested the immunogenicity in vivo, the chances are very high that it's going to work. I mean, these are
These are the ideas personalized medicine. It's a bespoke kind of construct that's made of each patient's own cells. So it's likely fine. And I just I want to I just want to underscore the amazing the thing that just just blows my mind every time I think about it.
You know that last figure in that paper is basically showing just one initial thing that we found that these guys can do, which is to help neurons heal across a scratch wound in two-dimensional culture. Just to think about that, the tracheal cells that are sitting in your body and they sort of sit there quietly for decades doing their thing and using their cilia to waft little particles and mucus and stuff out of your lungs.
The fact that if if uh liberated from their environment and given a chance to kind of reboot their their their multicellularity they now have the ability to go around and repair defects in other types of cells. We would have never known that it's just amazing to me that that that they have that capacity and it makes me wonder what what else you know what other cells are sitting around the your body with capacities to to heal other components and to have other beneficial
Pro regenerative types of outcomes on different parts of the body like that. That idea of releasing the native healing potential of your own cells and letting them do new things that might be beneficial for the body. I think is incredibly powerful. I think we're just seeing the first glimpses of that here. Can you talk about how this fits into the larger framework of your work? Because as I heard Gazem say,
Take a skin cell and turn it into a pluripotent cell. It reminds me of our previous conversations. Yeah, yeah, there's a few. The kind of applications of these are in several different directions. On the one hand, we certainly want to use this for very specific practical purposes, so we think that once we gain a better understanding of their kind of native functions and a little bit better on the programming end,
We will be able to address all sorts of very specific conditions and we can sort of run down some of the early ideas that we have. But there's a bigger picture here, which is using this biorebotics platform as a kind of simplified model system in which to crack the morphogenetic code. Think about all of the problems of biomedicine, including birth defects, traumatic injury, or thus failing to heal from traumatic injury.
cancer, degenerative disease, all of these things have one thing in common, which is that they would go away if we had the ability to tell groups of cells what to build, right? That's the major rate limiting step for regenerative medicine is that we do not understand how cellular collectives make decisions. We're pretty good on the hardware side for individual cells, right? So we know how cells differentiate, we know what the various
Lots of various genes do and how they interact with each other and so on but but this this idea of how do collections of cells make decisions that they're going to make a hand versus a foot versus something else and more importantly how we communicate our patterning goals to them that is if you want to build a new organ or you want to repair an existing organ or you want to make something that has never existed before
what information do you need to give to these cells and what interface can you use to get your goals across to the cellular collective? And that I think is critically important for unlocking the promise of regenerative medicine. And so that's what we're starting off here, because you really have to, before you can use all these fancy programming techniques, and that includes not just the traditional symbio that people are using, but also the stuff that we do in our lab, which is bioelectrical,
What are the baseline plasticities and competencies of these cells? What do they already know how to do and why? Why do they make decisions to take specific paths through anatomical space and build specific kinds of anatomies and so on? And so I think that's in the greater scheme of our lab's work, which is to understand how to communicate with the collective intelligence of cells. This is a very important
a model system in which we can now ask, okay, what kinds of stimuli, what kinds of information can we be giving to these cells to get them to build various things? Much like with the xenobots, you know, these first papers were all about characterizing their background kind of native competencies. We didn't engineer the heck out of them with new genes and all this stuff. We can and we probably will in the future. But step one is to understand how do collections of cells make decisions about what they're going to do? Michael, can you also
indicate again these are two different papers here that have an overarching theme but outline how are they distinct and how are they the same so one has to do with anthropobots or biobots and then the other has to do with this embryonic communication and the resilience so please yeah the the the common there are many common themes but one one important one is collective decision making so it's again this idea of so so in the case of the anthropobots it's a question of
understanding how groups of normal cells with normal human genome derived molecular hardware are going to decide to work together to make a specific new coherent construct with new behaviors, new functionalities, and so on. In the case of Angela and the cross embryo morphogenetic assistance, it's the idea that standard developmental biology studies how
cells cooperate to make a nice embryo. Well, it turns out that this actually works on a higher level as well. So groups of embryos also work together to complete morphogenesis. And in both of these cases, what we want to understand is where is the information? What is the collective intelligence of these cells? What kind of problems are they able to solve? So in the case of the Anthrobots, they're able to
uh they find themselves in the new environment in the new kind of a new scenario and they're able to put together a very coherent form that is able to to live for weeks and and then have certain functions and so on in the key in an angeles case what you're seeing is again a kind of collective problem solving but this time at the level of whole animals.
So not down at the cellular level, which is standard developmental biology. It's kind of, you know, maybe this is the beginnings of a kind of a sort of hyperdevelopmental biology where what you're really trying to work out is the rules by which whole bodies communicate to better achieve that better solve the problem of of embryo genesis.
because one of the things that, well, many people have studied in our lab focuses on in particular is biological intelligence in the sense of problem solving. That means when you're confronted with a new scenario that you haven't seen before, especially a new scenario that you haven't seen before, are you able to complete your goals in the case of development? Are you able to make the target morphology that you want to make a correct embryo or some other functional thing in the case of, in the case of anthropos?
And so that's what we're seeing in both of these projects. We're seeing new unexpected competencies at different levels, at the level of cells and then at the level of organisms, to do something helpful and coherent in novel circumstances. Angela, can you please talk to how this robustness increases with more embryos? So one embryo fails more often than if you have a collective and how you found that out?
Yeah, so whenever I do my treatment groups, so when I start stressing out these embryos, an N of one will almost never survive by itself. I probably did like 50 dishes of just one embryo each and never did I ever really have a survivor. So there was something about that, just the fact that it's getting the same amount of drug as my larger groups.
And then as we kind of scale away from there so now I'm increasing my group sizes.
If we're just looking at survival by itself, I see that survival starts increasing once you hit like, I think something around a group of 50, then you'll see an increase in survival. And then as you go up to like 100, you get like 80% survival. And then once you hit like a group of 300, you get almost everyone surviving.
So the whole premise of this was I had been given a teratogen and someone told me, hey, this works in my hands, I want to try it out. And no matter what I did, I could not replicate that. And I could not figure out why because I had this person telling me exactly what they did. And it came down to the number of embryos that she was using versus the number of embryos that I was using. And so to me, I was like, that's crazy that, you know, everything else was held constant.
But the only thing that was different between our two experiments was just purely the number of embryos that we were using. So you stumbled upon this. Yeah. Interesting. Interesting. Yeah. And I just, you know, I want to, I want to emphasize a couple of things here, which again are really, really striking to me.
One is that what this means and what Angela just pointed out is that a lot of people when they report toxicity values for different drugs. So there are thousands and thousands of papers in the literature saying that here's this new potential toxin or a drug or something else that can end up in the environment. And there are these tests using frog and zebrafish embryos that attempt to quantify how disruptive it is to development.
What we now know, and a lot of people don't control for actually the number of embryos that you had in your cohort. They just do a percentage and call it a day. So what this is telling us is that many, many studies in the literature are not actually reporting the raw
a danger value for these chemicals, they're reporting the corrected value after the group has been able to resist it, right? And so what you just heard was that if you do these things on single embryos, the actual teratogenicity is very high. It's extremely potent, but you start to be blinded to that effect the more embryos you put in, because what you're seeing is how dangerous it is after the embryos have had a good chance to correct for it.
And so that's really important that now we know that when we
examine the potential of various interventions to cause developmental defects. We have to ask what's the, what's the raw effect size, right? What's the actual teratogenicity and then how well do the embryos do to correct for it? So that's kind of a very practical thing that, that, that we now know that I think is important. And the other thing that is really striking is that the standard story of developmental biology and where the information comes from for you to be able to build a normal embryo,
is basically supposedly just two things. It's your actual genome and then it's the maternal components that are in the egg. So basically your parental genome that are, you know, there's some stuff provided for you in the egg for the zygote and that's it. And the idea is that the genome has everything you need to complete development and development is rightly so described as a very robust process. Most of the time it goes correctly despite its incredible complexity.
And what we're seeing here is that that purely vertical view, the idea is that that you could be one embryo, uh, sort of, uh, far away from anything else. You've got your own genome and that gives you everything you need to know. That story is clearly only partially true that yes, you've got, you've got the hardware that you need, but actually that, that hardware is not as robust as you think without other individuals around. So development in a sense is a group phenomenon. It's a, um,
That traditional robustness level of development is actually a collective, it's a collective property. It doesn't work as well in a purely vertical sense, being passed down from a genome to one organism. And does that contradict the current thinking in developmental biology? Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon
Jokes aside, Verizon has the most space to save on phones and plans where everyone in the family can choose their own plan and save. So bring in your bill to your local Miami Verizon store today and we'll give you a better deal.
Well, in the sense that this kind of effect has not been described before. I mean, people have seen things like alley effects and in terms of groups of animals surviving in some environment better than individuals. So that has been noted, but it wasn't known why that happens.
And I do think that it contradicts the emphasis on purely vertical transmission and the idea that one embryo has everything it needs in its own genome. I mean, these are otherwise fairly conventional mechanisms that we're studying. For example, one of the other amazing things about that paper is that they were able to, Angela and Megan were able to literally visualize waves of information
passing across embryos. We have videos of where you can actually see them. And in order to make it easier, what we did was exert a very specific event. So basically like a needle poke into one of the embryos. So that way, you know exactly when it starts, right? And then you can actually see using this calcium indicator fluorescent dye, you can actually see this wave of information passing through from the point of injury through the animal
Now Angela, this wave of information, what is the mechanism? What is its physical component? Is it an electrical field? Is it just calcium ions being thrown? Is it something else? Yeah, so our current hypothesis for
The mechanism behind all of this is that some sort of injury occurs and then the embryo that's receiving this injury elicits a calcium response. This calcium response then releases ATP into the media and then surrounding embryos are able to uptake this ATP and then elicit their own calcium response. So there's kind of a
Yeah, this reminds me of trees, some trees in some forests, when one gets infected, it sends a signal through the roots and the others start producing antibodies before they even receive the, it wouldn't be a Tratogen. In that case, it would be some virus or whatever, maybe, or fungus. So is this similar?
It is, and animals and plants both can signal to others when they're being preyed upon. So there are examples like this where some predator is munching on a leaf or something, then there are volatiles released where other plants in the environment can feel it. One interesting thing about this though is that we're dealing here with fairly complex processes. So it's not a binary yes or no, right? It's not a binary, am I being attacked or aren't I? It's sort of
Well, normally I would build a head of a certain shape and size, and now I'm unable to do that. And so passing information on how to build a tadpole head requires lots of information. And this is one of the big mysteries going forward, is how do you encode all that information in a single, you know, a signaling molecule that's passed?
Right, because it's not just the yes or no, you have to actually, I think you have to actually encode considerable amount of information to for the embryos to help each other. So whether that's happening through some sort of modulation of pulsing through the water, or it's got to be something other than just simple, like here's your ATP concentration, and that's if it's a single number, you're not going to encode head morphogenetic data in one number.
And the other thing to mention is that these projects are also going to come together because one of the things that I would really love to see is how this shapes out in the anthropobot case. So one of the things that we will be doing in the future is looking to see, do the anthropobots communicate with each other? Do they communicate with the other tissues that they find themselves in an environment with? How general is this? Because obviously it begins in the frog model,
Of course, as Angela said earlier, ultimately, you might want to basically fake it for biomedical purposes. So whatever signal is allowing tissues and organs to form properly, in these high density groups, you would want to be able to induce that on demand in the patient. And so the next step leads through biological tissues and especially anthropos to see whether that kind of phenomenon is general and whether we can harness it for biomedicine.
So what I see is if this wasn't a breakthrough enough, both of these papers, so one, the anthropobots and then some of these anthropobots, I believe heal other cells or other tissue. Is that correct? Okay. Yeah. And the number two, the inter embryo communication or the SEMA effect that you have where different tiny babies can tell other tiny babies like, hey,
Protect yourself and hey, let me help you with your morphogenesis. Not only that, but number three, there may be large discrepancies in the literature or misleading effects because you mentioned this word raw taro. Sorry, raw. Can you repeat it for me, please? Teratogenicity. Yeah, I'm not a biologist. So raw teratogenicity that someone may be throwing out some harmful chemicals to a single embryo.
And then someone else may be testing it on 10 embryos, but then I need to be clear here. When you're saying that there's some discrepancy in the literature, are you saying that they're reporting the raw amount? So let's say 10 milligrams of some teratogen, but they test them in 10 embryos. Do they then divide that by 10 or are you saying that they don't do that because they don't even tell you the amount of embryos to begin with?
Well, what I'm saying is that because no one had known before that the number of embryos actually determines how effective your teratogen is going to be, it means that when they report, I mean, other than testing it on different size cohorts, there's no way to know. You can't simply divide it by the number of embryos.
And so when somebody says a certain, and by the way, in Angela's work, this is important. It's not just about chemical teratogens. She also tested RNA, so mutant proteins. And so this is a much more general thing. This is not just about chemicals.
And so anything, including potentially a wide range of genetic mutations or drugs or other kinds of interventions, you know, it could be, who knows, maybe it works for radiation, maybe works for temperature induced defects. We don't know. But the idea is that what you're seeing isn't the real effect. What you're seeing is the effect after it has been corrected by the group.
And so we really need to be sensitive to that. We need to understand that that depending on the size of the group, you may be under reporting the actual disruptive power of this thing, because if the group had corrected it, there's a similar there's actually a really interesting similar phenomenon which affects evolution, which is that a lot of these model systems and animals have
ways during embryogenesis of ways of repairing certain defects. So this is for just as an example, and many people have published other examples, but in our in our work from years ago, if you scramble the craniofacial organs of a tadpole, so you know, the eyes on the side of the head, the jaws are off to the side, like everything's everything's scrambled, they will actually find their way back to the right locations, right? They have the ability to individual embryos have some ability to fix these things.
And so what that means for evolution, just think about if when selection gets hold of that embryo and everything is in the correct place and it's a beautiful embryo.
selection doesn't actually know was it beautiful because the genetics were amazing or was it beautiful because actually it started kind of a mess but it's really good at fixing things and so that that problem is not just for human scientists it's also for the evolutionary process itself that you're often not seeing the actual phenomenon what you're seeing is what's left over after the competent
material which is which is cells and tissues have had their say and we have been talking about this for for a long time in terms of individual cells but now in this work you're seeing that it's also a property of groups of embryos you know this ability to mask defects and to um uh and to really uh can kind of uh not let you see the full impact of what the disruption would have been.
So what if someone says, okay, Angela, this is all nice and good, but how does this apply to our species where we have predominantly one embryo? So in other words, we're not all octomoms. So who cares, let alone 50 moms or 100 moms?
Yeah. So I guess for humans, we look at this phenomenon and it can occur post like birth. So things like looking at how skin to skin affects mother's newborns, right? Like that's a communication just in a different type of way, no longer like this chemical, teratogen, embryo to embryo stuff, but there's still this communication interaction between mother and child.
Also, if we look at like other things such as emotionally, right, like humans talk to each other, we take care of each other. Talking is a type of communication. So we see that in the case of humans, it might not be a particular molecule that's being passed along, but it could be different forms of this interaction, whether that is contact or word based. I think that the phenomenon still holds in people.
And by the way, you can see in the, in the, in the, uh, in the anthropocase, you can see a microcosm of that happening because these, these anthropos are actually helping to heal the neurons that the neural scratch that they come across. Right. And so, so again, it's this, it's this notion of, um,
Yeah, something I was going to ask is what are the boundaries associated with this morphogenetic assistance? So you've established the cellular and what about
Organ level or tissue level or cytoskeletal or what about us as people? Is there some analogous mechanism through which we as humans as people we're influencing one another right now?
Yeah, quite possibly. This business of having had to be exposed to the teratogen before you become helpful as part of the group, maybe there's a human analogy too. Mark Solms told me that to be a good psychoanalyst, you have to have been psychoanalyzed yourself and gone through that process. Maybe this is truly scale-free in the sense that you see it in cells and tissues and
We have a new project starting now where we're going to look at that, and we're going to look at what the limits are of this. So in the frogs and so on, we're going to look at how general this phenomenon is. And then the same thing on the anthropobot side,
We're working with a company called Astonishing Labs, which again is going to let us really go wider and try to understand, okay, so they heal neural scratches, what else? What else can they heal? And how much of that is innate and how much of that can we bring out and how much of it is programmable and how much of it is inducible and so on? Gazem?
I have some notes here that I wrote down from going through your paper. I have it written down here as capable of navigating and promoting repair, at least in cultured human neural cell sheets. And they're formed without genetic editing, which seems to highlight the morphogenetic plasticity of wild type cells. Okay, so something I want to know is in science fiction, there's this trope where you swallow a pill and it has nanobots inside and these little robots operate and heal you.
So the deleterious effects are vanquished, at least ameliorated, maybe injuries are repaired because of the operations of these tiny machines. So do you see a future where your anthropobots can serve this purpose? Or in other words, were these movies talking about your work, Gazem, and they just didn't know it? Yeah, I mean, I think that it's been kind of kind of kind of eyes on a lot for kind of harnessing what
nature can do in context that are not expected or unconventional. So exploring human body, it's sort of as difficult as exploring the outer space. And we have spacecraft and we're doing a lot of efforts on that front. But I think that sort of brings
It really triggers the science fiction community to think about how that kind of exploration can be extended into other unknown frontiers like the human body. So yeah, I mean, we are really interested in seeing if the inter robots can be deployed in tissues that are otherwise inaccessible.
to direct operation or accessible but invasive like surgery. One of the things we are really trying to understand here and we talked about the morphogenetic code is to see if we can leverage this computational ability that's inherent in biology. So what that
And that comes into layers, right? So the computational ability to build itself. So by taking information from the environment and processing that to grow its body, as well as once its morphogenesis is complete or at adult sort of stable state, to take information from the environment and process that and turn it into behavior.
So I guess in the context of navigating the body, it would be more the latter, a evolved and well developed and sort of matured entrobot. Can it go through different tissues and collect information or trigger change? So that's there are a lot of different sort of avenues there one could follow. We have so far only looked at the healing in the context of
normal damage but yeah very hoping to look at other things like can they clear plague from the arteries by again been encountered with a like a piece of plague like adipose tissue can it detect that and then release some sort of a molecule that would help it
bulldoze. So combining the ability to release molecules as well as its physical thrust. So like bulldozing while releasing some sort of a molecule that would melt them, melt out the adipose tissue. Um, or can it, um, again, want to make it very clear that these are some things we're kind of aiming for and, uh, would require more research for us to demonstrate in the lobby have not yet.
But these are pretending you're just keeping all the goods for yourself, pacing out the papers. I'm going to tell the Emerald Gates Foundation.
Since we're talking about sci-fi, we can elaborate on... I mean, I personally think that anthropobots or other synthetic living biological tissues can, if you can scale them up, could even be used for construction, like for actual inhabitable structures. So I actually have a background in architecture and what really brought me into this field of... That's interesting.
Yeah, I mean, I recognize that in biology there is this ability to embody, you know, self-construct with sort of this embodied morphogenetic code. And as Mike was talking about, this comes in layers. We're often sort of conditioned maybe from, you know, middle school biology classes to think about DNA as the rule book for everything. But what we're discovering
is that yes, genetic code is important, but there are additional layers. You know, there's epigenetics or bioelectricity that together make up this morphogenic code. And how can we edit this code to steer these cells into completely new architectures that never existed before? Whether to use it, either to use it in medicine, in these ways that we're talking about,
Or again, since we're talking about sci fi, I'm personally interested in scaling this up into building like self constructing bricks. And then, you know, by using those bricks, can we build living architectures? Because, you know, yeah, I mean, when you look at the global warming, like more than 40% of the CO2 released to the environment is coming from the construction industry, just trying to like build things. Yes, yes.
And that's using the knowledge that we have developed as humans for thousands of years in the field of architecture, civil engineering.
But at the same time, when you look at nature, it's also able to build structures that scale like oak trees or a lot of, you know, like you see a whale and a biology can build, you know, large things without let alone releasing carbon, but by sucking carbon from the environment. So it's literally carbon negative, the exact opposite of what we are doing as humans.
And now, like in the 21st century, we're learning that that's actually also programmable. It's not set in stone. You're not limited to only what's evolved out there. But as a biologist, engineer, designer, you can actually edit nature and create new structures. So why not create structures that we can use to solve some of the other problems like sustainable construction?
Yeah, or even space exploration. I mean right now it's real difficult to leave the planet because of the gravitational pull. So the more weight you have
you know, aboard in your spacecraft, the more you're being pulled back, you'd have larger rockets to get you out. But then those rockets also pull you down even more because of their weight and so, you know, intricate balance. And are you really going to put in a bunch of like dead weight, like bricks, you know, or concrete precursors to build in like outer space? Instead, can you just take a tube of engineered cells
Right. That's super cool. So Gazem, you don't know this, but my background's in math and physics and I often wonder what would I do if I had to do it over and I had to choose something else? What would I choose? And I
I think it would be civil engineering. Well, I often say this because I look at houses and how they're constructed and it looks the same as it's looked for the past few decades and it takes just as long. And I always think, man, I wish that that could be done much quicker. I'm also a germaphobe. So in addition to being in mathematical physics, there's germaphobia and I'm a nerd.
Dragon Ball Z and Dragon Ball Z. I don't know if you know this show, but you have this little capsule you bring with you and then you throw it on the ground and it becomes a house or it becomes whatever. Yeah. So that's really the dream. Yes. I like this idea of building a robot from smaller pieces. Also because like I'm a germaphobe and I want robots to clean washrooms. I just feel horrible for janitors. I feel like, oh my gosh, how do they do that? And they should be paid so much more. I wish a robot could just go and do that. And then a robot would deconstruct because then you'd wonder how do you clean the robot?
Yeah, so I hope that you could make some inroads there. Yeah, I mean, we have microphages that go down like chase bacteria and swallow it and degrade it. So we can maybe transfer the same principle to bio bots. What's the lifespan or functional lifespan of the Anthrobots that you've been studying?
Yeah, so for them to build themselves, it's about two weeks. And then for this, so we at the very beginning talked about releasing them from the matrix and then having them sort of turn inside out that morphological reorganization that takes
set of another week or so so three weeks is for their developmental um developmental face which we now work we're working on a follow-up paper to characterize those stages more but once it's a sort of adult bot that is able to move around
There is a variability there, we've seen them living from sort of month to multiple months but what happens in every single case if you're not like giving them different drugs or anything in their wild type and throwback case what happens at the end of these
multi-month period is every single time just degradation into individual cells and to debris. So they're able to naturally biodegrade, which sort of helps with the concerns around, well, if you put them into body, then what's going to happen? Will they ever like create clogs? What we have seen in the lab is every single time they degrade into individual cells. So then that's not going to be any different than the individual cells that your body disposes off.
Now, I understand you didn't genetically modify these, but did you observe any other changes that were epigenetic or bioelectric? We haven't looked at their electricity. As Mike said, that's like definitely one of the next things to look at because there's so much there and that was
That concept was foreign to me until I came to Mike's lab. Up until my PhD, I was doing a lot of actually genetic editing, synthetic circuits to change morphology. And it was here that I'm realizing that there are all these other layers, like the complete morphogenetic code. So no, that's not something we've done yet, but we would love to look at.
But other epigenetic, so I mean this whole thing that we talked at the beginning about character formation, right? So none, no two entrobots are identical, but we have these morphotypes and behavioral classes that also happen to influence each other. So figures two, three, four in the paper.
I mean that has got to be a result of some sort of epigenetic influence because again every single anthropot starts with the exact same DNA but they end up in different sort of morphological flavors one is fully covered with cilia one sort of morphotype another morphotype is fully polarized so half of it is covered with cilia the other half is bald
The third type is, again, silly everywhere, but much more sparse, so like a checkerboard pattern. I mean, if all of these bots have started with the same DNA, what's causing this morphological, you know, variability? And that has got to be epigenetics. We don't yet know what those epigenetic knobs are. We have some supplementary data in the paper that discuss Matrigel, the matrix,
viscosity as a potential factor or their initial cell density so that's sort of reminiscent of Angela's work like based on how many cells we start with the resulting answer about population profile is the frame.
So those are the only two things we've looked at and we have significant differences. So those seem to be like maybe one of the first couple of knobs, but we still have a long way to understand what are the knobs and then what's the underlying sort of mechanism that's making those knobs to be the ones that are influential.
Yeah, one thing that I just wanted to add on top of that is that the original meaning of the word epigenetics was basically everything that's not the genome, right? So that includes bioelectricity, right? So traditionally that would include biophysical factors, biomechanical factors, ionic factors and so on. So when we say, at least when I say epigenetic, I don't just mean the chromatin modifications that people focus on today, you know, the methylation, acetylation, all that stuff.
but actually all these other things so we don't know I mean there could be of course there could be chromatin modification effects but but we're you know the next step is to look at the at the bioelectrics and and probably biomechanics too and other other aspects of the physiology of it. So there was a term gazem that was used and Michael I'd like you to explain it just for the audience to know it was matrix and the term escaping the matrix was used and so because we're dealing with people thinking you've cracked the code
You need to explain what's meant by that because people are going to think everyone here is 95 years old and you just made yourselves look younger. That's right. It's the cosmic matrix. No, it's not the cosmic matrix. It's yeah. So cells produce the stuff called extracellular matrix and it's basically just a collection of important molecules that sit on the outside of cells and between cells and they
It has all sorts of functions, including as a repository for information, so much like ants leave each other messages in their environment, right? It's a thing called stigmurgy when you can use the environment as a scratch pad. So extracellular matrix in vivo is this kind of like rich set of molecules that are hanging out between and outside of cells that can also be used as information.
What happened when you tried to block communication are you unable to because you didn't know what the communication was.
Yeah, so based off of our hypothesis that communication was occurring through this calcium ATP signaling mechanism, we did try to use different inhibitors of calcium and ATP. And what we see is that when we inhibit either or,
The survival of our embryos actually decrease so they basically become singleton so they act as if they're being raised by themselves so far by blocking those messages the embryos now think oh no i have no neighbors even though they are still in a group of one hundred three hundred. And so it would go down. To the level of what it would be if it was a single embryo.
So yeah, so what we saw was simply by blocking these communication avenues that these embryos are no longer able to basically sense each other. Now they think that they're just being raised by themselves, despite that not being the case. You were able to test it also without the teratogens, but with the communication blockers?
Yes, yeah, yeah. So we did just regular media in which we grow up our embryos and just added the inhibitor to see, okay, does the inhibitor have some sort of effect? And we see that by itself, the inhibitor doesn't. But with a teratogen on top of the inhibitor, then we see this communication basically going to zero. And so what were the controls that you used to establish that the group size was the factor and not some other environmental variable that is somehow correlated with the group size?
Yeah, so there is a number of things that I tried to normalize. So we tried to scale up dish size. So whether the embryos have more space or not, that was a concern. So I just would increase group sizes or dish sizes as group sizes increase. Obviously, the media and the tritogen, so how much each embryo is being exposed to was a major concern because
the first thought is, okay, well, if you have an N of one, that N of one is being hit with all the tritogen versus if you have an N of 300. Now you have more individuals to help you break that up. And so to address that, I scaled up both the amount of media that each embryo is getting as well as the data now, kind of like what you and Mike were saying, where one X, an embryo of one will get a one X versus an embryo of three will get three X of that. And so we try to normalize for as much as we can so that whatever we see is
But also the other important thing there is that in addition to these drug, you know, if it were just drugs, then you have to deal with a drug breakdown and all this stuff. So the other important set of data in that paper is doing the same thing, but with RNA injected directly into embryos. So there are no drugs. There's no issue of what happens to the drug in the medium. It's just every embryo gets the same amount of RNA that normally would destabilize development in a particular way.
And the other thing to add here, which is important, is that it's actually doubly surprising because the standard, if you were to ask somebody, okay, I have 50 embryos in addition, what I'm going to do is instead put 300 embryos in addition. What do you think is going to happen to their health? Typically, the expectation would be it should get worse because by having more embryos together, you have more opportunity for
So,
Angela, have you extended your study to see if any of the advantages observed in the early development with the SEMA effect have led to better or different outcomes as the embryos grow further? Yeah, so I mean, we've taken our embryos and we stress them out early, but then we follow their growth. So we follow their development until they hit basically stage 45, which is as long as IACUC allows us to grow.
Okay, how many of this group is still alive? So we do follow their development kind of as long as we can, at least from the one cell stage up until stage 45.
Yeah, so we have an age range from 21 to 76.
years old humans donating their sort of the lung tissue at the bifurcation point. And this these progenitor cells that sit at the base of the trachea, the tracheal epithelium, looking directly to the matrix in the body, extracellular matrix in the body. Those cells from all these different patients were able to give rise to entrobots.
One thing to clarify here is like for a specific entrobot population, we start with in a single sort of dish, let's say, we start with 15,000 cells. And we only get actually like hundreds of bots. And the dish is sort of a centimeter across. So it's a very, very tiny dish.
So not every single cell becomes an anthropobite, but every anthropobite based on what we're seeing is coming from a single cell. So in other words, from a human donor, it's not guaranteed that every single human lung progenitor cell will become an anthropobite, but enough many of them. And this is due to the senescence.
But enough many of them become that it's every single time you have a high throughput method that give us a sizable population. Well, this is also fascinating and I'm very much looking forward to seeing the follow-up research. Man, oh man, the two papers will be linked in the description. They're also on screen. And by the way, if you're just listening to this, I recommend you watch it on YouTube because at any point when
Something is brought up. For instance, Michael, you referenced a result of graph that's on screen at the same time. So what I want to know is what's next for you all? Both personally, I understand that some of you are getting your PhD soon. So both personally and then also as follow up research with regard to what we're talking about today. Gazem, I'll start with you and then Angela and then Michael, I want to know about what's next for you, especially the mind everywhere project. So Gazem, please.
Sure. I actually just defended my PhD a few weeks ago. Congratulations. Thank you. I'm in a transition phase and what's next is to trying to understand the capabilities of native capabilities of these end robots, as well as starting to engineer them for specific outcomes based on sort of what target morphological and functional
How does it feel to be done your PhD or at least to almost be done completely? Thank you. It's still surreal. It's been more than five years that I've been working on this project and I've
It's been sort of the because when I first started I was I really wanted to work with Mike because I knew that he was really interested in cracking the morphogenetic code and understanding
You know, like for me, understanding how the hell the, you know, the natural architectures build themselves, like, how is it exactly happening? What are the control sort of parameters? So that's why I was really interested in conducting the research research in this lab and learning about all these new sort of non genetic approaches to doing that was
new for me. So, um, and I've just been really surprised by the level of, uh, change we were able to induce in these cells. I mean, this is really radical going from a single cell to something as complex as an answer. But, uh, so far what I had done, like in my masters was, uh, the more sort of genetic approach. Okay. If I put in this gene, this will happen that gene. So those were really small changes.
Angela, what's next for you?
Yeah, so I will hopefully be following to them and graduating fairly soon. Now that this paper is done and it's out for the world to see. It's kind of been a long process. I started it my rotation like six years ago and it survived a pandemic and everything. So it feels good to at least be close to a finish line.
And then, yeah, I hope to be able, so afterwards I'll continue in Mike's lab for a little bit after just to wrap things up and hopefully continue to pursue this idea of looking at how genomics aren't really responsible for everything of an organism.
Yeah, we're hopefully going to continue to look into this project, look at different avenues of communication, look at some of these things that we didn't really have time to look at with this first paper, but hopefully follow up experiments are going to be very exciting. And by the way, what surprised you most about this research, like during it or from the reception? I understand that, hey, it's being released today, so it wouldn't be reception from the public, but you get the idea. Yeah, so I think
For me, this whole idea of how our genome doesn't encode for our anatomy, right? So if you think about a tadpole and a frog, they have the same genome, but their morphologies are completely different. And so kind of diving into this little hole of like looking into, okay,
Given that, you know, the whole field is focused on genomics and genetics. But what else is there that can contribute to our morphology? So what other instructive information can we receive or that we can give in order to form to form properly? Angela, the frog that you used, it's a certain species called the Xenopus laevis. And for people who don't know, that was the embryos. The embryos we keep talking about are from that species. Now, that's been a staple of Michael's work for at least a decade.
Why? What separates this frog, this species, from other frogs and other species? What useful properties does it have? Yeah, so I think Xenophys laevis is one of the reasons that we use it is because the genome is well studied. So there's something called ZenBase, which has all the genomic information. It has expression. It has a lot of staining that people are interested in. So having that database, I think, is very useful.
Also, another perk of Zenopost is that the embryos develop outside of the organism and you can see it. The embryos are quite large. You can see them with the naked eye. And then with the help of a microscope, you can track them and you can watch them develop from a single cell to a two cell to a four cell stage. So it's an ease almost.
Being able to just watch it and having a database to compare it to, I think, makes it a very strong model organism. And I mean, there's going to be perks and cons to every organism, I think. Obviously, I have a preference for those, but yeah. Angela, something else is that with the RNA sequencing, you uncovered that there were transcriptional changes associated with the SEMA effect.
So were you able to identify any genes that responded differently in the small groups versus the large groups?
Yeah, so in our paper, we do an RNA-seq. So we look at the RNA of these embryos. And essentially what I do is I have a large group of 300 embryos and a small group of 100 embryos. So we're comparing the group sizes of both treated and untreated individuals. For the purpose of the paper, we focus on the two group sizes that are being treated. And we see is that there's a total of like 16 genes that are up and down regulated.
And there are specific types of pathways that are being used in the small groups and specific types of pathways being used in the large groups. So that almost indicates that if you're in a large group, you're coping with the stress in one way versus when you're in a smaller group, you're coping with the stress in a different way. Transcriptional changes mean that the gene is then expressed differently or what? What does it mean?
Yeah, so with RNAseq, we're looking at how many copies of the gene there is essentially. So looking at the profile of the RNA. So what RNA is present and how much of it is present. Great. Michael, what's next for you? Yeah, well, you know, the first thing I want to say is just how incredibly proud I am of both of Gizem and Angela and the rest of the teams because there were lots of other people, collaborators, undergrads and other postdocs and
but just like they've done such an amazing job pushing forward this project and everybody needs to understand it is really hard not only to it like innovating in science is hard and getting something to work is hard and and and both angeline gizem have been you know faced all kinds of issues to get this stuff to work but but then just just the idea of being a young person at that stage of the career and
i'm having something that's so different and kind of counter expectation it's a lot of pressure and it's a lot of it's a lot of responsibility and i just you know i think people should understand that like it's hard for anybody else listening to this you know who might be in that position might be considering going into it i don't know what what you all feel but to me it's as hard and and does.
As much pressure as it is it's the most fun you can have so you know that's just what you want everybody understand like it's it's it's not just it's not just standard incremental signs like this was like they did an amazing thing so i'm so that's the first thing i wanted to just say that you know i'm super super proud of you both.
Yeah. Let's see. Next. Well, obviously, we're going to continue with some of the stuff that we talked about. So the practical uses on the Anthrobots, the practical uses of them, the ability to try to understand how to program them towards new and controllable shapes and functions. And also, because in my group, we're very interested in basal cognition in general and diverse intelligence more broadly,
Really understanding what are the properties of the the proto cognitive properties of this new model system what what do they know how to do do they can they form memories can they learn from their environment can they do they have preferences about different different lifestyles are different outcomes that that can be fall them and so on and none of this is known we've made no claims yet as to their level on the sort of you know the spectrum of
of cognition. We have no idea where they fit, but the one thing I know for sure is that we don't guess. We have to do experiments and find out. So we're going to find out. So those are the anthropo stuff.
Um, on the, on the SEMA front, uh, clearly trying to, of course, uh, better understand how it works and how the information is encoded that goes between the, um, the, the, uh, embryos and really to make models of this as a collective intelligence. So we already know we've already, we've already made models that, and other people have to that treat individual cells within the growing embryo as a collective intelligence. But it turns out there are multiple levels to this. Maybe not surprising in the end that, um, that maybe the group is
also has the ability to compute the path through anatomical space. So really understanding this, and then on the biomedical side, learning to induce it at will. Because as you said, you know, in the human case, where you have a patient, not necessarily an embryo, but a patient, they're not part of a 500, you know, a connected 500 individual cohort. But could we could we fake it? Could we you know, is there a way to is there a way to give that information that they would have had if they had been?
So those are the kinds of things. And then again, to really understand how the, you know, we think individual cells are using the bioelectric networks as a cognitive medium. What are embryos using such that the group has the ability to solve certain problems? Is it a field of ATP and calcium and who knows what else, right? So these are all things that we're going to be tackling next. I think it's going to be very exciting. And the Mind Everywhere project?
Yeah, I mean, both of these projects are key elements of that.
Yeah, so that TAME paper was version 1.0, so I'm working on the next one. It's going to be a little while still, but we've learned a lot since that first one. We're getting, I think, better conceptual foundations, better computer models of what it takes to scale cognition. This idea of the cognitive glue, what is it that enables competent individuals like cells or even molecular networks
To scale up into larger iq individuals that solve problems in other spaces bigger bigger boundaries of the self and so on so we have been to these two projects are part of it there are many other projects that contribute as well so yeah moving moving along. Are you planning on writing a book michael to introduce your studies to the audience.
Yeah, so I'm committed to one book. Oney Pagan and I are writing a book on bioelectricity. We have a contract with Norton. It's due towards the end of 2024. So that's something that we're definitely doing. So that's a basic book on bioelectricity and it's kind of import for medicine and so on.
In my head i sort of rattling around have two other books one that's on this basil cognition kind of topic that sort of talks about the scaling of intelligence from very minimal systems all the way up and then maybe one after that we'll see. First i can't even imagine finishing this first one that i'm doing so i need to get past that.
Thank you. Thank you all for coming. Thanks for having us. Yeah. Thank you for having us. Thank you so much. Yeah. Thanks. Thanks for having us. Yeah. Great discussion. Thank you. Great questions. I was surprised when you said you were not a biologist. I was sure you must have had like some sort of maybe at least like undergrad or master's training in biology because the questions are really good. Thank you. Thank you. Spot on.
The podcast is now concluded. Thank you for watching. If you haven't subscribed or clicked that like button, now would be a great time to do so as each subscribe and like helps YouTube push this content to more people. You should also know that there's a remarkably active discord and subreddit for theories of everything where people explicate toes, disagree respectfully about theories and build as a community our own toes.
Links to both are in the description. Also, I recently found out that external links count plenty toward the algorithm, which means that when you share on Twitter, on Facebook, on Reddit, etc., it shows YouTube that people are talking about this outside of YouTube, which in turn greatly aids the distribution on YouTube as well.
Last but not least, you should know that this podcast is on iTunes, it's on Spotify, it's on every one of the audio platforms. Just type in theories of everything and you'll find it. Often I gain from re-watching lectures and podcasts and I read that in the comments. Hey, toll listeners also gain from replaying. So how about instead re-listening on those platforms? iTunes, Spotify, Google Podcasts, whichever podcast catcher you use.
If you'd like to support more conversations like this, then do consider visiting patreon.com slash curtjymungle and donating with whatever you like. Again, it's support from the sponsors and you that allow me to work on Toe full-time. You get early access to ad-free audio episodes there as well. For instance, this episode was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough.
▶ View Full JSON Data (Word-Level Timestamps)
{
"source": "transcribe.metaboat.io",
"workspace_id": "AXs1igz",
"job_seq": 6590,
"audio_duration_seconds": 5789.8,
"completed_at": "2025-12-01T00:23:30Z",
"segments": [
{
"end_time": 20.896,
"index": 0,
"start_time": 0.009,
"text": " The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze."
},
{
"end_time": 36.067,
"index": 1,
"start_time": 20.896,
"text": " Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates."
},
{
"end_time": 64.548,
"index": 2,
"start_time": 36.34,
"text": " Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a total listener, you get a special discount. Head over to economist.com slash to subscribe. That's economist.com slash to for your discount."
},
{
"end_time": 91.101,
"index": 3,
"start_time": 66.22,
"text": " We have a world exclusive today. By the time this airs, it will have just been announced that two of your papers with you, Michael Levin, have been published with your co-authors, which are here, Ghazem Gumuskaya and Angela Tang. So Michael, I'd like you to explain what these papers are. And then Ghazem and Angela, I'd like you to explain respectively the significance of those papers. So Michael, please. Yeah."
},
{
"end_time": 109.258,
"index": 4,
"start_time": 91.664,
"text": " Sure. Yeah, it's really interesting to me that these are all coming out at the same time because there's a kind of a fundamental similarity here. One, so Gizems has to do with something we call Anthrobots, which are these, it's a new biorobotics platform. They're made of"
},
{
"end_time": 123.66,
"index": 5,
"start_time": 109.855,
"text": " Human human cells and other kind of a self motile construct that has all kinds of possibilities for for medicine and for telling us about evolution and development and so on and so we'll talk about that."
},
{
"end_time": 151.749,
"index": 6,
"start_time": 124.019,
"text": " Angela's paper is about the ability of embryos to communicate with each other and to help each other resist various teratogenic influences. So things that would normally cause developmental defects, it turns out that in a group, embryos are able to work together to better overcome those kinds of influences. And so what these things have in common is really our attempts to understand where biological information comes from."
},
{
"end_time": 170.162,
"index": 7,
"start_time": 152.176,
"text": " Because in the one case, so in the anthropoid case, we have this coherent construct that is using a completely wild type human genome to make a form and function that are not the typical things you see in a normal human target morphology. In the case of Angela's paper,"
},
{
"end_time": 197.346,
"index": 8,
"start_time": 170.674,
"text": " What you see is that the robustness of development is not just the property of a single embryo with its own genome, but actually a group phenomenon where a number of, in fact, the larger the group, the better, large collections of embryos are able to together solve the problem of morphogenesis better than individuals or small groups. And so in both cases, it's this really interesting dynamic of where biological control information comes from."
},
{
"end_time": 202.756,
"index": 9,
"start_time": 198.49,
"text": " All right, great. Gazam, if you don't mind, please, what is the significance of that?"
},
{
"end_time": 229.087,
"index": 10,
"start_time": 204.343,
"text": " Sure, so like at a high level what is really exciting about this work I think is this paradigm shift in our thinking about biology and nature. We tend to think about nature as this thing sitting outside waiting to be investigated by scientists and written books about but in reality what I came to realize as a designer sort of eight years ago that nature can actually be a design medium"
},
{
"end_time": 249.531,
"index": 11,
"start_time": 229.087,
"text": " through this new field called synthetic biology, which recognizes that biological structures have embodied computational frameworks that determine their architecture and their function. So I became really interested personally in trying to get new architectures to build themselves with"
},
{
"end_time": 277.125,
"index": 12,
"start_time": 249.531,
"text": " sort of composed of biological tissues. And why biology? Because biology has a lot of properties that the sort of traditional sort of construction methods don't. Well, properties like carbon negativity, it can take carbon from the environment in building itself. This idea of self-construction, you know, starting from a single seed and becoming a fully fleshed structure, healing,"
},
{
"end_time": 296.732,
"index": 13,
"start_time": 277.346,
"text": " Ability to take information from the environment, process that and give a output. Regeneration, so a lot of different properties that we don't see in human designed and built frameworks. So specifically for Anthrobots, why it's really exciting."
},
{
"end_time": 322.432,
"index": 14,
"start_time": 297.005,
"text": " Traditionally, what's been done in the field of synthetic morphogenesis, which is this idea of creating new biological structures, is to try to edit the DNA, the genetic code, to create new patterns with cells. But that has been really limited in the type of complexity and the scale that the structures can be generated."
},
{
"end_time": 347.585,
"index": 15,
"start_time": 322.978,
"text": " Antrobots is looking into biology and the morphogenetic code as a sort of more layered enterprise. So it's not only genetics, but also epigenetics, which consists of bioelectricity, as we'll discuss in the episode, as well as other sort of like methylation type of changes in the genome."
},
{
"end_time": 359.974,
"index": 16,
"start_time": 348.217,
"text": " So with Entrobots, we wanted to leverage that. Can we start with one structure that is known to build something in the body? In this case, human trachea."
},
{
"end_time": 389.872,
"index": 17,
"start_time": 360.367,
"text": " But then can be by giving it environmental inputs, can we get it to create a completely novel structure without touching the genome? So we still don't know what epigenetic factors and of which there are a lot of different possibilities that exactly give rise to the answer about. So the process is something we're investigating. But what we've seen is that we were able to get these cells to create a new architecture that we had designed a priori."
},
{
"end_time": 419.497,
"index": 18,
"start_time": 389.872,
"text": " Um, so yeah, like in summary, there are the, the, the first sort of fully cellular living self-constructing biological robots and they build themselves from single cells and, um, they don't have sort of electrical wiring or mechanical parts as a traditional sort of robot would have. Um, but they are still sort of programmed anatomies in that sense we call them."
},
{
"end_time": 446.459,
"index": 19,
"start_time": 419.838,
"text": " biobots. That's where the sort of robotics aspect is coming from. And also like the bots we know of, they can do useful work. So we showed that they are able to traverse wounded human neuronal tissues and induce repair in those wounds in the course of like three days. So in a nutshell, that's the aspiration and the summary."
},
{
"end_time": 465.145,
"index": 20,
"start_time": 446.92,
"text": " Okay, yeah, I mean that's actually that's a really important point because when we made the Zenobots, some people thought that this might be some sort of result that's specific to amphibians and we know amphibians are kind of plastic with this embryonic tissues or tend to be plastic. And so there's this temptation to think of this as a"
},
{
"end_time": 491.374,
"index": 21,
"start_time": 465.333,
"text": " Angela, your work on embryos assisting one another, can you please explain why is that consequential?"
},
{
"end_time": 510.469,
"index": 22,
"start_time": 491.578,
"text": " I think the significance of this work is that it's taking a step in addressing this knowledge gap that exists within the field with respect to these instructive lateral interactions that can be used to correct any developmental defects at a level that occurs above just the cells, tissues, DNA, or organs."
},
{
"end_time": 534.531,
"index": 23,
"start_time": 510.862,
"text": " So we're looking at whole organisms and how they can interact with each other and aid each other. And all of this, I guess, is important because if we're able to understand and harness these instructive cues, then we can use cells and tell them what to build and when to stop building and use them to fix defects and disease. So there's kind of this application for medicine potentially in the future."
},
{
"end_time": 561.476,
"index": 24,
"start_time": 535.162,
"text": " Yeah, and you said that when you had first brought it up to Michael, you thought it was crazy. And you weren't sure what he was going to say about it. So why is it so ludicrous? I guess in the field like biology, everyone studies the genome. So the idea that our genome is going to predict what we're going to look like in our future. So our genome leads to our phenotype. And that's kind of the underlying one of the underlying principles of biology."
},
{
"end_time": 592.142,
"index": 25,
"start_time": 562.193,
"text": " My work is kind of looking at, okay, well, what about things outside of the genome and kind of contradicting that a little bit because it's looking at now our genomes are not being changed, but we're able to change all these other things without affecting the genome in any way. So I thought that this is crazy. Like I'm not doing anything to the genome, but all of these different defects are forming. And the only thing that's different is the number of embryos in a cohort. Um, so I think to me that was very shocking."
},
{
"end_time": 611.971,
"index": 26,
"start_time": 592.619,
"text": " Okay, so Gazem, I would like you to explain this simply. Suppose we were to isolate the cell from a human lung and we put it in a petri dish. Okay, does this cell exhibit the characteristics that are consistent with an anthropobot? Now, I imagine not. So what modifications are necessary for this cell to attain the anthropobotic status?"
},
{
"end_time": 630.316,
"index": 27,
"start_time": 612.415,
"text": " Sure, so our starting question was how do we get to this target morphology, which is this multicellular spheroid with cilia on the surface and where that target morphology comes from is the xenobats, except that we wanted to change how we get there."
},
{
"end_time": 648.643,
"index": 28,
"start_time": 630.316,
"text": " So we talked about self-construction. We wanted a single cell to give rise to that target architecture. And then we also talked about necessity to use human cells as well as necessity to use adult cells. So our starting material for that reason could have been a"
},
{
"end_time": 674.582,
"index": 29,
"start_time": 648.951,
"text": " Progenitor cell from any one of the saliated epithelial tissues in the body. So not just the airway. You also have the saliated epithelia in the oviductal region or in the brain. We started with human long epithelium because the accessibility of the cells due to the research that's out there. There is a lot of sort of material available for long research due to cancer and other"
},
{
"end_time": 690.503,
"index": 30,
"start_time": 675.179,
"text": " other diseases. So for that reason, our starting material was the human airway epithelium progenitor cells. So these cells are already committed to becoming any one of the cell types in the human airway. So they could be"
},
{
"end_time": 720.247,
"index": 31,
"start_time": 690.503,
"text": " secretory cells or ciliated cells. And what was already known in the literature is that if you culture these cells in a rich matrix, extracellular material rich matrix, they will form the spheroid and a single cell will give rise to that spheroid. The problem was that in that configuration, the cilia would be looking inside to the lumen because the goal of those types of"
},
{
"end_time": 739.36,
"index": 32,
"start_time": 720.247,
"text": " culture methods which touches the field of organoids is to recapitulate the native tissue architecture. So when you look at the human lung to trachea you have cilia inside and you have these sort of basal cells outside. So the protocol that was already developed in the literature was mimicking that."
},
{
"end_time": 754.974,
"index": 33,
"start_time": 739.36,
"text": " But for us, for our purposes, to get the cilia to look outside so they can be motile like the xenobats, we had to come up with a way to get them to flip inside out. So that's what we've accomplished and we've tried a lot of different approaches for this."
},
{
"end_time": 784.275,
"index": 34,
"start_time": 754.974,
"text": " and what ended up working for us was to sort of trick those ciliated cells that develop and look inside to migrate outward. So how we've accomplished that was by changing two things, by changing the growth phase. So these spheroids are again embedded in a matrix and that's what's making the cilia to look inside. So if you remove that matrix and instead supply a low phase liquid based environment,"
},
{
"end_time": 813.848,
"index": 35,
"start_time": 784.275,
"text": " So human cells are already known for being plastic in a certain manner?"
},
{
"end_time": 830.589,
"index": 36,
"start_time": 814.616,
"text": " So, all tissues like cells that have the progenitor status are known to be plastic into a degree."
},
{
"end_time": 850.265,
"index": 37,
"start_time": 830.947,
"text": " I mean the plasticity drops as you move in. Mike and Angela can also speak to this. What's known is that it drops from the embryonic state to the sort of adult and elderly state, but it's never sort of a binary thing. It is for some organisms, but for humans."
},
{
"end_time": 870.333,
"index": 38,
"start_time": 850.265,
"text": " It's not a binary thing because there are a lot of protocols out there that use the progenitor cells, which are semi-stem-like and semi-committed cells that form different structures in vitro. It's just a question of how do you get it to form the structure of interest for you."
},
{
"end_time": 895.828,
"index": 39,
"start_time": 870.333,
"text": " And the method that's used in the literature a lot is to use genetic circuits to do this, to get that cell to execute certain morphogenic functions. But using that method, getting to something as complex as a ciliated spheroid would not be possible. And that's why we wanted to play into the native plasticity of these cells. Now, before we get to Angela,"
},
{
"end_time": 925.896,
"index": 40,
"start_time": 896.186,
"text": " Michael, I would like you to talk about the correlation between the structure and the motility. Yeah, so when you look at any kind of animal, you always wonder what the relationship is between the structure and the function. In other words, typically in a standard kind of creature or model system that we study, there's a long history of evolutionary selection that makes sure that the structure that it has is actually"
},
{
"end_time": 943.507,
"index": 41,
"start_time": 926.22,
"text": " Perfectly suited for the functionality that that you wanted to have and so this means moving around in the three dimensional world but also physiological kinds of actions and so on and so in this case what we're dealing with here is something super interesting it's a it's a creature that although it has a perfectly standard human genome."
},
{
"end_time": 962.312,
"index": 42,
"start_time": 943.763,
"text": " is not something that's ever been specifically under selection before to be a good anthropobot. There's never been any anthropobots. And so what we're looking at is a new functionality, a new set of behaviors and a new structure that underlies it. But we didn't know ahead of time what that relationship was going to be."
},
{
"end_time": 984.445,
"index": 43,
"start_time": 962.756,
"text": " I'm in the team did this really amazing job characterizing two things first of all characterizing all the different shapes and that they come in that these enter what's coming and there are a few discrete forms and so this is there they're actually I'm called a morphe types because they're actually discrete types of shape that you have in terms of the distribution of the Celia and and you know what the overall shape is and so on"
},
{
"end_time": 996.681,
"index": 44,
"start_time": 984.855,
"text": " And then we characterize the behaviors and they have a variety of different paths that they can take through the medium. And that looks very much like there's a there's a notion in behavior science of"
},
{
"end_time": 1017.432,
"index": 45,
"start_time": 997.244,
"text": " of an ethogram, which is basically just a diagram of the different behaviors that an animal can do and which of those behaviors can statistically likely to follow which other behaviors, you know, so you'll like for some sort of fish, you'll do like there's this particular kind of mating behavior that does and so if it if it darts a certain way, then next it'll do something else and that's a 50% chance of doing something else."
},
{
"end_time": 1028.677,
"index": 46,
"start_time": 1017.432,
"text": " So you can you can start to build up this diagram and so what we did was to study these these anthro was as if they were a new kind of creature because in an important sense they were we didn't know what all the behaviors were going to be."
},
{
"end_time": 1058.985,
"index": 47,
"start_time": 1029.121,
"text": " We didn't know what the relationship with the with the shape was going to be. And so they because then worked out how the different shapes produce the different motions and really what what the behaviors are and how they how they tend to follow each other, you know, one after the other. And I really think that this is also just the beginning, because, you know, this is the first paper on anthropos. This is a very basic characterization. We need to do all kinds of interesting things now to see how they react to their environment, how they"
},
{
"end_time": 1078.575,
"index": 48,
"start_time": 1059.189,
"text": " Were you surprised that they split into these distinct forms?"
},
{
"end_time": 1096.067,
"index": 49,
"start_time": 1079.087,
"text": " There were many possibilities. So one possibility is that they would have no behavior at all. Another possibility would be that they would only have one type of behavior. Another possibility would be that there would be multiple behaviors, but any given bot could only do one of those behaviors. So you really don't know what to expect."
},
{
"end_time": 1125.742,
"index": 50,
"start_time": 1096.254,
"text": " and the fact that these behaviors are discrete that you can characterize and this is an unbiased statistical analysis showed us that there were in fact classes of behavior. The fact that they fall into these classes is super interesting and I look forward to discovering what other classes of behavior there are and then really fleshing out the ethology of this new form to understand why it changes behaviors when it does."
},
{
"end_time": 1138.865,
"index": 51,
"start_time": 1126.032,
"text": " Angela, there's a term in your paper called cross embryo morphogenetic assistant. So it's this mouthful, which is abbreviated to SEMA. So did you point SEMA? And what is it? What does it mean? Yeah, so"
},
{
"end_time": 1163.575,
"index": 52,
"start_time": 1139.548,
"text": " Mike and I wanted something that was catchy and very obvious. So the long form of it, what you just said, is kind of very obvious. It's how embryos can communicate with each other and assist each other. And then just to make it not so much of a mouthful, we decided that we would just call it SEMA for short. And then the whole phenomenon is this idea that"
},
{
"end_time": 1188.046,
"index": 53,
"start_time": 1164.053,
"text": " these larger groups"
},
{
"end_time": 1216.049,
"index": 54,
"start_time": 1188.046,
"text": " this is how I handled it, and now you too can survive the stressor. So what we see is that as we increase our cohort size, we see an increase in survival depending on the tradigen that we're using, but also as our group size increase, we also see a decrease in the types of certain types of defects that we know. So this is both increasing survival and decreasing the frequency of defects."
},
{
"end_time": 1245.913,
"index": 55,
"start_time": 1217.142,
"text": " Okay, so I'd like to attempt to explain this extremely elementary. So people have embryos. Embryos are like the seeded babies, single cell babies or extremely small babies. Okay, when you have just one of them and you give it something called a teratogen, which is something that disrupts the development of one of these little guys, so they grow up abnormal. It turns out that when you have multiple embryos next to one another,"
},
{
"end_time": 1271.084,
"index": 56,
"start_time": 1246.51,
"text": " that if you give it a teratogen, it's more resilient to these perturbations. It's more stable. So it's as if they're communicating somehow these embryos are communicating and saying, Hey, don't mess up your arm like this. Here's how to make a correct arm. Is that approximately correct? Yeah, that's, um, I think the, how we're envisioning this happening kind of"
},
{
"end_time": 1292.824,
"index": 57,
"start_time": 1271.783,
"text": " Michael, there's this concept called group wisdom. Is there some variation of that?"
},
{
"end_time": 1319.582,
"index": 58,
"start_time": 1293.336,
"text": " Yeah, you know, we know there's this phenomenon called the wisdom of the crowds. And it's there used to be this old thing where they'd be in certain in stores, it'd be this giant jar of jelly beans, right. And people would try to guess how many jelly beans. And so the result there is that every individual person is quite often nobody has any idea how many jelly beans are there. But if you just average all the guesses, it turns out that the crowd is spot on."
},
{
"end_time": 1349.804,
"index": 59,
"start_time": 1319.957,
"text": " And so this phenomenon has been known for a long time. And our initial hypothesis was that something like that was happening, that basically every embryo was contributing some amount of information, and together they were able to have a much more steady view of what a correct embryo was supposed to be. And remember, these are frog embryos, so they start out as frog eggs, and they're about a millimeter in diameter, and there's many of them developing in a petri dish."
},
{
"end_time": 1369.804,
"index": 60,
"start_time": 1350.094,
"text": " So originally that's the thought we had, but then Angela found a very interesting piece of data which suggests that the real story is more complex and more interesting, which is that if you were to compose a group of some embryos that were exposed to teratogens and some embryos that were never exposed,"
},
{
"end_time": 1396.425,
"index": 61,
"start_time": 1370.162,
"text": " The simple model would suggest that it would actually work quite well because the animals that were never exposed to the teratogens, well, they have a very good idea of how to make their body and their organs and they should do a great job of informing the others of any information that they might be missing in terms of that teratogen. But it turns out that actually that doesn't work. The optimal effect is when everybody was affected by the teratogen and it's the"
},
{
"end_time": 1426.391,
"index": 62,
"start_time": 1396.8,
"text": " It's the result of having confronted this, this influence and having overcome it. That is actually what's necessary. Everyone has to have seen it in order for the group to know what to do, which is, which is super interesting. And it means that it's, it's a much more active process. It isn't just that, you know, I know what I know, and I'm just going to spread that information and everybody else can make use of it. No, it's actually that most likely what's happening is that that information is being derived on the fly as embryos are encountering"
},
{
"end_time": 1451.084,
"index": 63,
"start_time": 1426.647,
"text": " these external threats, they are deriving signals that are important for them to reinforce the normal path that they take through this space of possible configurations and they're sharing that information with each other. Now Michael, is it less harmful because you're giving the same amount of teratogen to two embryos that you would to one or are you titrating proportionately? So you'd give 3x if there are three embryos?"
},
{
"end_time": 1474.599,
"index": 64,
"start_time": 1452.824,
"text": " That's a great question. Of course, we titrated everything. We scale up the amount of teratogen based on how many embryos there are. Everybody's still getting the same amount of influence. Angela, in the paper, there's the term inter embryonic communication and inter embryonic interaction. What would be the difference between those two?"
},
{
"end_time": 1490.623,
"index": 65,
"start_time": 1475.64,
"text": " Yeah, so I'm thinking inter-embryo communication is, in our paper, we talk about like a potential molecule calcium and ATP that can be used as a communicator. So it's almost like a little message that's being sent between embryos."
},
{
"end_time": 1518.046,
"index": 66,
"start_time": 1491.254,
"text": " Inter embryo interactions is when these embryos are able to grow together. So now if you have an N of one, you may still be sending out messages, but you have no other embryos to interact with. Versus in these larger groups, we have both the message and people to receive the message. You also mentioned that it doesn't rely on genetic homogeneity. So how much of a variation is tolerable?"
},
{
"end_time": 1544.667,
"index": 67,
"start_time": 1519.684,
"text": " Yeah, so within an individual cohort, we have, I don't know, like, they're not all the same, they're not all homogeneous. And none of that's not really important to us either. So there's a part of our paper in which we look at embryos or embryos from two different lineages. So we have wild type and albino embryos. So they have different phenotypes and different genomes."
},
{
"end_time": 1572.824,
"index": 68,
"start_time": 1545.435,
"text": " And what we see is when we treat them both with stressor, there's no difference between the two. So that indicates that, you know, you don't need to have the same genome to be impacted the same way. Yeah. And can you explain what a wild type is? Because I know before reading this paper, I never encountered that term outside of Pokemon. And in Pokemon, wild type just means you find it in the wild. It doesn't belong to a trainer. I'm like, OK, what does a wild type cell mean?"
},
{
"end_time": 1586.357,
"index": 69,
"start_time": 1573.49,
"text": " That's pretty spot on. So wild type genotype or wild type anything really just means how this organism exists in nature. So it's unperturbed, there's no changes that we've done to it, no genetic modification."
},
{
"end_time": 1605.401,
"index": 70,
"start_time": 1586.92,
"text": " I see, but you could still have wild type. So it has to be some relative term because every cell is mutated. So it can't just be like a non mutated cell. It has to be with respect to something like its relevance in a scientific experiment. You can't just find some cells and say these are wild type. You have to say these are wild type relative to something else or no."
},
{
"end_time": 1630.64,
"index": 71,
"start_time": 1606.323,
"text": " Yeah, no, you're correct. The wild type of a given species just means that that is the standard genome that you will find out in nature. It has not been modified by the experiment. So it's just a genetics term. It means that it's not a mutant of some sort. It's just the natural. Now, having said that, of course, you're absolutely right. Even within the natural population, you're obviously going to get variability."
},
{
"end_time": 1654.087,
"index": 72,
"start_time": 1630.64,
"text": " so it's not and in fact with these frogs these are not an inbred population the way that you might have with with certain model systems these are this is an outbred population and so of course there are genetic differences even between individual frogs in our colony but wild type just means that there's no there's not been any specific mutation performed on them. Like in the lab by a scientist. Great great. Gazem"
},
{
"end_time": 1674.855,
"index": 73,
"start_time": 1654.497,
"text": " How do you think different cell types would affect the formation and the behavior of the biobots? And I just noticed I used the word biobot, okay, not anthropobot, because that's in the titles, but then in the paper. So I also want to know, why did you all title it biobots, but then also coined the term anthropobot? But that's a separate question. So how do you think different cells would affect the formation of these anthropobots slash biobots?"
},
{
"end_time": 1696.647,
"index": 74,
"start_time": 1676.049,
"text": " I mean I guess like just kind of taking a step back and establishing the terminology there. Biobot is not a term we've coined, it's something that has already been discussed in the literature. What that essentially refers to is programmable anatomy."
},
{
"end_time": 1719.48,
"index": 75,
"start_time": 1696.647,
"text": " I mean, that's sort of using biological cells, either completely biological, like in the case of entrobots or xenobots, or some sort of a hybrid between biological and a mechanical or chemical scaffold carrying those cells and providing mechanical support. So there could be a lot of different types of biobots and there have been."
},
{
"end_time": 1744.206,
"index": 76,
"start_time": 1721.169,
"text": " I guess the field really started flourishing around 2013. What we have been trying to do is to create bio-bats that are fully cellular. The first example of that is xenobats. In creating xenobats, no external gel or mechanical or electrical substance was used."
},
{
"end_time": 1757.346,
"index": 77,
"start_time": 1744.206,
"text": " I'm so and robots are another example of that again fully biological and the name is coming from the human origin so by about the more general term and and robot are examples of them."
},
{
"end_time": 1781.203,
"index": 78,
"start_time": 1757.602,
"text": " Exactly. All fully cellular bio-bots, yes. And there could be more in the future that could be named differently. And so your original question was like how different types of cells could create different types of bio-bots. So it all depends on what you're trying to accomplish. No one bio-bot is better reversed than the other one. It just depends on what your goals are."
},
{
"end_time": 1810.094,
"index": 79,
"start_time": 1781.203,
"text": " Um, for Xenobots, for example, like, uh, the goal was to create something that could, um, or rather I should say if your goal is to create something that could survive outside, like in the wild, um, Xenobots is a better way to go than Anthrobots because Anthrobots are, um, they, because they're derived from human cells, they have sterility requirements. Um, and they are better for say medicine. Uh, whereas like Anthrobots,"
},
{
"end_time": 1838.217,
"index": 80,
"start_time": 1810.265,
"text": " I'm sorry to interrupt. They have what kind of requirement? So sterility, which means that they cannot come across, they cannot interface with pathogens of any kind, bacteria or viruses or fungi. This is a general requirement for mammalian cell culture. When we do cell culture for any purpose, making biobots or investigating diseases,"
},
{
"end_time": 1856.544,
"index": 81,
"start_time": 1838.217,
"text": " There are strict sterility requirements. We work inside these sort of hoods with laminar flow that push air out to prevent anything kind of coming and landing on the cells. So with these strict sterility requirements, for example, if you're trying to build something that will go into the"
},
{
"end_time": 1879.838,
"index": 82,
"start_time": 1856.544,
"text": " reverse and try to like detect the presence of a certain toxin and report back you would not want to use a mammalian cell or an anthropobite and like xenobot would be perfect for that. Or like another example if you're trying to build something where you want to mass fabricate then you would want to have a property like self-construction as I mentioned because then you have you know each"
},
{
"end_time": 1907.551,
"index": 83,
"start_time": 1880.179,
"text": " Biobot building itself, which means you can build like thousands of them in parallel without having to, you know, do anything like otherwise you would have to individually sculpt thousands. So it all depends on what your goals are. And that also impacts the decision of what cell type to use. So yeah, I mean, it's a lot of different cell types could be used for a lot of different Biobots. It all depends on what the design specifications are."
},
{
"end_time": 1931.664,
"index": 84,
"start_time": 1908.251,
"text": " We're just trying to bring the idea of design into biology, bring the two together and harnessing some of the properties that only biology has, such as healing after damage or self-replication or self-construction, and bring that into this process of fabrication of new structures."
},
{
"end_time": 1958.404,
"index": 85,
"start_time": 1931.664,
"text": " Could you please speak to the potential medical applications?"
},
{
"end_time": 1984.053,
"index": 86,
"start_time": 1959.838,
"text": " Yeah, so again, one of our design specifications was that we want to use this in medicine. And like Angela was talking about, having them be vile type was really important for us. So very not sort of inserting any foreign DNA that could when in turn deployed in the human body could have off target effects."
},
{
"end_time": 2006.135,
"index": 87,
"start_time": 1984.053,
"text": " So for that reason it was important for us to keep the human DNA wild type so we've tried from more than twenty different human donors and in every single time we're able to create an answer about and that's across a lot of different ages and you know genders and races and we've seen that this works with a lot you know diverse like human genomes."
},
{
"end_time": 2026.834,
"index": 88,
"start_time": 2006.937,
"text": " So what sort of more specifically in the medical field that we are hoping to accomplish is can we take a cell from a human donor? So currently, as we mentioned, we're taking these from the human airway epithelium, but down the line, you can envision taking a cell from a human patient's skin."
},
{
"end_time": 2040.913,
"index": 89,
"start_time": 2026.834,
"text": " And then turning it into a induced pluripotent stem cell, which in turn would basically revert the clock and have the ability to then be differentiated into all these different kinds of tissues in the human body, including the human lung."
},
{
"end_time": 2066.732,
"index": 90,
"start_time": 2041.34,
"text": " So in terms of application, that's what we're envisioning. That's not something we've shown in the paper, but that protocol has already been worked out in the literature. So starting with a human skin cell and then turning it into an into an anthropo that is geared towards a specific application based on what that patient might need. And then when we put that into the body, it is a synthetic construct. It's something that doesn't"
},
{
"end_time": 2094.411,
"index": 91,
"start_time": 2067.21,
"text": " Yeah, I was going to ask about if you all have tested biocompatibility or immunogenicity if you're already envisioning it for medical applications."
},
{
"end_time": 2117.671,
"index": 92,
"start_time": 2095.555,
"text": " Not yet our preliminary experiments have been with human cells but in vitro only. So next step for us would be ex vivo tissue so human cells extracted from humans and then after that it would be in vivo or proxies for in vivo experiments so that would be step three."
},
{
"end_time": 2133.37,
"index": 93,
"start_time": 2118.968,
"text": " Yeah, but the but the I mean, these cells are already coming from the they've already been inside the patient. So while while we haven't specifically tested the immunogenicity in vivo, the chances are very high that it's going to work. I mean, these are"
},
{
"end_time": 2149.497,
"index": 94,
"start_time": 2133.524,
"text": " These are the ideas personalized medicine. It's a bespoke kind of construct that's made of each patient's own cells. So it's likely fine. And I just I want to I just want to underscore the amazing the thing that just just blows my mind every time I think about it."
},
{
"end_time": 2174.428,
"index": 95,
"start_time": 2149.497,
"text": " You know that last figure in that paper is basically showing just one initial thing that we found that these guys can do, which is to help neurons heal across a scratch wound in two-dimensional culture. Just to think about that, the tracheal cells that are sitting in your body and they sort of sit there quietly for decades doing their thing and using their cilia to waft little particles and mucus and stuff out of your lungs."
},
{
"end_time": 2203.473,
"index": 96,
"start_time": 2174.428,
"text": " The fact that if if uh liberated from their environment and given a chance to kind of reboot their their their multicellularity they now have the ability to go around and repair defects in other types of cells. We would have never known that it's just amazing to me that that that they have that capacity and it makes me wonder what what else you know what other cells are sitting around the your body with capacities to to heal other components and to have other beneficial"
},
{
"end_time": 2229.906,
"index": 97,
"start_time": 2203.933,
"text": " Pro regenerative types of outcomes on different parts of the body like that. That idea of releasing the native healing potential of your own cells and letting them do new things that might be beneficial for the body. I think is incredibly powerful. I think we're just seeing the first glimpses of that here. Can you talk about how this fits into the larger framework of your work? Because as I heard Gazem say,"
},
{
"end_time": 2258.456,
"index": 98,
"start_time": 2230.384,
"text": " Take a skin cell and turn it into a pluripotent cell. It reminds me of our previous conversations. Yeah, yeah, there's a few. The kind of applications of these are in several different directions. On the one hand, we certainly want to use this for very specific practical purposes, so we think that once we gain a better understanding of their kind of native functions and a little bit better on the programming end,"
},
{
"end_time": 2288.097,
"index": 99,
"start_time": 2258.541,
"text": " We will be able to address all sorts of very specific conditions and we can sort of run down some of the early ideas that we have. But there's a bigger picture here, which is using this biorebotics platform as a kind of simplified model system in which to crack the morphogenetic code. Think about all of the problems of biomedicine, including birth defects, traumatic injury, or thus failing to heal from traumatic injury."
},
{
"end_time": 2312.824,
"index": 100,
"start_time": 2288.37,
"text": " cancer, degenerative disease, all of these things have one thing in common, which is that they would go away if we had the ability to tell groups of cells what to build, right? That's the major rate limiting step for regenerative medicine is that we do not understand how cellular collectives make decisions. We're pretty good on the hardware side for individual cells, right? So we know how cells differentiate, we know what the various"
},
{
"end_time": 2332.995,
"index": 101,
"start_time": 2313.387,
"text": " Lots of various genes do and how they interact with each other and so on but but this this idea of how do collections of cells make decisions that they're going to make a hand versus a foot versus something else and more importantly how we communicate our patterning goals to them that is if you want to build a new organ or you want to repair an existing organ or you want to make something that has never existed before"
},
{
"end_time": 2361.305,
"index": 102,
"start_time": 2333.285,
"text": " what information do you need to give to these cells and what interface can you use to get your goals across to the cellular collective? And that I think is critically important for unlocking the promise of regenerative medicine. And so that's what we're starting off here, because you really have to, before you can use all these fancy programming techniques, and that includes not just the traditional symbio that people are using, but also the stuff that we do in our lab, which is bioelectrical,"
},
{
"end_time": 2389.428,
"index": 103,
"start_time": 2361.305,
"text": " What are the baseline plasticities and competencies of these cells? What do they already know how to do and why? Why do they make decisions to take specific paths through anatomical space and build specific kinds of anatomies and so on? And so I think that's in the greater scheme of our lab's work, which is to understand how to communicate with the collective intelligence of cells. This is a very important"
},
{
"end_time": 2418.575,
"index": 104,
"start_time": 2389.787,
"text": " a model system in which we can now ask, okay, what kinds of stimuli, what kinds of information can we be giving to these cells to get them to build various things? Much like with the xenobots, you know, these first papers were all about characterizing their background kind of native competencies. We didn't engineer the heck out of them with new genes and all this stuff. We can and we probably will in the future. But step one is to understand how do collections of cells make decisions about what they're going to do? Michael, can you also"
},
{
"end_time": 2444.667,
"index": 105,
"start_time": 2419.241,
"text": " indicate again these are two different papers here that have an overarching theme but outline how are they distinct and how are they the same so one has to do with anthropobots or biobots and then the other has to do with this embryonic communication and the resilience so please yeah the the the common there are many common themes but one one important one is collective decision making so it's again this idea of so so in the case of the anthropobots it's a question of"
},
{
"end_time": 2471.101,
"index": 106,
"start_time": 2444.974,
"text": " understanding how groups of normal cells with normal human genome derived molecular hardware are going to decide to work together to make a specific new coherent construct with new behaviors, new functionalities, and so on. In the case of Angela and the cross embryo morphogenetic assistance, it's the idea that standard developmental biology studies how"
},
{
"end_time": 2495.862,
"index": 107,
"start_time": 2471.698,
"text": " cells cooperate to make a nice embryo. Well, it turns out that this actually works on a higher level as well. So groups of embryos also work together to complete morphogenesis. And in both of these cases, what we want to understand is where is the information? What is the collective intelligence of these cells? What kind of problems are they able to solve? So in the case of the Anthrobots, they're able to"
},
{
"end_time": 2515.606,
"index": 108,
"start_time": 2496.374,
"text": " uh they find themselves in the new environment in the new kind of a new scenario and they're able to put together a very coherent form that is able to to live for weeks and and then have certain functions and so on in the key in an angeles case what you're seeing is again a kind of collective problem solving but this time at the level of whole animals."
},
{
"end_time": 2535.845,
"index": 109,
"start_time": 2515.828,
"text": " So not down at the cellular level, which is standard developmental biology. It's kind of, you know, maybe this is the beginnings of a kind of a sort of hyperdevelopmental biology where what you're really trying to work out is the rules by which whole bodies communicate to better achieve that better solve the problem of of embryo genesis."
},
{
"end_time": 2563.626,
"index": 110,
"start_time": 2536.032,
"text": " because one of the things that, well, many people have studied in our lab focuses on in particular is biological intelligence in the sense of problem solving. That means when you're confronted with a new scenario that you haven't seen before, especially a new scenario that you haven't seen before, are you able to complete your goals in the case of development? Are you able to make the target morphology that you want to make a correct embryo or some other functional thing in the case of, in the case of anthropos?"
},
{
"end_time": 2591.681,
"index": 111,
"start_time": 2564.292,
"text": " And so that's what we're seeing in both of these projects. We're seeing new unexpected competencies at different levels, at the level of cells and then at the level of organisms, to do something helpful and coherent in novel circumstances. Angela, can you please talk to how this robustness increases with more embryos? So one embryo fails more often than if you have a collective and how you found that out?"
},
{
"end_time": 2617.09,
"index": 112,
"start_time": 2592.449,
"text": " Yeah, so whenever I do my treatment groups, so when I start stressing out these embryos, an N of one will almost never survive by itself. I probably did like 50 dishes of just one embryo each and never did I ever really have a survivor. So there was something about that, just the fact that it's getting the same amount of drug as my larger groups."
},
{
"end_time": 2631.988,
"index": 113,
"start_time": 2617.09,
"text": " And then as we kind of scale away from there so now I'm increasing my group sizes."
},
{
"end_time": 2649.053,
"index": 114,
"start_time": 2632.466,
"text": " If we're just looking at survival by itself, I see that survival starts increasing once you hit like, I think something around a group of 50, then you'll see an increase in survival. And then as you go up to like 100, you get like 80% survival. And then once you hit like a group of 300, you get almost everyone surviving."
},
{
"end_time": 2676.578,
"index": 115,
"start_time": 2649.053,
"text": " So the whole premise of this was I had been given a teratogen and someone told me, hey, this works in my hands, I want to try it out. And no matter what I did, I could not replicate that. And I could not figure out why because I had this person telling me exactly what they did. And it came down to the number of embryos that she was using versus the number of embryos that I was using. And so to me, I was like, that's crazy that, you know, everything else was held constant."
},
{
"end_time": 2695.435,
"index": 116,
"start_time": 2677.193,
"text": " But the only thing that was different between our two experiments was just purely the number of embryos that we were using. So you stumbled upon this. Yeah. Interesting. Interesting. Yeah. And I just, you know, I want to, I want to emphasize a couple of things here, which again are really, really striking to me."
},
{
"end_time": 2720.23,
"index": 117,
"start_time": 2695.674,
"text": " One is that what this means and what Angela just pointed out is that a lot of people when they report toxicity values for different drugs. So there are thousands and thousands of papers in the literature saying that here's this new potential toxin or a drug or something else that can end up in the environment. And there are these tests using frog and zebrafish embryos that attempt to quantify how disruptive it is to development."
},
{
"end_time": 2743.865,
"index": 118,
"start_time": 2720.469,
"text": " What we now know, and a lot of people don't control for actually the number of embryos that you had in your cohort. They just do a percentage and call it a day. So what this is telling us is that many, many studies in the literature are not actually reporting the raw"
},
{
"end_time": 2771.817,
"index": 119,
"start_time": 2744.258,
"text": " a danger value for these chemicals, they're reporting the corrected value after the group has been able to resist it, right? And so what you just heard was that if you do these things on single embryos, the actual teratogenicity is very high. It's extremely potent, but you start to be blinded to that effect the more embryos you put in, because what you're seeing is how dangerous it is after the embryos have had a good chance to correct for it."
},
{
"end_time": 2776.937,
"index": 120,
"start_time": 2772.244,
"text": " And so that's really important that now we know that when we"
},
{
"end_time": 2804.326,
"index": 121,
"start_time": 2777.619,
"text": " examine the potential of various interventions to cause developmental defects. We have to ask what's the, what's the raw effect size, right? What's the actual teratogenicity and then how well do the embryos do to correct for it? So that's kind of a very practical thing that, that, that we now know that I think is important. And the other thing that is really striking is that the standard story of developmental biology and where the information comes from for you to be able to build a normal embryo,"
},
{
"end_time": 2830.794,
"index": 122,
"start_time": 2804.514,
"text": " is basically supposedly just two things. It's your actual genome and then it's the maternal components that are in the egg. So basically your parental genome that are, you know, there's some stuff provided for you in the egg for the zygote and that's it. And the idea is that the genome has everything you need to complete development and development is rightly so described as a very robust process. Most of the time it goes correctly despite its incredible complexity."
},
{
"end_time": 2857.961,
"index": 123,
"start_time": 2831.101,
"text": " And what we're seeing here is that that purely vertical view, the idea is that that you could be one embryo, uh, sort of, uh, far away from anything else. You've got your own genome and that gives you everything you need to know. That story is clearly only partially true that yes, you've got, you've got the hardware that you need, but actually that, that hardware is not as robust as you think without other individuals around. So development in a sense is a group phenomenon. It's a, um,"
},
{
"end_time": 2882.125,
"index": 124,
"start_time": 2858.507,
"text": " That traditional robustness level of development is actually a collective, it's a collective property. It doesn't work as well in a purely vertical sense, being passed down from a genome to one organism. And does that contradict the current thinking in developmental biology? Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon"
},
{
"end_time": 2906.544,
"index": 125,
"start_time": 2886.527,
"text": " Jokes aside, Verizon has the most space to save on phones and plans where everyone in the family can choose their own plan and save. So bring in your bill to your local Miami Verizon store today and we'll give you a better deal."
},
{
"end_time": 2926.749,
"index": 126,
"start_time": 2909.343,
"text": " Well, in the sense that this kind of effect has not been described before. I mean, people have seen things like alley effects and in terms of groups of animals surviving in some environment better than individuals. So that has been noted, but it wasn't known why that happens."
},
{
"end_time": 2954.309,
"index": 127,
"start_time": 2927.005,
"text": " And I do think that it contradicts the emphasis on purely vertical transmission and the idea that one embryo has everything it needs in its own genome. I mean, these are otherwise fairly conventional mechanisms that we're studying. For example, one of the other amazing things about that paper is that they were able to, Angela and Megan were able to literally visualize waves of information"
},
{
"end_time": 2980.384,
"index": 128,
"start_time": 2954.684,
"text": " passing across embryos. We have videos of where you can actually see them. And in order to make it easier, what we did was exert a very specific event. So basically like a needle poke into one of the embryos. So that way, you know exactly when it starts, right? And then you can actually see using this calcium indicator fluorescent dye, you can actually see this wave of information passing through from the point of injury through the animal"
},
{
"end_time": 3001.732,
"index": 129,
"start_time": 2980.759,
"text": " Now Angela, this wave of information, what is the mechanism? What is its physical component? Is it an electrical field? Is it just calcium ions being thrown? Is it something else? Yeah, so our current hypothesis for"
},
{
"end_time": 3023.643,
"index": 130,
"start_time": 3002.244,
"text": " The mechanism behind all of this is that some sort of injury occurs and then the embryo that's receiving this injury elicits a calcium response. This calcium response then releases ATP into the media and then surrounding embryos are able to uptake this ATP and then elicit their own calcium response. So there's kind of a"
},
{
"end_time": 3053.66,
"index": 131,
"start_time": 3024.36,
"text": " Yeah, this reminds me of trees, some trees in some forests, when one gets infected, it sends a signal through the roots and the others start producing antibodies before they even receive the, it wouldn't be a Tratogen. In that case, it would be some virus or whatever, maybe, or fungus. So is this similar?"
},
{
"end_time": 3082.363,
"index": 132,
"start_time": 3054.189,
"text": " It is, and animals and plants both can signal to others when they're being preyed upon. So there are examples like this where some predator is munching on a leaf or something, then there are volatiles released where other plants in the environment can feel it. One interesting thing about this though is that we're dealing here with fairly complex processes. So it's not a binary yes or no, right? It's not a binary, am I being attacked or aren't I? It's sort of"
},
{
"end_time": 3105.657,
"index": 133,
"start_time": 3082.619,
"text": " Well, normally I would build a head of a certain shape and size, and now I'm unable to do that. And so passing information on how to build a tadpole head requires lots of information. And this is one of the big mysteries going forward, is how do you encode all that information in a single, you know, a signaling molecule that's passed?"
},
{
"end_time": 3128.148,
"index": 134,
"start_time": 3105.879,
"text": " Right, because it's not just the yes or no, you have to actually, I think you have to actually encode considerable amount of information to for the embryos to help each other. So whether that's happening through some sort of modulation of pulsing through the water, or it's got to be something other than just simple, like here's your ATP concentration, and that's if it's a single number, you're not going to encode head morphogenetic data in one number."
},
{
"end_time": 3156.084,
"index": 135,
"start_time": 3128.558,
"text": " And the other thing to mention is that these projects are also going to come together because one of the things that I would really love to see is how this shapes out in the anthropobot case. So one of the things that we will be doing in the future is looking to see, do the anthropobots communicate with each other? Do they communicate with the other tissues that they find themselves in an environment with? How general is this? Because obviously it begins in the frog model,"
},
{
"end_time": 3185.128,
"index": 136,
"start_time": 3156.084,
"text": " Of course, as Angela said earlier, ultimately, you might want to basically fake it for biomedical purposes. So whatever signal is allowing tissues and organs to form properly, in these high density groups, you would want to be able to induce that on demand in the patient. And so the next step leads through biological tissues and especially anthropos to see whether that kind of phenomenon is general and whether we can harness it for biomedicine."
},
{
"end_time": 3209.121,
"index": 137,
"start_time": 3185.623,
"text": " So what I see is if this wasn't a breakthrough enough, both of these papers, so one, the anthropobots and then some of these anthropobots, I believe heal other cells or other tissue. Is that correct? Okay. Yeah. And the number two, the inter embryo communication or the SEMA effect that you have where different tiny babies can tell other tiny babies like, hey,"
},
{
"end_time": 3237.005,
"index": 138,
"start_time": 3209.65,
"text": " Protect yourself and hey, let me help you with your morphogenesis. Not only that, but number three, there may be large discrepancies in the literature or misleading effects because you mentioned this word raw taro. Sorry, raw. Can you repeat it for me, please? Teratogenicity. Yeah, I'm not a biologist. So raw teratogenicity that someone may be throwing out some harmful chemicals to a single embryo."
},
{
"end_time": 3257.568,
"index": 139,
"start_time": 3237.312,
"text": " And then someone else may be testing it on 10 embryos, but then I need to be clear here. When you're saying that there's some discrepancy in the literature, are you saying that they're reporting the raw amount? So let's say 10 milligrams of some teratogen, but they test them in 10 embryos. Do they then divide that by 10 or are you saying that they don't do that because they don't even tell you the amount of embryos to begin with?"
},
{
"end_time": 3277.278,
"index": 140,
"start_time": 3259.104,
"text": " Well, what I'm saying is that because no one had known before that the number of embryos actually determines how effective your teratogen is going to be, it means that when they report, I mean, other than testing it on different size cohorts, there's no way to know. You can't simply divide it by the number of embryos."
},
{
"end_time": 3294.991,
"index": 141,
"start_time": 3277.654,
"text": " And so when somebody says a certain, and by the way, in Angela's work, this is important. It's not just about chemical teratogens. She also tested RNA, so mutant proteins. And so this is a much more general thing. This is not just about chemicals."
},
{
"end_time": 3321.203,
"index": 142,
"start_time": 3294.991,
"text": " And so anything, including potentially a wide range of genetic mutations or drugs or other kinds of interventions, you know, it could be, who knows, maybe it works for radiation, maybe works for temperature induced defects. We don't know. But the idea is that what you're seeing isn't the real effect. What you're seeing is the effect after it has been corrected by the group."
},
{
"end_time": 3342.705,
"index": 143,
"start_time": 3321.817,
"text": " And so we really need to be sensitive to that. We need to understand that that depending on the size of the group, you may be under reporting the actual disruptive power of this thing, because if the group had corrected it, there's a similar there's actually a really interesting similar phenomenon which affects evolution, which is that a lot of these model systems and animals have"
},
{
"end_time": 3367.449,
"index": 144,
"start_time": 3342.978,
"text": " ways during embryogenesis of ways of repairing certain defects. So this is for just as an example, and many people have published other examples, but in our in our work from years ago, if you scramble the craniofacial organs of a tadpole, so you know, the eyes on the side of the head, the jaws are off to the side, like everything's everything's scrambled, they will actually find their way back to the right locations, right? They have the ability to individual embryos have some ability to fix these things."
},
{
"end_time": 3376.203,
"index": 145,
"start_time": 3367.739,
"text": " And so what that means for evolution, just think about if when selection gets hold of that embryo and everything is in the correct place and it's a beautiful embryo."
},
{
"end_time": 3398.916,
"index": 146,
"start_time": 3376.51,
"text": " selection doesn't actually know was it beautiful because the genetics were amazing or was it beautiful because actually it started kind of a mess but it's really good at fixing things and so that that problem is not just for human scientists it's also for the evolutionary process itself that you're often not seeing the actual phenomenon what you're seeing is what's left over after the competent"
},
{
"end_time": 3423.029,
"index": 147,
"start_time": 3399.224,
"text": " material which is which is cells and tissues have had their say and we have been talking about this for for a long time in terms of individual cells but now in this work you're seeing that it's also a property of groups of embryos you know this ability to mask defects and to um uh and to really uh can kind of uh not let you see the full impact of what the disruption would have been."
},
{
"end_time": 3439.019,
"index": 148,
"start_time": 3423.677,
"text": " So what if someone says, okay, Angela, this is all nice and good, but how does this apply to our species where we have predominantly one embryo? So in other words, we're not all octomoms. So who cares, let alone 50 moms or 100 moms?"
},
{
"end_time": 3463.251,
"index": 149,
"start_time": 3440.299,
"text": " Yeah. So I guess for humans, we look at this phenomenon and it can occur post like birth. So things like looking at how skin to skin affects mother's newborns, right? Like that's a communication just in a different type of way, no longer like this chemical, teratogen, embryo to embryo stuff, but there's still this communication interaction between mother and child."
},
{
"end_time": 3490.862,
"index": 150,
"start_time": 3463.251,
"text": " Also, if we look at like other things such as emotionally, right, like humans talk to each other, we take care of each other. Talking is a type of communication. So we see that in the case of humans, it might not be a particular molecule that's being passed along, but it could be different forms of this interaction, whether that is contact or word based. I think that the phenomenon still holds in people."
},
{
"end_time": 3508.746,
"index": 151,
"start_time": 3491.834,
"text": " And by the way, you can see in the, in the, in the, uh, in the anthropocase, you can see a microcosm of that happening because these, these anthropos are actually helping to heal the neurons that the neural scratch that they come across. Right. And so, so again, it's this, it's this notion of, um,"
},
{
"end_time": 3538.695,
"index": 152,
"start_time": 3509.087,
"text": " Yeah, something I was going to ask is what are the boundaries associated with this morphogenetic assistance? So you've established the cellular and what about"
},
{
"end_time": 3549.718,
"index": 153,
"start_time": 3539.104,
"text": " Organ level or tissue level or cytoskeletal or what about us as people? Is there some analogous mechanism through which we as humans as people we're influencing one another right now?"
},
{
"end_time": 3580.589,
"index": 154,
"start_time": 3550.879,
"text": " Yeah, quite possibly. This business of having had to be exposed to the teratogen before you become helpful as part of the group, maybe there's a human analogy too. Mark Solms told me that to be a good psychoanalyst, you have to have been psychoanalyzed yourself and gone through that process. Maybe this is truly scale-free in the sense that you see it in cells and tissues and"
},
{
"end_time": 3606.476,
"index": 155,
"start_time": 3580.947,
"text": " We have a new project starting now where we're going to look at that, and we're going to look at what the limits are of this. So in the frogs and so on, we're going to look at how general this phenomenon is. And then the same thing on the anthropobot side,"
},
{
"end_time": 3625.708,
"index": 156,
"start_time": 3607.142,
"text": " We're working with a company called Astonishing Labs, which again is going to let us really go wider and try to understand, okay, so they heal neural scratches, what else? What else can they heal? And how much of that is innate and how much of that can we bring out and how much of it is programmable and how much of it is inducible and so on? Gazem?"
},
{
"end_time": 3652.261,
"index": 157,
"start_time": 3626.391,
"text": " I have some notes here that I wrote down from going through your paper. I have it written down here as capable of navigating and promoting repair, at least in cultured human neural cell sheets. And they're formed without genetic editing, which seems to highlight the morphogenetic plasticity of wild type cells. Okay, so something I want to know is in science fiction, there's this trope where you swallow a pill and it has nanobots inside and these little robots operate and heal you."
},
{
"end_time": 3679.923,
"index": 158,
"start_time": 3652.961,
"text": " So the deleterious effects are vanquished, at least ameliorated, maybe injuries are repaired because of the operations of these tiny machines. So do you see a future where your anthropobots can serve this purpose? Or in other words, were these movies talking about your work, Gazem, and they just didn't know it? Yeah, I mean, I think that it's been kind of kind of kind of eyes on a lot for kind of harnessing what"
},
{
"end_time": 3703.37,
"index": 159,
"start_time": 3680.435,
"text": " nature can do in context that are not expected or unconventional. So exploring human body, it's sort of as difficult as exploring the outer space. And we have spacecraft and we're doing a lot of efforts on that front. But I think that sort of brings"
},
{
"end_time": 3723.217,
"index": 160,
"start_time": 3703.865,
"text": " It really triggers the science fiction community to think about how that kind of exploration can be extended into other unknown frontiers like the human body. So yeah, I mean, we are really interested in seeing if the inter robots can be deployed in tissues that are otherwise inaccessible."
},
{
"end_time": 3745.691,
"index": 161,
"start_time": 3723.473,
"text": " to direct operation or accessible but invasive like surgery. One of the things we are really trying to understand here and we talked about the morphogenetic code is to see if we can leverage this computational ability that's inherent in biology. So what that"
},
{
"end_time": 3773.046,
"index": 162,
"start_time": 3746.067,
"text": " And that comes into layers, right? So the computational ability to build itself. So by taking information from the environment and processing that to grow its body, as well as once its morphogenesis is complete or at adult sort of stable state, to take information from the environment and process that and turn it into behavior."
},
{
"end_time": 3801.425,
"index": 163,
"start_time": 3773.541,
"text": " So I guess in the context of navigating the body, it would be more the latter, a evolved and well developed and sort of matured entrobot. Can it go through different tissues and collect information or trigger change? So that's there are a lot of different sort of avenues there one could follow. We have so far only looked at the healing in the context of"
},
{
"end_time": 3820.213,
"index": 164,
"start_time": 3801.886,
"text": " normal damage but yeah very hoping to look at other things like can they clear plague from the arteries by again been encountered with a like a piece of plague like adipose tissue can it detect that and then release some sort of a molecule that would help it"
},
{
"end_time": 3846.698,
"index": 165,
"start_time": 3820.213,
"text": " bulldoze. So combining the ability to release molecules as well as its physical thrust. So like bulldozing while releasing some sort of a molecule that would melt them, melt out the adipose tissue. Um, or can it, um, again, want to make it very clear that these are some things we're kind of aiming for and, uh, would require more research for us to demonstrate in the lobby have not yet."
},
{
"end_time": 3856.084,
"index": 166,
"start_time": 3846.988,
"text": " But these are pretending you're just keeping all the goods for yourself, pacing out the papers. I'm going to tell the Emerald Gates Foundation."
},
{
"end_time": 3884.667,
"index": 167,
"start_time": 3856.51,
"text": " Since we're talking about sci-fi, we can elaborate on... I mean, I personally think that anthropobots or other synthetic living biological tissues can, if you can scale them up, could even be used for construction, like for actual inhabitable structures. So I actually have a background in architecture and what really brought me into this field of... That's interesting."
},
{
"end_time": 3914.718,
"index": 168,
"start_time": 3885.623,
"text": " Yeah, I mean, I recognize that in biology there is this ability to embody, you know, self-construct with sort of this embodied morphogenetic code. And as Mike was talking about, this comes in layers. We're often sort of conditioned maybe from, you know, middle school biology classes to think about DNA as the rule book for everything. But what we're discovering"
},
{
"end_time": 3941.101,
"index": 169,
"start_time": 3915.742,
"text": " is that yes, genetic code is important, but there are additional layers. You know, there's epigenetics or bioelectricity that together make up this morphogenic code. And how can we edit this code to steer these cells into completely new architectures that never existed before? Whether to use it, either to use it in medicine, in these ways that we're talking about,"
},
{
"end_time": 3969.428,
"index": 170,
"start_time": 3941.34,
"text": " Or again, since we're talking about sci fi, I'm personally interested in scaling this up into building like self constructing bricks. And then, you know, by using those bricks, can we build living architectures? Because, you know, yeah, I mean, when you look at the global warming, like more than 40% of the CO2 released to the environment is coming from the construction industry, just trying to like build things. Yes, yes."
},
{
"end_time": 3977.961,
"index": 171,
"start_time": 3969.974,
"text": " And that's using the knowledge that we have developed as humans for thousands of years in the field of architecture, civil engineering."
},
{
"end_time": 4002.125,
"index": 172,
"start_time": 3978.49,
"text": " But at the same time, when you look at nature, it's also able to build structures that scale like oak trees or a lot of, you know, like you see a whale and a biology can build, you know, large things without let alone releasing carbon, but by sucking carbon from the environment. So it's literally carbon negative, the exact opposite of what we are doing as humans."
},
{
"end_time": 4027.602,
"index": 173,
"start_time": 4002.363,
"text": " And now, like in the 21st century, we're learning that that's actually also programmable. It's not set in stone. You're not limited to only what's evolved out there. But as a biologist, engineer, designer, you can actually edit nature and create new structures. So why not create structures that we can use to solve some of the other problems like sustainable construction?"
},
{
"end_time": 4038.217,
"index": 174,
"start_time": 4027.602,
"text": " Yeah, or even space exploration. I mean right now it's real difficult to leave the planet because of the gravitational pull. So the more weight you have"
},
{
"end_time": 4064.974,
"index": 175,
"start_time": 4038.49,
"text": " you know, aboard in your spacecraft, the more you're being pulled back, you'd have larger rockets to get you out. But then those rockets also pull you down even more because of their weight and so, you know, intricate balance. And are you really going to put in a bunch of like dead weight, like bricks, you know, or concrete precursors to build in like outer space? Instead, can you just take a tube of engineered cells"
},
{
"end_time": 4091.817,
"index": 176,
"start_time": 4065.35,
"text": " Right. That's super cool. So Gazem, you don't know this, but my background's in math and physics and I often wonder what would I do if I had to do it over and I had to choose something else? What would I choose? And I"
},
{
"end_time": 4111.596,
"index": 177,
"start_time": 4092.398,
"text": " I think it would be civil engineering. Well, I often say this because I look at houses and how they're constructed and it looks the same as it's looked for the past few decades and it takes just as long. And I always think, man, I wish that that could be done much quicker. I'm also a germaphobe. So in addition to being in mathematical physics, there's germaphobia and I'm a nerd."
},
{
"end_time": 4141.869,
"index": 178,
"start_time": 4112.176,
"text": " Dragon Ball Z and Dragon Ball Z. I don't know if you know this show, but you have this little capsule you bring with you and then you throw it on the ground and it becomes a house or it becomes whatever. Yeah. So that's really the dream. Yes. I like this idea of building a robot from smaller pieces. Also because like I'm a germaphobe and I want robots to clean washrooms. I just feel horrible for janitors. I feel like, oh my gosh, how do they do that? And they should be paid so much more. I wish a robot could just go and do that. And then a robot would deconstruct because then you'd wonder how do you clean the robot?"
},
{
"end_time": 4162.961,
"index": 179,
"start_time": 4142.824,
"text": " Yeah, so I hope that you could make some inroads there. Yeah, I mean, we have microphages that go down like chase bacteria and swallow it and degrade it. So we can maybe transfer the same principle to bio bots. What's the lifespan or functional lifespan of the Anthrobots that you've been studying?"
},
{
"end_time": 4179.292,
"index": 180,
"start_time": 4163.507,
"text": " Yeah, so for them to build themselves, it's about two weeks. And then for this, so we at the very beginning talked about releasing them from the matrix and then having them sort of turn inside out that morphological reorganization that takes"
},
{
"end_time": 4195.094,
"index": 181,
"start_time": 4179.292,
"text": " set of another week or so so three weeks is for their developmental um developmental face which we now work we're working on a follow-up paper to characterize those stages more but once it's a sort of adult bot that is able to move around"
},
{
"end_time": 4216.766,
"index": 182,
"start_time": 4195.094,
"text": " There is a variability there, we've seen them living from sort of month to multiple months but what happens in every single case if you're not like giving them different drugs or anything in their wild type and throwback case what happens at the end of these"
},
{
"end_time": 4242.944,
"index": 183,
"start_time": 4216.766,
"text": " multi-month period is every single time just degradation into individual cells and to debris. So they're able to naturally biodegrade, which sort of helps with the concerns around, well, if you put them into body, then what's going to happen? Will they ever like create clogs? What we have seen in the lab is every single time they degrade into individual cells. So then that's not going to be any different than the individual cells that your body disposes off."
},
{
"end_time": 4260.981,
"index": 184,
"start_time": 4244.07,
"text": " Now, I understand you didn't genetically modify these, but did you observe any other changes that were epigenetic or bioelectric? We haven't looked at their electricity. As Mike said, that's like definitely one of the next things to look at because there's so much there and that was"
},
{
"end_time": 4284.326,
"index": 185,
"start_time": 4261.442,
"text": " That concept was foreign to me until I came to Mike's lab. Up until my PhD, I was doing a lot of actually genetic editing, synthetic circuits to change morphology. And it was here that I'm realizing that there are all these other layers, like the complete morphogenetic code. So no, that's not something we've done yet, but we would love to look at."
},
{
"end_time": 4305.299,
"index": 186,
"start_time": 4284.326,
"text": " But other epigenetic, so I mean this whole thing that we talked at the beginning about character formation, right? So none, no two entrobots are identical, but we have these morphotypes and behavioral classes that also happen to influence each other. So figures two, three, four in the paper."
},
{
"end_time": 4332.585,
"index": 187,
"start_time": 4305.742,
"text": " I mean that has got to be a result of some sort of epigenetic influence because again every single anthropot starts with the exact same DNA but they end up in different sort of morphological flavors one is fully covered with cilia one sort of morphotype another morphotype is fully polarized so half of it is covered with cilia the other half is bald"
},
{
"end_time": 4362.381,
"index": 188,
"start_time": 4332.585,
"text": " The third type is, again, silly everywhere, but much more sparse, so like a checkerboard pattern. I mean, if all of these bots have started with the same DNA, what's causing this morphological, you know, variability? And that has got to be epigenetics. We don't yet know what those epigenetic knobs are. We have some supplementary data in the paper that discuss Matrigel, the matrix,"
},
{
"end_time": 4379.002,
"index": 189,
"start_time": 4362.381,
"text": " viscosity as a potential factor or their initial cell density so that's sort of reminiscent of Angela's work like based on how many cells we start with the resulting answer about population profile is the frame."
},
{
"end_time": 4397.602,
"index": 190,
"start_time": 4379.48,
"text": " So those are the only two things we've looked at and we have significant differences. So those seem to be like maybe one of the first couple of knobs, but we still have a long way to understand what are the knobs and then what's the underlying sort of mechanism that's making those knobs to be the ones that are influential."
},
{
"end_time": 4425.179,
"index": 191,
"start_time": 4398.097,
"text": " Yeah, one thing that I just wanted to add on top of that is that the original meaning of the word epigenetics was basically everything that's not the genome, right? So that includes bioelectricity, right? So traditionally that would include biophysical factors, biomechanical factors, ionic factors and so on. So when we say, at least when I say epigenetic, I don't just mean the chromatin modifications that people focus on today, you know, the methylation, acetylation, all that stuff."
},
{
"end_time": 4453.507,
"index": 192,
"start_time": 4425.179,
"text": " but actually all these other things so we don't know I mean there could be of course there could be chromatin modification effects but but we're you know the next step is to look at the at the bioelectrics and and probably biomechanics too and other other aspects of the physiology of it. So there was a term gazem that was used and Michael I'd like you to explain it just for the audience to know it was matrix and the term escaping the matrix was used and so because we're dealing with people thinking you've cracked the code"
},
{
"end_time": 4476.408,
"index": 193,
"start_time": 4454.019,
"text": " You need to explain what's meant by that because people are going to think everyone here is 95 years old and you just made yourselves look younger. That's right. It's the cosmic matrix. No, it's not the cosmic matrix. It's yeah. So cells produce the stuff called extracellular matrix and it's basically just a collection of important molecules that sit on the outside of cells and between cells and they"
},
{
"end_time": 4499.957,
"index": 194,
"start_time": 4476.408,
"text": " It has all sorts of functions, including as a repository for information, so much like ants leave each other messages in their environment, right? It's a thing called stigmurgy when you can use the environment as a scratch pad. So extracellular matrix in vivo is this kind of like rich set of molecules that are hanging out between and outside of cells that can also be used as information."
},
{
"end_time": 4518.251,
"index": 195,
"start_time": 4499.957,
"text": " What happened when you tried to block communication are you unable to because you didn't know what the communication was."
},
{
"end_time": 4535.043,
"index": 196,
"start_time": 4519.514,
"text": " Yeah, so based off of our hypothesis that communication was occurring through this calcium ATP signaling mechanism, we did try to use different inhibitors of calcium and ATP. And what we see is that when we inhibit either or,"
},
{
"end_time": 4561.715,
"index": 197,
"start_time": 4535.043,
"text": " The survival of our embryos actually decrease so they basically become singleton so they act as if they're being raised by themselves so far by blocking those messages the embryos now think oh no i have no neighbors even though they are still in a group of one hundred three hundred. And so it would go down. To the level of what it would be if it was a single embryo."
},
{
"end_time": 4580.179,
"index": 198,
"start_time": 4561.971,
"text": " So yeah, so what we saw was simply by blocking these communication avenues that these embryos are no longer able to basically sense each other. Now they think that they're just being raised by themselves, despite that not being the case. You were able to test it also without the teratogens, but with the communication blockers?"
},
{
"end_time": 4607.875,
"index": 199,
"start_time": 4580.776,
"text": " Yes, yeah, yeah. So we did just regular media in which we grow up our embryos and just added the inhibitor to see, okay, does the inhibitor have some sort of effect? And we see that by itself, the inhibitor doesn't. But with a teratogen on top of the inhibitor, then we see this communication basically going to zero. And so what were the controls that you used to establish that the group size was the factor and not some other environmental variable that is somehow correlated with the group size?"
},
{
"end_time": 4630.145,
"index": 200,
"start_time": 4609.053,
"text": " Yeah, so there is a number of things that I tried to normalize. So we tried to scale up dish size. So whether the embryos have more space or not, that was a concern. So I just would increase group sizes or dish sizes as group sizes increase. Obviously, the media and the tritogen, so how much each embryo is being exposed to was a major concern because"
},
{
"end_time": 4660.486,
"index": 201,
"start_time": 4630.52,
"text": " the first thought is, okay, well, if you have an N of one, that N of one is being hit with all the tritogen versus if you have an N of 300. Now you have more individuals to help you break that up. And so to address that, I scaled up both the amount of media that each embryo is getting as well as the data now, kind of like what you and Mike were saying, where one X, an embryo of one will get a one X versus an embryo of three will get three X of that. And so we try to normalize for as much as we can so that whatever we see is"
},
{
"end_time": 4689.104,
"index": 202,
"start_time": 4661.681,
"text": " But also the other important thing there is that in addition to these drug, you know, if it were just drugs, then you have to deal with a drug breakdown and all this stuff. So the other important set of data in that paper is doing the same thing, but with RNA injected directly into embryos. So there are no drugs. There's no issue of what happens to the drug in the medium. It's just every embryo gets the same amount of RNA that normally would destabilize development in a particular way."
},
{
"end_time": 4712.671,
"index": 203,
"start_time": 4689.599,
"text": " And the other thing to add here, which is important, is that it's actually doubly surprising because the standard, if you were to ask somebody, okay, I have 50 embryos in addition, what I'm going to do is instead put 300 embryos in addition. What do you think is going to happen to their health? Typically, the expectation would be it should get worse because by having more embryos together, you have more opportunity for"
},
{
"end_time": 4736.203,
"index": 204,
"start_time": 4712.961,
"text": " So,"
},
{
"end_time": 4765.674,
"index": 205,
"start_time": 4736.698,
"text": " Angela, have you extended your study to see if any of the advantages observed in the early development with the SEMA effect have led to better or different outcomes as the embryos grow further? Yeah, so I mean, we've taken our embryos and we stress them out early, but then we follow their growth. So we follow their development until they hit basically stage 45, which is as long as IACUC allows us to grow."
},
{
"end_time": 4788.285,
"index": 206,
"start_time": 4766.049,
"text": " Okay, how many of this group is still alive? So we do follow their development kind of as long as we can, at least from the one cell stage up until stage 45."
},
{
"end_time": 4804.497,
"index": 207,
"start_time": 4788.78,
"text": " Yeah, so we have an age range from 21 to 76."
},
{
"end_time": 4833.097,
"index": 208,
"start_time": 4805.094,
"text": " years old humans donating their sort of the lung tissue at the bifurcation point. And this these progenitor cells that sit at the base of the trachea, the tracheal epithelium, looking directly to the matrix in the body, extracellular matrix in the body. Those cells from all these different patients were able to give rise to entrobots."
},
{
"end_time": 4857.449,
"index": 209,
"start_time": 4833.353,
"text": " One thing to clarify here is like for a specific entrobot population, we start with in a single sort of dish, let's say, we start with 15,000 cells. And we only get actually like hundreds of bots. And the dish is sort of a centimeter across. So it's a very, very tiny dish."
},
{
"end_time": 4880.282,
"index": 210,
"start_time": 4857.449,
"text": " So not every single cell becomes an anthropobite, but every anthropobite based on what we're seeing is coming from a single cell. So in other words, from a human donor, it's not guaranteed that every single human lung progenitor cell will become an anthropobite, but enough many of them. And this is due to the senescence."
},
{
"end_time": 4908.49,
"index": 211,
"start_time": 4880.674,
"text": " But enough many of them become that it's every single time you have a high throughput method that give us a sizable population. Well, this is also fascinating and I'm very much looking forward to seeing the follow-up research. Man, oh man, the two papers will be linked in the description. They're also on screen. And by the way, if you're just listening to this, I recommend you watch it on YouTube because at any point when"
},
{
"end_time": 4934.77,
"index": 212,
"start_time": 4908.558,
"text": " Something is brought up. For instance, Michael, you referenced a result of graph that's on screen at the same time. So what I want to know is what's next for you all? Both personally, I understand that some of you are getting your PhD soon. So both personally and then also as follow up research with regard to what we're talking about today. Gazem, I'll start with you and then Angela and then Michael, I want to know about what's next for you, especially the mind everywhere project. So Gazem, please."
},
{
"end_time": 4963.319,
"index": 213,
"start_time": 4935.981,
"text": " Sure. I actually just defended my PhD a few weeks ago. Congratulations. Thank you. I'm in a transition phase and what's next is to trying to understand the capabilities of native capabilities of these end robots, as well as starting to engineer them for specific outcomes based on sort of what target morphological and functional"
},
{
"end_time": 4986.817,
"index": 214,
"start_time": 4963.66,
"text": " How does it feel to be done your PhD or at least to almost be done completely? Thank you. It's still surreal. It's been more than five years that I've been working on this project and I've"
},
{
"end_time": 5000.691,
"index": 215,
"start_time": 4988.114,
"text": " It's been sort of the because when I first started I was I really wanted to work with Mike because I knew that he was really interested in cracking the morphogenetic code and understanding"
},
{
"end_time": 5027.705,
"index": 216,
"start_time": 5001.084,
"text": " You know, like for me, understanding how the hell the, you know, the natural architectures build themselves, like, how is it exactly happening? What are the control sort of parameters? So that's why I was really interested in conducting the research research in this lab and learning about all these new sort of non genetic approaches to doing that was"
},
{
"end_time": 5056.22,
"index": 217,
"start_time": 5028.063,
"text": " new for me. So, um, and I've just been really surprised by the level of, uh, change we were able to induce in these cells. I mean, this is really radical going from a single cell to something as complex as an answer. But, uh, so far what I had done, like in my masters was, uh, the more sort of genetic approach. Okay. If I put in this gene, this will happen that gene. So those were really small changes."
},
{
"end_time": 5075.555,
"index": 218,
"start_time": 5056.715,
"text": " Angela, what's next for you?"
},
{
"end_time": 5096.254,
"index": 219,
"start_time": 5076.271,
"text": " Yeah, so I will hopefully be following to them and graduating fairly soon. Now that this paper is done and it's out for the world to see. It's kind of been a long process. I started it my rotation like six years ago and it survived a pandemic and everything. So it feels good to at least be close to a finish line."
},
{
"end_time": 5116.971,
"index": 220,
"start_time": 5096.988,
"text": " And then, yeah, I hope to be able, so afterwards I'll continue in Mike's lab for a little bit after just to wrap things up and hopefully continue to pursue this idea of looking at how genomics aren't really responsible for everything of an organism."
},
{
"end_time": 5144.497,
"index": 221,
"start_time": 5117.21,
"text": " Yeah, we're hopefully going to continue to look into this project, look at different avenues of communication, look at some of these things that we didn't really have time to look at with this first paper, but hopefully follow up experiments are going to be very exciting. And by the way, what surprised you most about this research, like during it or from the reception? I understand that, hey, it's being released today, so it wouldn't be reception from the public, but you get the idea. Yeah, so I think"
},
{
"end_time": 5163.131,
"index": 222,
"start_time": 5144.735,
"text": " For me, this whole idea of how our genome doesn't encode for our anatomy, right? So if you think about a tadpole and a frog, they have the same genome, but their morphologies are completely different. And so kind of diving into this little hole of like looking into, okay,"
},
{
"end_time": 5192.295,
"index": 223,
"start_time": 5163.422,
"text": " Given that, you know, the whole field is focused on genomics and genetics. But what else is there that can contribute to our morphology? So what other instructive information can we receive or that we can give in order to form to form properly? Angela, the frog that you used, it's a certain species called the Xenopus laevis. And for people who don't know, that was the embryos. The embryos we keep talking about are from that species. Now, that's been a staple of Michael's work for at least a decade."
},
{
"end_time": 5219.497,
"index": 224,
"start_time": 5193.609,
"text": " Why? What separates this frog, this species, from other frogs and other species? What useful properties does it have? Yeah, so I think Xenophys laevis is one of the reasons that we use it is because the genome is well studied. So there's something called ZenBase, which has all the genomic information. It has expression. It has a lot of staining that people are interested in. So having that database, I think, is very useful."
},
{
"end_time": 5244.462,
"index": 225,
"start_time": 5219.94,
"text": " Also, another perk of Zenopost is that the embryos develop outside of the organism and you can see it. The embryos are quite large. You can see them with the naked eye. And then with the help of a microscope, you can track them and you can watch them develop from a single cell to a two cell to a four cell stage. So it's an ease almost."
},
{
"end_time": 5269.753,
"index": 226,
"start_time": 5245.026,
"text": " Being able to just watch it and having a database to compare it to, I think, makes it a very strong model organism. And I mean, there's going to be perks and cons to every organism, I think. Obviously, I have a preference for those, but yeah. Angela, something else is that with the RNA sequencing, you uncovered that there were transcriptional changes associated with the SEMA effect."
},
{
"end_time": 5277.585,
"index": 227,
"start_time": 5270.52,
"text": " So were you able to identify any genes that responded differently in the small groups versus the large groups?"
},
{
"end_time": 5305.094,
"index": 228,
"start_time": 5278.899,
"text": " Yeah, so in our paper, we do an RNA-seq. So we look at the RNA of these embryos. And essentially what I do is I have a large group of 300 embryos and a small group of 100 embryos. So we're comparing the group sizes of both treated and untreated individuals. For the purpose of the paper, we focus on the two group sizes that are being treated. And we see is that there's a total of like 16 genes that are up and down regulated."
},
{
"end_time": 5331.493,
"index": 229,
"start_time": 5305.708,
"text": " And there are specific types of pathways that are being used in the small groups and specific types of pathways being used in the large groups. So that almost indicates that if you're in a large group, you're coping with the stress in one way versus when you're in a smaller group, you're coping with the stress in a different way. Transcriptional changes mean that the gene is then expressed differently or what? What does it mean?"
},
{
"end_time": 5360.435,
"index": 230,
"start_time": 5332.329,
"text": " Yeah, so with RNAseq, we're looking at how many copies of the gene there is essentially. So looking at the profile of the RNA. So what RNA is present and how much of it is present. Great. Michael, what's next for you? Yeah, well, you know, the first thing I want to say is just how incredibly proud I am of both of Gizem and Angela and the rest of the teams because there were lots of other people, collaborators, undergrads and other postdocs and"
},
{
"end_time": 5382.927,
"index": 231,
"start_time": 5360.776,
"text": " but just like they've done such an amazing job pushing forward this project and everybody needs to understand it is really hard not only to it like innovating in science is hard and getting something to work is hard and and and both angeline gizem have been you know faced all kinds of issues to get this stuff to work but but then just just the idea of being a young person at that stage of the career and"
},
{
"end_time": 5404.292,
"index": 232,
"start_time": 5382.927,
"text": " i'm having something that's so different and kind of counter expectation it's a lot of pressure and it's a lot of it's a lot of responsibility and i just you know i think people should understand that like it's hard for anybody else listening to this you know who might be in that position might be considering going into it i don't know what what you all feel but to me it's as hard and and does."
},
{
"end_time": 5423.524,
"index": 233,
"start_time": 5404.292,
"text": " As much pressure as it is it's the most fun you can have so you know that's just what you want everybody understand like it's it's it's not just it's not just standard incremental signs like this was like they did an amazing thing so i'm so that's the first thing i wanted to just say that you know i'm super super proud of you both."
},
{
"end_time": 5452.415,
"index": 234,
"start_time": 5424.053,
"text": " Yeah. Let's see. Next. Well, obviously, we're going to continue with some of the stuff that we talked about. So the practical uses on the Anthrobots, the practical uses of them, the ability to try to understand how to program them towards new and controllable shapes and functions. And also, because in my group, we're very interested in basal cognition in general and diverse intelligence more broadly,"
},
{
"end_time": 5476.049,
"index": 235,
"start_time": 5452.415,
"text": " Really understanding what are the properties of the the proto cognitive properties of this new model system what what do they know how to do do they can they form memories can they learn from their environment can they do they have preferences about different different lifestyles are different outcomes that that can be fall them and so on and none of this is known we've made no claims yet as to their level on the sort of you know the spectrum of"
},
{
"end_time": 5487.261,
"index": 236,
"start_time": 5476.323,
"text": " of cognition. We have no idea where they fit, but the one thing I know for sure is that we don't guess. We have to do experiments and find out. So we're going to find out. So those are the anthropo stuff."
},
{
"end_time": 5516.749,
"index": 237,
"start_time": 5487.841,
"text": " Um, on the, on the SEMA front, uh, clearly trying to, of course, uh, better understand how it works and how the information is encoded that goes between the, um, the, the, uh, embryos and really to make models of this as a collective intelligence. So we already know we've already, we've already made models that, and other people have to that treat individual cells within the growing embryo as a collective intelligence. But it turns out there are multiple levels to this. Maybe not surprising in the end that, um, that maybe the group is"
},
{
"end_time": 5544.121,
"index": 238,
"start_time": 5516.749,
"text": " also has the ability to compute the path through anatomical space. So really understanding this, and then on the biomedical side, learning to induce it at will. Because as you said, you know, in the human case, where you have a patient, not necessarily an embryo, but a patient, they're not part of a 500, you know, a connected 500 individual cohort. But could we could we fake it? Could we you know, is there a way to is there a way to give that information that they would have had if they had been?"
},
{
"end_time": 5572.807,
"index": 239,
"start_time": 5544.121,
"text": " So those are the kinds of things. And then again, to really understand how the, you know, we think individual cells are using the bioelectric networks as a cognitive medium. What are embryos using such that the group has the ability to solve certain problems? Is it a field of ATP and calcium and who knows what else, right? So these are all things that we're going to be tackling next. I think it's going to be very exciting. And the Mind Everywhere project?"
},
{
"end_time": 5579.855,
"index": 240,
"start_time": 5573.575,
"text": " Yeah, I mean, both of these projects are key elements of that."
},
{
"end_time": 5609.121,
"index": 241,
"start_time": 5580.282,
"text": " Yeah, so that TAME paper was version 1.0, so I'm working on the next one. It's going to be a little while still, but we've learned a lot since that first one. We're getting, I think, better conceptual foundations, better computer models of what it takes to scale cognition. This idea of the cognitive glue, what is it that enables competent individuals like cells or even molecular networks"
},
{
"end_time": 5629.65,
"index": 242,
"start_time": 5609.121,
"text": " To scale up into larger iq individuals that solve problems in other spaces bigger bigger boundaries of the self and so on so we have been to these two projects are part of it there are many other projects that contribute as well so yeah moving moving along. Are you planning on writing a book michael to introduce your studies to the audience."
},
{
"end_time": 5651.425,
"index": 243,
"start_time": 5630.623,
"text": " Yeah, so I'm committed to one book. Oney Pagan and I are writing a book on bioelectricity. We have a contract with Norton. It's due towards the end of 2024. So that's something that we're definitely doing. So that's a basic book on bioelectricity and it's kind of import for medicine and so on."
},
{
"end_time": 5674.804,
"index": 244,
"start_time": 5651.425,
"text": " In my head i sort of rattling around have two other books one that's on this basil cognition kind of topic that sort of talks about the scaling of intelligence from very minimal systems all the way up and then maybe one after that we'll see. First i can't even imagine finishing this first one that i'm doing so i need to get past that."
},
{
"end_time": 5696.527,
"index": 245,
"start_time": 5675.128,
"text": " Thank you. Thank you all for coming. Thanks for having us. Yeah. Thank you for having us. Thank you so much. Yeah. Thanks. Thanks for having us. Yeah. Great discussion. Thank you. Great questions. I was surprised when you said you were not a biologist. I was sure you must have had like some sort of maybe at least like undergrad or master's training in biology because the questions are really good. Thank you. Thank you. Spot on."
},
{
"end_time": 5721.681,
"index": 246,
"start_time": 5697.722,
"text": " The podcast is now concluded. Thank you for watching. If you haven't subscribed or clicked that like button, now would be a great time to do so as each subscribe and like helps YouTube push this content to more people. You should also know that there's a remarkably active discord and subreddit for theories of everything where people explicate toes, disagree respectfully about theories and build as a community our own toes."
},
{
"end_time": 5739.787,
"index": 247,
"start_time": 5721.681,
"text": " Links to both are in the description. Also, I recently found out that external links count plenty toward the algorithm, which means that when you share on Twitter, on Facebook, on Reddit, etc., it shows YouTube that people are talking about this outside of YouTube, which in turn greatly aids the distribution on YouTube as well."
},
{
"end_time": 5764.94,
"index": 248,
"start_time": 5739.787,
"text": " Last but not least, you should know that this podcast is on iTunes, it's on Spotify, it's on every one of the audio platforms. Just type in theories of everything and you'll find it. Often I gain from re-watching lectures and podcasts and I read that in the comments. Hey, toll listeners also gain from replaying. So how about instead re-listening on those platforms? iTunes, Spotify, Google Podcasts, whichever podcast catcher you use."
},
{
"end_time": 5789.804,
"index": 249,
"start_time": 5764.94,
"text": " If you'd like to support more conversations like this, then do consider visiting patreon.com slash curtjymungle and donating with whatever you like. Again, it's support from the sponsors and you that allow me to work on Toe full-time. You get early access to ad-free audio episodes there as well. For instance, this episode was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough."
}
]
}No transcript available.