#272 | Albert-László Barabási; How ideas spread into society, Nutritional dark matter, Networks

Albert-László Barabási [00:00:00] Network effects. And I would go as far as to say all the companies that we admire today, from Apple to Google, to, to Facebook, Twitter, even Cisco and the big router companies, they're all network based companies. Their business model is based on the network network. You take the network of away and the company doesn't exist.

Vance Crowe [00:00:25] I'm Bruce figure a veterinarian living in Sylvia, Kansas, and you're listening to the Vance Crow podcast. Welcome back to the podcast. I'm glad you're here. Today we talk with one of the world's leading network scientists, and I had a chance to sit down with Laslo Bar Babai to be able to discuss, Hey, how do ideas spread into networks? And longtime listeners of the podcast know, this is something deeply and intensely interesting to me because of my work at Monsanto trying to change cultural perception about modern farming techniques. But what I didn't expect in this conversation is that we would talk about flavors and tastes and the, the challenges that are going on with the American diet, really all of the Western diet with ultra processing foods. This is a really interesting conversation with a man that is both a network scientist and an artist, and now using his laboratories at some of the most important schools on the east coast to really do cutting edge science to figure out not only how do genetics impact us, but how do the food and the molecules we don't know about really have an impact on us. He refers to them as to the nutritional dark matter, and he is full of different and interesting concepts like this. So before we get to that, I wanna talk just for a second about the Legacy Interviews. The other day, I had a listener call me up and say, you know what you should ask your listeners to do? And that is, if they're not right now in the position to buy a legacy interview, to call up somebody and say, Hey, you have a parent or a grandparent that I think really matters a lot to you.

Vance Crowe [00:02:00] And I wanted you to know about this service because I've heard from people that have used the service that more people need to know about it. And right now our biggest avenue to get to them is through the podcast. So if you have been thinking about a legacy interview, but you don't necessarily have somebody for us, then maybe consider referring a friend and you can send them to Legacy Interviews dot com. And there we can either do an online interview over the same medium that I used to interview Laslo, or we can do it in person in our studio here in St. Louis. So that is a website address again, is Legacy Interviews dot com. Alright, without further ado, let's go to this interview with Laslo Barabasi. Laslo Barabasi, welcome to the podcast. Thanks

Albert-László Barabási [00:02:45] For having me.

Vance Crowe [00:02:47] Listeners probably wouldn't know this, but we are in the presence of greatness. You are deeply embedded in the world of networks and how they work and discovering new things about them, which is a field that I find completely fascinating. It's informed a whole bunch of my work in the past. So why don't we start off by having one of the field's greatest scientists describe what is a network and why should people care about it?

Albert-László Barabási [00:03:10] Well, that's, that's a low ball question, right? Because, because at the end, our very biological existence is determined by networks, right? If you think about the genes and the metabolites in our body, the wall kind of evolution is based on the emergence and the existence of networks that support life. If we wanna stay at the example of life, our brain is a network, obviously, of network of neurons. If you think about ecological systems, those are, you know, interconnected species that support the ecosystem as we have it. But of course, we have actually networks in social media. We have technological networks like the internet and powered and so on. And the reason why we have a network science today is because about 20 min, 20 years ago, myself and several of my colleagues have discovered that even though these networks are very, very different from each other, sometimes the node is a molecule, sometimes it's a person, sometimes it's internet router, sometimes it's a is a food. And the links can be social interactions, chemical reactions, you know, physical cables, and you name it. Despite all of these giant differences, there's a common architecture to many networks. And by understanding the common architecture, you have a better way of describing analyzing them and eventually predicting their behavior.

Vance Crowe [00:04:37] So in, I think people are really fairly acquainted with like Twitter, right? And you can, you have the different ways of following people. You, you maybe, and you can make a graph that says, all right, I'm a node and everyone that I follow is a node. And we can draw a line in between those. But when people are looking at a graph like that, it really just kinda looks fancier. It looks like art. What do you, what are you able to divine by going and looking at those graphs? What kind of information can emerge?

Albert-László Barabási [00:05:06] Well, Twitter is a very good example, and which is Twitter is a space where everyone can be heard, right? You have an option to actually put your message out. And in principle, it could reach anyone else by people sharing this. So both the worldwide web and the Twitter kind of emerged as this democratic medium. In reality, it is far from being democratic, right? And that's hidden in the network. When you look at the network, you will actually find that there are a couple of large giant hubs. These are the people with tens of millions and close to a hundred million followers from Obama to Britney Spears. So, so, so that, so on one end, there are these huge hubs that connect the wall network together, and most nos have a very few nodes, only one or two followers, or maybe in a dozen sense. So one, and it turns out this is not unique to Twitter. When you look in the cell, there are a couple of genes that interact with almost every other component of the cell or a couple of proteins. And most proteins only interact with a few others. And when you look at the word why web, you got Google and Facebook that are giant, right? And then my webpage hardly anybody connects to, so, so what, what you see in Twitter in large scale is very, very similar to what you see in other networks. And that determines how Twitter is useful or being used. Obviously, if you have a message that you wanna get out, you want to get it in the hands of the hubs because if they kind of retweet your message, then a giant number of people will actually know about it.

Albert-László Barabási [00:06:40] And, and when you think about it, how ideas are spreading on the Twitter, how ideas, how, how kind of how you acquire links and so on, you are playing fundamental lessons in network science.

Vance Crowe [00:06:55] Yeah, there's, it's like one of those things that feels very unfair that the rich get richer, right? But once you have a network of a certain size, then it becomes easier and easier for your network to grow. So if you have two followers, it's a lot harder to get to 10 followers. But once you have a thousand followers, getting 10 more is easy to do.

Albert-László Barabási [00:07:12] Yes. This is actually one of my claims to fame professionally, and we call it preferential attachment. That is the growth rate of a node in any network is proportionate to how many links it has. And this is very profound because fundamentally it tells you the more links you have, the more friends you have, the more chances you have for somebody to introduce you to someone, right? And, and that mechanism plays itself out in all different type of networks. Sometimes the me the mechanism is obvious. Why is that happening? Sometimes it's subtle, like in biological networks, but you just pointed to one of the mechanisms through which these major hubs emerge.

Vance Crowe [00:07:54] I remember the first time I ever sat down and tried to do the, you can probably even name it, where you have a certain number of telephones there, right? And as you add another telephone, the value of that network gets more and more valuable because, you know, you go from having three telephones, so there's only three people that can communicate to four. And now all of those interconnections that as you start adding those on, it feels a little bit like you're staring at the edge of the universe when you realize just how quickly connections can grow and, and, and what exponential growth is really all about.

Albert-László Barabási [00:08:25] Yes. So what you just described is what we call professionally as the METCO law, right? The idea that you, like if I have, it came up in the time when the Xerox machines were at the just emerging and the idea that if I have a Xerox machine, it's totally useless because there's no one I can xerox to, right? But, so for the value to increase, more and more people should have it, and therefore everybody will have increased value. And that's a very traditional network effect. The, the problem has been with that type of calculation, that based the valuation early on in the internet era of many companies, and they wrongly assume that that's the real value of a network, is that it assumes that in a network everybody's connected to everyone else. In reality, networks are sparse. So let me give an example. There are 7 billion people on earth, and a typical person like myself or yourself know about a thousand people on first name basis. So only you are connected to only a tiny, tiny fraction of that large number of potential nodes that you connect to. So that seems to be a limitation, but it shows you that the network is sparse. So you're not really accessing everyone, but it also, there is value in it, because if I look at who your friends are, who you connect to, I know everything about you. So there is, there is, in your friendship circle, there's lots of information about you because your friends are not random. They relate to you, they have the same socioeconomic status, they have the same interest and so on as you are. So that's where the power of the network comes in.

Albert-László Barabási [00:09:58] Even though the network can be very, very large, and as you said, infinite and looks like exponentially growing at the local level, there is lots of very specific information that is really specific to you and can be used to predict your behavior, can be used to sell stuff to you, can, can be used to kind of show you information and so on. And really, if you think about the social media companies, all they're using is using network effects. And I would go as far as to say all the companies that we admire today, from Apple to Google, to, to Facebook, Twitter, even Cisco and the big router companies, they're all network based companies. Their business model is based on the network. You take the network of away and the company doesn't exist. So this is really what happened in the 21st century that the successful businesses must earn us network effects. And if you wanna do it right, you need to speak network science.

Vance Crowe [00:11:02] When, so I had a, an a hypothesis about how ideas spread into the world. So I was, I was hired by Monsano who looked around at the world and said, people think we're evil. They think that everything we're doing and putting out there is bad, and we have no idea how to change that. And so when I came into the company, my first question was, well, why is it that everybody knows what they know? And I, and I don't mean that from like, like a, like a communications concept. I meant like, you know, everything I had ever heard about Monsano, every single person I knew knew that they were evil. So the question became how is it that information spreads and that people get these ideas? And, and if you wanted to change that, where in the network would you join it in order to be able to push new information

Albert-László Barabási [00:11:48] In

Vance Crowe [00:11:48] There? Is this a knowable thing or is this only a, a hypothesis in my brain?

Albert-László Barabási [00:11:55] No, it's totally, that's what many colleagues of mine do, and we occasionally did it as well as more like in the consulting space, is that how do you, how do you understand what are the roots of information? How does information pertaining to certain ideas spread? And the problem with this is that most people kind of assume that I, that information and ideas are like a virus that if you, if you've been exposed to it, then you have it and you're sick to it, a sig bite, right? But in reality, that's what we learned, that in that when it comes to information, it does not follow the simple contagion model that if you get in touch, you will get it, but it follows what we call a complex contagion, which means that information resides in communities. Let me be very specific. If you just hear about the book once, you may say, oh, that's interesting. And then you move on. If you hear it two or three times from your friends, you start paying attention. And in order for you to buy a book, you have to have, you have to be exposed from within your circle to three to five times. Unless you are really expert on that field and you know, you know, you, you know that you need it. So, so what we see is that when it comes to information and beliefs, they first invade communities and they get ingrained with in that particular community. And other communities may not even be aware of it or they would not care, right? And, and most messages like, like you do, are actually very, very ingrained in those communities. And, and it is those communities that you first have to identify, which is fundamentally a network science problem, because all networks are really for kind of clustered into communities like this network behind me.

Albert-László Barabási [00:13:38] You see there's a community here, there's another one there, and so on. And then, so you need to identify which of those communities are really holding those beliefs, and for which they're totally irrelevant, that belief. And then you need to start convincing those communities and engaging with them and understand the roots of that belief. So, so let's get rid of the idea that I, that of the concept that ideas are like COVID and we should think of it, that they're really each idea in order to get held in the comm, first of all, ideas exist within the communities and requires quite a bit of work to understand where they matter and where they don't.

Vance Crowe [00:14:18] So you, I'm so glad to hear you say this. So I, the way that we ended up ultimately conceptualizing it was, let's just imagine a single community, and there's a few people that are at the top of that community, like a mountain peak. And as they have ideas that they put into the world, it spreads down to the masses. So if you have, you know, high status or high value along the number of people that know it, the, when you get down to the masses, you're in the lower part of the, the peak. Well, I started realizing like the world is organized into tribes. And what we wanted to do was first find people that would derive value from changing their point of view. So they had to have, like, they already viewed the world one way, but if you could get them to change their opinion, they would be able to extract some kind of a value. So we called those tribes, and I've always had trouble putting it into a visualization, but I really feel like it's a fitness landscape, right? Like where there's many mountain peaks and just reaching one mountain peak isn't that valuable, but finding a way to connect one mountain peak to another, that's where all the value is. So trying to find those very few people in small groups and then letting them work on their own communities was the most valuable way to go about it. Would you have done it differently?

Albert-László Barabási [00:15:32] Absolutely. And the trick is, you know, that often the influencers within the community are not hubs otherwise. So to to be very, very influential in a certain community, you don't have to have 10 million followers, you just have to have a thousand followers from that community, right? And, and that's where people often get it wrong, is that they assume that I just need to get the major hubs, but the major hubs may not be really credible within the communities that you need to worry about. And then there's another wall story within, when it comes to belief systems is that if there are people who are impossible to change, they can have a very long and lasting effect on that community. So there's lots of work in the literature, particularly in the political and you know, the election space to show that if there's a small minority who really very strongly believes a certain type of message, it's those are who can, who cannot be flipped, you know, by, by the neighborhood, they can eventually win in the world long time. So and so, so that you need to kind of go ahead and start looking into these fine differences. And I think that's what network science brings you, your ideas, how you started out are totally intuitively correct and you had the right picture. I don't know if you had the right operation to really identify who those people are and where are those communities, and that's what network scientists provide you.

Vance Crowe [00:16:56] Well, okay. And there, there is where the big gap was because, you know, if you're working for a corporation, they're like, all right, let's figure out how to do this and let's go as fast as we can to automate it. So they would say, here, we'll give you access to all of social media, anything you want, we'll buy the APIs and we'll give it to you. But it was so much data, there was so much there that it was nearly impossible. So I ended up finding that it was probably easier to be an artist than it was to be a factory in that where you're individually choosing them.

Albert-László Barabási [00:17:25] Yes, yes. You, you still need to be an artist when you work with big data, but what I just what we, what you need to be aware of that these two set does exist and people were able to successfully use it in other domains.

Vance Crowe [00:17:41] So In what ways do you see your work being applied? I actually, I wanted to go over, so you were talking about that small group of people that are able to really impact a network. I think Nasim Taleb calls those the intransigent minority, right? Where it only takes a few people to completely dig in their heels to move the center. We kind of have this concept that the, the world is decided by what groups of people want, but it's actually a small group of people that can make it so uncomfortable for everybody else on subjects. They don't really care about that. They just give up. Is that what you were referring to?

Albert-László Barabási [00:18:14] That's right, that's right. And the simulation showed that, that if you have a set of notes in the system, in a voter type of environment who are just some unwilling to, to change their behavior, they would eventually flip the wall network with time.

Vance Crowe [00:18:29] So the thing that I was really interested to talk with you about is that you have taken network science, you have the art behind you, you've, you've really pushed into the world, but you got into nutrition, which does not seem intuitive to me outside of the fact that it seems like current nutrition advice seems like astrology in some way. So how did you find this, this world?

Albert-László Barabási [00:18:50] Sure. So to understand that we, we need to go, go back about 15 years and understand that my lab in the last 15 years has been really focused on what we call network medicine, which is how the molecules in the, have the proteins in the cell interact with each other and how those interactions and the, and the breakdown in those interactions lead to disease. Why do we need networks to understand the disease? Well, the genome project has provided us all the, the list of full list of genes and proteins and metabolites in the cell. But this is like giving you the list of the parts in your car and, and you know, laying it down in your living room and it's here said, here is your portion, or here is your, and I say, please go ahead and assemble it, but I'm not gonna give you a wiring diagram, right? And, or you know, you have the pieces, now tell me what's wrong with my car. So, so fundamentally for us to understand how the cell works, we need that extra piece, which is the wiring diagram because the breakdown of a component is often not a problem for the component itself. It's a only problem for what, what it does with the other components that, for example, the gas doesn't get where it needs to do and so on. So, so therefore network medicine is a field that aims to understand these networks, map out these networks and have, try understand diseases through that. And we had some major successes in the last 10 years beyond the fact that there's a division of network medicine now at Harvard with about 200 researchers working the medical doctors mostly applying the network concept to disease.

Albert-László Barabási [00:20:22] But I also found that companies that if you, for example, have RA today, you will have one of our tools, your doctor testing you and will tell you what medicine you should actually be getting and you will not do the trial error approach. So network medicine is already impacting people, but one of the things I realized about five years ago is that network medicine is all about fundamentally genetics and interactions between the genetic components, but we are not considering the role of the environment. And the environment plays a very, very huge role in the chances where you will or you will not develop the disease. So even if you have all the mutations, all the genetic predisposition to have heart disease, if you change your eating patterns, you can totally delete that and, and kind of overcome those genetic effects. So the question I ask, how could we build the environmental effects in particular the role of the food into these beautiful network diagrams that we're building and into the predictive power that we have over there. So I said, well that's easy. Food is molecules, right? So, so let's just figure out how do the molecules in the food interact with the cells? And, and I thought that it would be easy to do because of 10 years earlier we did a project on flavors that is that what flavor chemicals are in each food. And in fact, this network behind me is the flavor network whose notes each note corresponds to a flavor, I'm sorry, an ingredient like, like, you know, like apple or orange and two ingredients are connected if they share the same flavor molecules.

Albert-László Barabási [00:21:58] And we did so back then because there was a hypothesis, molecular cuisine telling us that things this well together if they share flavor chemicals. So it's a bit like color matching, but if you put the same color twice in the image, then it looks like it looks better, right? Then definitely

Vance Crowe [00:22:14] Like oregano and basil have some sort of overlap, which is why they're always paired together.

Albert-László Barabási [00:22:18] That's right. That's right. So they're chemically overlapping and we were hoping to discover pears that may not be trivial and we did, and so on and so forth. But then we thought, oh, we have all the molecules that's in the fruit. And then when I started good looking into that, I realized I was that wrong. Yes, we knew the flavor molecules because the industry has invested lots of money of figuring out what are the volatile components in the food. But when it comes to what are the components, what is the bulk of the food? All we know are the nutritional components. Well you would say, well that's a lot, right? Yes indeed. The nutritional components are sugars, fats, certain type of vitamins, amino acids and so on. Altogether about 150 components that the US USDA tracks in the US food supply. And there are big tables telling you that what, what, what, what has what. But one of the things we realized is that that food has, at that time we thought that more than 20,000 components. Now we have in my lab a database that has more than 135,000 different molecules that are found in the food that we eat. And this is what we call the nutrition or dark matter because no one tracks that. You have no idea what really whether what certain chemical is in your food. And people believe that's not a problem because these are inert molecules, they don't interact with your cells and therefore they just go in and go out is the nutritional components that we absorb and that's that wrong. And that's what we understood that actually two third of the nutritional dark matter has already known health effects.

Albert-László Barabási [00:23:56] And let me give an example. So we know polyphenols are good for us, right? There are in plants and were recommended to eat lots of plants. They're generating what we call the secondary metabolism of the plants. They're very, very diverse and they have lots of positive health effects. They're believed to be antioxidant. However, what is interesting about polyphenols is that our metabolism doesn't see them that is there, they're not eaten by the metabolic pathways, but traditionally how nutrients get to us. But they're first broken down a little bit by the box in their body, in their gut, and then, and they enter the BroadStream and their role is purely regulatory, which means that they attach to certain proteins in our cells and they turn on and off certain processes.

Vance Crowe [00:24:45] Can I interrupt you here really quick? So I think like I, I did a ton of research trying to understand why you gain weight and why don't you and how does food enter your body? If I understand what you're saying, this is like somewhat mind shattering already, but that when you eat food, let's say you eat a piece of bread, your body sees that as mostly carbohydrates, maybe a little bit of fat, and it's breaking them apart and it's saying, okay, the carbohydrates are gonna go as quickly as possible into the bloodstream and the fat molecules will go and they'll be processed in some way. But that we've always looked at it as if you're not a protein carb or a fat, you're not a part of the metabolism, you're not being broken down as food. And so it's not tracked and it's not counted. Are we on the same page here? Okay.

Albert-László Barabási [00:25:28] That's right. That's right. So much of the nutritional components that we call nutritional elements are engaging with the metabolism. There's typically source of energy or source of some other metals that the metabolism need to function, right? Or some other chemicals. But the nutritional dark matter is mainly or more, most exclusively regulatory, right? And, and you need that without that actually because they are turning on lots of, lots of processes, including metabolic processes, they can mod modulate metabolic processes. So they tell the cells what to do, right? And how to do that and how fast to do that and so on. And, and that is totally on tracked, right? You don't have databases to tell you what polyols are in what plants and, and these 130,000 molecules that we curate in our lab, we have associations with multiple foods, but there is no central database that we contain that we spend quite a bit of time and money to kind of assemble that and it is not complete. So the nutritional dark matter is what is totally missing from the way we think about food. Yet that's where most of the health effects are. Let me put it this way. Nutritional components, you must have, it's like the the gas in your car, right? If you don't have it, the car will not run. So in that sense, we're absolutely right as that to focus that you must have the right nutritional components in the food. Without it you will be sick and you have lots of problems. But if you wanna understand how food really works, they are the necessary but not the sufficient component to understand the impact of on health.

Albert-László Barabási [00:27:07] And so, so my lab has been involved in the last kind of five years of on one end mapping out all the chemicals that we have, evidence that could be in food and what food and how much of it should be there on one end. This is kind of mapping out the nutritional dark matter and in the other map and developing tools through which you can actually take the, in each of these molecules and tell you what health effects it could have, what disease could cause, what could this disease actually could help. So to think of food as drug not only as a problem, but as a way of modulating disease to food.

Vance Crowe [00:27:46] Like how do you even begin to do this? Are you taking food and just putting it into a shredder and then seeing what can we discover? How, how does this even work?

Albert-László Barabási [00:27:54] We do lots of things. Let's talk about the nutritional dark matter, right? So on one end, the first thing we did is that let's, let's accumulate the knowledge that exists already. Well how, where is that knowledge? Well, there are some central databases that contain information about food molecules. They're highly incomplete, but they exist. The other thing we did is that we wrote a code that effectively read the full scientific literature in the last 50 years and identified every single time when somebody mentions a food and a chemical and, and we let the computers using different AI methods to identify that association that, that this paper says that this chemical could have this type of health effects, or if it says food and chemicals saying that there is evidence that this chemical is in the food. This is what we call a knowledge graph and it's an AI based tool that provided a very detailed database for us of like what are the health, food and chemical association. So, but then there was not enough. We also engaged with a number of mass spec facilities and we did actually send them several dozen foods and we asked them to do what we call untargeted metabolomics. That is rod. The problem is that currently mass spec tools are really optimized for gimme a hundred foods and you are interested in one molecule and I can tell you if it's there or not and how much of it is there. They're really poorly performing. If I just say, you here is, here is carrot, tell me what's in it. And, but they did it for us, right?

Albert-László Barabási [00:29:27] And so we did it mostly as a proof of principle because we wanted to see how far we can get. So right now we are at the point that we have this massive databases that is very, very incomplete and we have a plan to go forward. And as we speak, we are actually trying to convince some funders to help us fund a nonprofit organization to achieve that. That is to map out what's in the food. Because now what we can see, the bazo data that we collected and the methodologies that we developed with the combination of that data and AI in about five years time, we could map out all the food that is in the molecule. So we have the path forward, we are kind of pitching this fund to funders to fund another, a company, but a nonprofit organization to achieve that and make it available to all of us because I per personally believe that if we have detailed information about food, it, you know, it will be a total game changing of how we approach health through food. Think about it, we spend $3 billion to map out the human genome, the human genome all together, all the mutations. If you know all the mutations about yourself, you have a five to 10% predictive power about what diseases you will have. The the rest is food and environment. And we have spent $0 to try to really map out the nutritional dark matter. So that's something that humanity has to change and we need to kind of find a way to finance to really make a systematic effort to have public knowledge of what's in our food.

Albert-László Barabási [00:31:02] Because until we have that, it's meaningless to have an the, an informed information about how the food affects our health.

Vance Crowe [00:31:10] Well, and you see just how passionate people are about the discussion and and my earlier analogy to astrology seems apt here, right? Like people will say, oh, I think the reason that we're all becoming fat is because of seed oils or because of GMOs or pesticides or the fact that there's too much sugar or there's people shouldn't be eating meat more than once a week. And so, but when you break it down and you go all the way, like right now, the only thing we can say is, oh, it breaks down to proteins, carbs and fats. That's the answer. If it can't be found in that equation, it's really not worth knowing. Like you're one of the first people I've ever heard approach this. Well, you are the first person I've ever heard approached this not from doing it as like, let's just exclude these parts of our diet and then just see what happens. But to try and go all the way down to this level. Do you, do you, I mean clearly you think at the bottom of, of this deconstructing if model that you're gonna, you're gonna discover things by getting all the way down at the molecular level. It seems kind of like a big gamble.

Albert-László Barabási [00:32:13] We, well of course we will discover it and, and of course we're gonna be bring it back to health. You know, much of the nutrition, like particularly the self health in the nutritional space, mostly for communication purposes have, has been always seeking single bullet solutions. There is no single bullet when it comes to food. That's the unfortunate situation, right? And because the food, as we said, you know, like on a daily basis you are exposed to 20 to a hundred thousand molecules that modulate in a combinatorial way your health and no, no molecule ever comes alone. And because of that, it's almost impossible to really identify what one molecule does unless you do the approach that we do right? Is to say, let me understand first what's in the food, what do they come together with, right? And how each of them could affect our health. Let me give the example of the, the Beto Carin Beto Carin, which is, which has been found by multiple studies to have a positive effect on cardiovascular diseases. That is people who eat food reach in beta carotine seems to have a better cardiovascular health that naturally right? You know, and you know, it's also orange rights. So it gives you the, the carrots are the, the, the major sources. So that led to the normal, like the logical conclusion that if I would give you Beto keratin, you would actually have a healthier cardiovascular system.

Albert-László Barabási [00:33:46] And they did that and all the studies failed in that space. And, and that is because we don't even know if it's Beto carotene because every time a food a plant would have Beto carotene, it would also have about 400 other molecules that would never be, and so is the Beto Carotine just a biomarker that it comes together with the other things that are causing the health effects or it is the source of that positive health and most likely it's a biomarker, right? It comes together with the other ones and it's orange so it's visible, right? So, so, and I think that's how we really need to approach food that there is now a single solution. And furthermore, there's not a single solution for you because what works for you may not work for me and might might not work for your grandmother, right? So eventually we need to get to the point of what we call preci precision and personalized nutrition. And if somebody tells you they know how to do that, they're lying right now. Right? So, but that's the reality that how you are actually processing food depends on your bacteria in your stomach, which is measurable, right? It depends on the many mutations that you have, right? And as well as the composition of the food that that, that you are, you are effectively eating, you need to hold all of these two things together to really give you advice that is meaningful. We are very far from that, but we have a path towards that.

Vance Crowe [00:35:21] I mean, it it, to hear you describe it, it seems like we're kind of at the for humors of of medicine, right? Like where all we understand is phlegm and you know, blood or something like that. Like we just these vague notions. Do you think, how far in five years do you think you can get towards actually understanding what should you laslo need for your food versus what I should have?

Albert-László Barabási [00:35:45] Well, it depends on how much investment we make. So think about it, more than half of the biomedical research money currently about 55% is devoted to genomics. Genomics is responsible to 10, 10 to 10% of your disease conversation. We have a genomics institute at NIH with billions of dollars of yearly budget. We don't have a nutrition institute at N nih, right? So one day, I mean we're gonna get five years from now would be when would the biomedical community end up funding the agency acknowledge that food is not an afterthought when it comes to health, but it is one of the source of our health, end of disease. And, and, and if that transition happens, I personally think would be far more transformative than the human genome project.

Vance Crowe [00:36:41] Yeah, it's, I mean there's so much involved in trying to figure out what those molecules are doing because I, I mean, my impression is that the molecules change, maybe not the molecules, the chemistry of the whole thing changes. Like for example, when you grill meat, right? All of a sudden you're taking that meat from being in one state to another. How, how will cooking be a, a part of what you're looking at?

Albert-László Barabási [00:37:05] Of course it does change, right? But it doesn't change the bulk of the chemicals in there, right? If you're just on the surface and inside actually transforming some of them. And there is a whole literature on what changes and what doesn't when it comes to cooking and when it comes to processing and when it comes to ultra processing and so on. So I don't see that as a problem. I think it's, I see as a task to be addressed and tax that is being addressed. Actually there's far more work now in the literature about how cooking changes the food than how much, how much about what is really in the food to begin with, right? So, so, so I don't think that's an insurmountable problem. It's a problem that is real. We decided to approach this that we first look at the raw materials because you gotta get started somewhere and that's the beginning of everything. And, and so our food on project right now is mainly focused on the raw materials, but if we find data about burger composition as a mixture or a cook burger will not toss it away. We'll put it in the database. But much of the, much of the the research we've been in raw materials, but how far can you take? Well let me give an example. So we were curious about as polyphenols and their impact on health and on one end we were able to say what pronos, in case rosemary acid could actually have an effect on platelet formation, which is one of the big issues related to cardiovascular diseases.

Albert-László Barabási [00:38:39] And we didn't only say that that chemical should have that effect, but we're able to say exactly what genes and what proteins would be affected if you have more of that. And then we did ex my colleagues on jolo at Harvard did the experiment and show that that's exactly what happens, what we're predicting how adding to the cell line, for example, rosic acid will change the cell in the direction of kind of affecting cardiovascular signatures. So, but then we went further than that. We said let's take RA arar psoriasis and we use these network medicine tools to identify about 30 molecules that we believe that could affect the inflammation process in ra. And then we sent it to an independent industrial lab to test it in cell lines. And out of the 30 molecules, these are all foodborne molecules. Only one had no effect. About five of them were anti-inflammatory and the rest were pro-inflammatory. Both of these have value for us because now we know which molecules in your, in your food could affect RA and we know which one could make it worse and which one actually helps it. So if I need to rebalance your diet first I need to say, let me understand what you're eating and let's make a few changes to have more anti-inflammatory molecules and have fewer pro-inflammatory molecules. That's the dream, right? And again, that goal is not to simply say eat this as an RA patient, I need to first understand what you're eating to begin with and, and how you change that.

Albert-László Barabási [00:40:16] It's the same example that if you are, if you go to, if you go to a fisherman community, you're not gonna tell them to eat more fish, right? Because they're already eating enough fish and, but if you go to the middle of America, you may say, I think you should eat more fish, right? So, so these advices need to be in reference to what we already know about you and how we can actually correct what you have. Because let's face it, we know what is a healthy diet. It's a me diet, right? You can reproduce the MEE diet in US without problems, right? You it's mostly vegetable based diet diet, but it can, it has lots of garlic, it has olive oil and can actually support meat as well. So we know how to eat healthy. The problem is that if you're gonna tell America you have to switch to a ME diet, no one will really follow. So the only interventions you can have health wise is to say, let me understand where you are right now with your eating patterns, what you and your family eat, and let me understand what are the minimal changes that we can do on that to improve a disease that you may already have. That's the only hope we have right now to change people's habit because if you just tell you what is the healthy diet, most likely you will not care until you get

Vance Crowe [00:41:31] Sick. What are foods that you know you should eat more of that are difficult for you personally to, to take on?

Albert-László Barabási [00:41:40] There is, I, it's difficult to say that such a thing. So let me put it this way. Before, before I started working on the food on project, I would be the average Joe down the street. I would eat at Burger Kings and lots of other places. I would eat everything that was put in front of me. Since I started working on the food on project, I didn't become a vegetarian, but I became fundamentally a salad eater. If you cook me a a steak, I will have it because hospitality trumps beliefs, right? And it's okay, right? There's no problem. But every time I, for example, prepare a meal like I will for lunch today for myself, it will be a salad. So you can actually make changes that, that, that, and I wasn't a big salad eat before, but I really enjoy my salads. I learned to how to combine lots of different fruits and vegetables and herbs to really create a palette that I really enjoy. So there is no such a thing that I feel like I should eating something and I don't really enjoy that. Rather I kind of adjusted the way I prepare my salads and my foods to kind of fit the palette that I'm comfortable with.

Vance Crowe [00:42:53] So how did you discover, what was the, the way to make salads that you enjoyed? I mean, there are people right now sitting in tractors or washing dishes that just got done with a steak and they think, I don't, I don't have any interest in this game.

Albert-László Barabási [00:43:06] This doesn't

Vance Crowe [00:43:06] Sound right.

Albert-László Barabási [00:43:08] Well, that's a very interesting one and that's why I said there's no one size fits all, right? Because I think you have to have your own journey of where you are coming from and you know, if, if you wanna kind of start with a salad where you put slices of steak on for on top of it, go for it, right? If you replace the french fries with salad, you're already on a path towards winning this game, right? So, so it's really the combination support, and this is where I think many of these advices get it wrong, is that they want to kind of give a one bullet solution, but there is no one bullet solution that fits for all of us, right? We all come from very different backgrounds and eating habits and advice has to be adjusted to that.

Vance Crowe [00:43:49] Yeah. I think for cooking there, making it an educational game seems to me to be the most intriguing. You know, like, I like your flavor one, I'm, I'm hyper interested and we should talk more about this, but I, I, I've got a, a daughter that's almost two years old and I decided that what I was gonna try and do was introduce her to as many smells from spices as I could and then try and figure out how do you create a language around smells and tastes because we don't really have a language around it. You know, we have sweet, sour, spicy, but that's about it. And this has been a, a hell of a journey for me because, you know, how do you describe the smell of oregano? It's very, very difficult. I How would you go about tackling this?

Albert-László Barabási [00:44:35] I I need to recommend you a book offline. I'm just reading one, it's a little booklet that from the 1800, actually 1700 originally that, that for every color has a description. This was before they could print color and this was the book that Darvin took on his voyage to actually describe the colors of the plants and the animals that he saw. It's a beautiful little book, you know, like most a hundred pages. And then later on they started at the colors to what they refer to. So the new additions they have it. So it's a beautiful question for us, for humanity of how we use one information carrier, which is language to describe a totally orthogonal information carrier like smell and taste. And, and as you said, our vocabulary when it comes to smell and taste, and particularly the smell is very, very limited, right? Much more taste. I think we have a wider palette, but when it comes to smell, it's very, very limited. But we, maybe we can get inspired of what the color people have done, right?

Vance Crowe [00:45:42] I, I think it's, I mean, so, you know, my 2-year-old can't really speak very well, right? But I can tell by the look on her face if she's surprised. So one of the things that's happened in the last, I don't know, two months is I can tell her you're gonna, like, I have her close her eyes and then I tell her, okay, I'm gonna put garlic in front of you. Can you picture what garlic's gonna smell like? And then I put it in front of her and she can't hide if she's surprised or not. So if she, if she smells it and she's surprised, you know, that she didn't get it, but sometimes she's not surprised. And so it's, it's amazing like how you, like my child becomes my little experiment here, but being able to put those words to it seems to me to be able to open up and, and access ideas and prediction models that you can't have if you don't have, maybe you can if you don't have words, but I, I don't, I don't, I I certainly didn't have it. I don't have it without words.

Albert-László Barabási [00:46:37] And it's a beautiful experiment what we're doing. It reminds me always my wife's recommendation without children, she always said no experiments on children.

Vance Crowe [00:46:48] I mean, kids are experiments, right? She picks up the glass and drops it and

Albert-László Barabási [00:46:52] Yes, we're, yeah, we, kids by themselves are giant experiment that none of us are prepared for. And it's amazing that we need to have a driver's license to drive a car, but we don't need one to have a child. That being said, I, I love your project. So one day you should write the book about how children actually perceive smells. And I'm ambi also curious like whether any of them will come back to them later when they start choosing their own foods.

Vance Crowe [00:47:21] Yeah, that's, that's exactly what I want her to have access. 'cause when I was growing up, right, the spice rack in my mother's thing was like a giant thing of cinnamon and garlic and like maybe some oregano, but that was about it. But now today we have access to so many flavors. So can we talk about your flavor project and what, what you discovered in all of that?

Albert-László Barabási [00:47:41] Yeah, so that was interesting because as said that it was, it was given by, it was started by the opportunity that was a book, it was really a physical book. We're gonna get a digital version that says these, these ingredients have these flavor chemicals on them. And we ended up pretty much typing that up because we couldn't get a digital version. And, and then we, we asked it to say, you know, is it really true that ingredients taste well together if they share chemicals? And, but the question is how do you know that? Right? So this is the map that, that has that information, but we have actually a more detailed map that I'm happy to share with you. So I'm gonna share screen and show you that map. Right? And let me just blow this up and, and here we are. I'm gonna go to our art catalog. I was there. Okay, here we are. So that's kind of the flavor network, what you see now. So each node corresponds to a chemical like cilantro and it's connected to another in the ingredient and it's connected like pineapple and strawberry because they share not one, but multiple, multiple flavor chemicals. So,

Vance Crowe [00:48:56] So for people that are listening to this, I'll just describe it if for what I see, imagine that you are looking at like the map of a brain where there are all these little dots and they're interconnected. And on this particular thing that he's showing, they're different colors and I assume the different colors. Well, you tell me and there's like really tight networks and broad, so

Albert-László Barabási [00:49:15] Yellow, yellow are all fruits, right? These are the, the dark red is all different wines here. The lighter red is all different type of meats. The greens are all different type of vegetables, onion, garlic, shallot lee, right?

Vance Crowe [00:49:31] Oh, fascinating.

Albert-László Barabási [00:49:32] So these are different food categories that we eat and, and then they are connected by links if they share one or multiple flavor chemicals. And then the question was, do we really prefer to combine things that are connected on this map and think, or things that are not connected on this map that do not share. And well, one way to test it is go ahead and make all of this pairing and then have somebody taste it, which is not the type of, not the type of research we do in my lab. What we did instead is that we relied on all the national cuisines whose recipes are available online. And then we asked, I asked, is it possible that the, in the Italian recipe recipes, they tend to combine more often things that are actually connected in this map compared to those that don't. And the answer is yes, particularly in Italian and the Western European cuisine that we really, people really try to combine things that share flavor chemicals. And the best example to my mind is lasagna, where every single ingredient other than salt shares some other chemical flavor chemical with the other ingredients. But it turns out that that's only true for the Western European, and by definition the, the American cuisine world, the Asian is the opposite. The Asian is yin and yang. So they actually try to combine things that do not share chemicals that are in some way complimentary to each other.

Albert-László Barabási [00:51:04] So, so in that way, the research kind of helped us to really pinpoint these subtle differences between the different type of national cuisines. And here you have the two dimensional version of that. What you see behind me is a three dimensional statue of the same data. You see exactly the same data, but now rather in a, in a three dimensional form, much rather than a 2D.

Vance Crowe [00:51:27] So this is fascinating and this is a part of the artwork that you do. And I, we can, I would like to talk more about that, but I, it makes me think you must have some intuitions about what human beings ate before the modern age, right? We've only had broccoli for 6,000 years, right? But people were cooking or, or combining foods. What did, where, what were humans like, what were they eating and how did they come to these discoveries?

Albert-László Barabási [00:51:54] Well, that's a great question. I'm always puzzled what Europeans were eating in particular because much of the staple food in Europe is now based on American food. That is that, you know, from potatoes to tomatoes, nevermind coffee, you know, it's all food that was imported after Columbus from America. So including corn. So I'm always puzzled of what was really the European diet. The diet that we have today is really kind of the product of the 17 1800. That's when we start talking about national cuisines. That's why we're talking about much more kind of, even the French food that we see today had no resembles The French food that pe the French people eat in 1600, the 1600 French food was much more like today's Mexican, lots of sauces, everything mixed together rather than this clean ingredient based food that characterizes the French cuisine today. So, so that has drastically changed with industrialization, with the emergence of, of the imported foods. But on the other hand, what is important and may be important for the last issue that probably we should talk about, which is processed food, is that we were eating lots of raw ingredients. And one, one, what is important about raw ingredients is that we often don't think about it, but the chemical engine of a plant is the same as our chemical engine. It's like we're all running on gas, right? It's a carbon based metabolism that is identical with ours, right?

Albert-László Barabási [00:53:30] So, so it means that plants and animals have the same composition, the same nutritional balance, and the same content as we do. And I don't don't wanna say we're eating ourselves, but we do that right. In a, in a chemical way, right? And, and that's what has really drastically changed in the last 50 years with the emergence of food processing and ultra processing, where suddenly the food that we eat doesn't resemble us any longer.

Vance Crowe [00:54:08] I I So say more about when you talk about raw food, right? You're saying food that you just pick right off of the plant or that you eat the seeds right off of the tree?

Albert-László Barabási [00:54:19] Yes. So as this is actually a journey where we're very much part of it right now is that we wanted to understand not only, you know, what's in the food, but how much of those chemicals, what is the composition of each of these chemicals in the food. And even if you just take the nutritional components that, like if you look at the hundred gram of food, be that apple, be that meat, the range of chemicals, the range of compositions, or very order nine orders of magnitude. And what I mean, when you look at vitamin E, you have about 10 to the minus six grams per a hundred gram fruit. And when it comes to water, it's about 70 gram water in 9,000 gram of food, right? So we have actually about nine orders of magnitude variations in how much chemical, how much of one chemical is in a given amount of food. And this is not by chance, you cannot have 10 to, like, you cannot have 70 grams of vitamin in your food because no plant can generate such a high density of chemical E because the metabolic pathway can only generate only that tiny amount of vitamin E Right. Given a hundred gram of plant. So, so, so how much chemicals, how much, how much nutrients are in each of the food is fundamentally determined by the biochemical engine that is, is roughly the same both in plants and in humans.

Albert-László Barabási [00:55:50] So carbon based metabolism with small variations. So, so there is a very rigid architecture of what normally gets into you and what proportions and the proportions really matter. Let me give an example. So if I'm gonna give you 10 hydrogen atoms and one oxygen ato and I say, can you make me water? You would say, sure, I can make you one water, water molecule. Right? And then I don't know what to do with the rest of the hydrogen atoms because I'm gonna have eight, eight hydrogen atoms left over that I cannot combine it with anything. Right? So, so if we eat food that is natural, we get things in the same proportion that natural created and we were co-evolving with that nature, right? And, and so therefore it's the same metabolic, same carbon, carbon-based metabolism. We know how to absorb it. If, however, I give you chemically unbalanced food that has many more of certain chemicals than the others. Traditionally we thought it just goes out. But it doesn't, it deposits like cholesterol, right? If you have too much of it or fat, it doesn't simply go out. It gets into your bloodstream and it gets deposited the places that it has no place to be. Right? And that's where food causes disease. So in a way, one of the reason that over and over we go back and, and, and nutritional research and medical research goes back to the discovery and they, every generation discovers itself that natural food, meaning that whole food that is, that, you know, like raw food that is you're processed in your kitchen is the healthiest for you is mainly because of the fact that that has the same chemical composition as we have co-evolved with to really absorb over tens of thousands of years.

Albert-László Barabási [00:57:46] But something has changed in the last 50 years, which is UTA processed food that does not have the same chemical composition any longer as natural food. And when we look for example, at wood, ultra processed onion rings, for example, there are a bunch of chemicals all over the place. The ratios are totally messed up. 'cause that's what Woodrow processing does, right? They try to create the same, well

Vance Crowe [00:58:12] Let's define ultra processing.

Albert-László Barabási [00:58:14] Yes. So that's a very good question. So raw food is pretty clear, is what you actually get in the vegetable aisle of the, of, of your supermarket processing is what you do with, because we never eat raw food at you peel the oranges, right? You squeeze the orange juice out, you cut the tomatoes, those are processing, right? You can also freeze it, you can also boil it, you can also cook it and do other things. These are things that you can actually do in your home. And there are part of normal processing steps. And yes, of course if you burn something can have cart genetic effects, but typically you're not fundamentally altering the chemistry of the food. The, the, the nutrients and the chemicals will come in the same proportion as in the raw food. Ultra processing is when you take the food and you deconstruct it into chemical components to reassemble at a certain mo at this later moment. And the best example I can give you is orange juice. You may have actually, you know, natural orange juice in your supermarket shelves, right? And they may even say it's organic. But what they don't tell you is that when orange is being collected in Florida or wherever it's being collected, then the juice is being squeezed out and then it's being broken into three different components. The water is taken out, the pulp is removed, and the juice is actually pretty much dehydrated. And then these three components are being recombined at some other location, often hundreds of miles away, you know, into an orange juice that arrives to your supermarket.

Albert-László Barabási [00:59:51] But in order for that orange juice to taste like orange juice, you need to add lots of flavor chemicals because once you go through the dehydration and the separation of the pub, it doesn't taste, it doesn't feel like orange juice anymore. Even if you remix it that way. So then you start adding flavor chemicals and texture chemicals to create the illusion of the orange juice that's ultra processed food for you. And 70% of the calories that are available in the supermarket, 73 are actually ultra processed. And 60% of the US populations calorie intake on a daily basis is coming from ultra processed food and it's killing us.

Vance Crowe [01:00:35] I'm, I'm, as I'm hearing you describe this, I'm thinking about the statistics on COVID, right? The 80% of the people that were hospitalized were obese. And you have to figure that a huge percentage of the, the people that are obese are eating ultra processed foods. So it's amazing to think about how those two things could be highly correlated with one another. The fact that we have highly processed foods and the vulnerability of people to be hospitalized for COVID. I I had no idea that's where this conversation would take us today.

Albert-László Barabási [01:01:04] It it is, it is a very strong effect. And it just simply, it simply means that we have lots of people who have lots of health problems and COVID was very effective among those. And you are right. Much of the obesity is really based on ultra process with ultra process do there has lots of things that we know that it does to us. One of them, it actually makes us to co to consume more calories. So there are actually studies to show that if you give the same Putin ultra processs on unprocessed version to people or simply processed, they tend to consume more to more of the ultra processed one and get more calories. Partly because they've been optimized for us to like it, right? With the extra chemicals and the balance of the chemicals. But most of the effects of ultra process would, we don't really understand, and I would be lying if I say we do what we know however, is that about 10 years ago, a Brazilian group has started to categorize foods very carefully into these classes of UAR process, unprocessed or raw food. And when that data became available to researchers, they started to do redo the epidemic studies to say where do the health effects, for example, as associated with red meat come from, right? We know that red meat has lots of consequences of cardiovascular diseases. And what the studies show that raw meat that is unprocessed meat had no effect whatsoever in cardiovascular diseases. It was the ultra process component where the lion's share of the health effect comes from. And now we have lots of data in the last five years to show that if you take 10% of your calories and you get it that you normally get from Whole Foods, then you replace it with ultra processed versions of it, it would read about 13% increase on chance of cancer.

Albert-László Barabási [01:02:44] It would increase by 12 chance, chance of obesity, cardiovascular disease, and all the ga. So ultra process foods seems to be one of the reasons that we have so many health effects in the United States. We do not fully affect, understand its why and how it does that. We suspect it's because of the chemical imbalance and the additives and many other things. But we know that it affects us. And, and that's what we're kind of involved in right now. One of the biggest problem is with the ultra processed food is that we know it's not good for you, but it's much cheaper than the non-processed food. Right? And most important, you as a consumer, you are not empowered to decide when you take something off the shelf, whether that's ultra processed or married processed, that data doesn't exist now.

Vance Crowe [01:03:35] Oh, that's interesting because like for example, the orange juice, you have no idea that it was dehydrated and then put out and, and you can understand why the person producing orange juice or the company producing it would do it. But you can also say, well, it would be very valuable for consumers to be able to make that as a part of their buying decision.

Albert-László Barabási [01:03:56] Yes. And, and this is something that, that we realized that that's the major problem. So one of the things we've been working the last two years and its product just came out right now is that, and I can share a screen to show you, we actually put out a website more like a research website rather than where you can actually put in any type of food. Let me put goldfish, right? And it will tell you how ultra process is and, and what are the alternatives that you can have, how you pick one. This is actually a non ultra processed version. And as you go towards more and more at bars in goldfish, you can find things that are, get a score of 99 hugely alter process. So look at this Walgreens

Vance Crowe [01:04:40] Here, pull it up on the screen. You were saying you wanted to share it.

Albert-László Barabási [01:04:42] Oh, I did not share it. Apologies. I thought I did. So I'm gonna do again the find you first. Oh, I see. I have to click share. So can you see me now?

Vance Crowe [01:04:59] Oh yes.

Albert-László Barabási [01:05:00] Okay, perfect. So here is, so we just put out this research tool, we call it true food tech. And I just typed in goldfish into that. And it's telling me that in an odd way, the the goldfish that has pizza flavoring is, is not, is very little processing in that and the goldfish that says on the right hand side whole grain, right? It's hugely ultra processed. So, so, and if you, we actually showed the chemical ingredients, it's full of additives and everything to kind of enforce it. So, so our goal was to put at least for the research community into the consumers a tool out, it's an AI-based tools that looks at the nutritional components of the food that must be actually reported by law for any food. And, and the AI is very good at detecting changes in the chemical composition that would not be occurring in natural food. And it's a signature of ultra processing. So if you go to the next store next time or before you go shopping, if you wanna check out the brand that you naturally buy, whether it's UAR process, you can do so. And this website will also tell you that in the same store, if there are other brands who have much less processing, because one thing we actually showed, and that was really interesting for me. If you take people and just look at their daily diet, we eat about 20 items per day. You know, when you put your diet together and you just take one item that is the most ultra processed calories that you consume per day, and you replace that item with the same item, but the less processed version that leads to a 10% decrease in your chance of getting a metabolic disease, which is effectively a diabetes and its versions.

Albert-László Barabási [01:06:52] So just taking one item on your diet and replace it with a less processed version could actually have a long-term health impact.

Vance Crowe [01:07:00] Well, who in their wildest dreams would've imagined that Laslo bari the network scientists would be working on this project. But this is amazing. I had no idea the conversation would go here. Remind me one more time what the website is for this so people can check it out.

Albert-László Barabási [01:07:18] Sure, it's true. Food tech, TECH, the ending. That's kind of the trick to it. So true food is obvious, right? The Dark Tech is perhaps showing that we're coming from a tech environment to do that.

Vance Crowe [01:07:32] I'll put it in the show notes. And if people wanted to see some of the artwork that you did on the flavor profiles and read more about your work, where would you send them?

Albert-László Barabási [01:07:40] Well, it is the bar bahi lab.com. Right? And if you're interested, click on the art tab, right? Rather than if you care about food, click on the science lab part. If you care about art, click on the art tab.

Vance Crowe [01:07:56] Well, Laslo, this has been a true honor and a really excellent conversation. Thank you so much for coming on. Thanks

Albert-László Barabási [01:08:02] For giving me the opportunity. Happy to continue at some other time.