 So the idea that we are what we eat is certainly not a new one. Archimbaldi painted this painting along with others, depicting this idea back in the 1500s. And in 1863, the philosopher Feuerbach came up with this pithy quote, der Mensch ist Wasser ist, which is a great pun in German, what he's saying is that man is what he eats. And so this idea that what we eat is a tremendous impact on things like our weight, on our health is very well established. The new frontier is understanding what impact it has on our thoughts, on our mood, even on our memory. And Jeff Bland at the Personalized Medicine Lifestyle Institute, I met him back in 2016, and we've had a very fruitful and ongoing dialogue since then. And he has this quote that I really like where he says that food is a language that speaks to our genes. And what he means by this is that our genes, obviously, are fixed when we're conceived. They stay the same throughout our lifetime. But how they're expressed and the impact on our physiology is tremendously variable through this process called gene expression. And so the idea is that food speaks to our genes and causes changes in our bodies and maybe even in our minds. Now the language of food is very much a language of color. And I'm sure you've all heard the admonition that you should eat the rainbow, right? And if we consider a rainbow, a brightly colored fruits and vegetables, which we all know that we should be eating, all of the colors here, everything from the carotenoids that make the carrots orange to the anthocyanins that make the blackberries black, and all of the other natural compounds that are in these foods turn into, translate into chemicals that are changing how our bodies act. And so when we're thinking about this language of food, speaking to our genes, we really want to think about the colors in that food. However, untangling this language is incredibly challenging. And this heroic study at Harvard is still one of the largest of its kind, where they looked at 120,000 men and women over a period of 20 years, so 1.5 million person years of follow-up. And what they looked at was the impact of each food item on what made these people gain or lose weight over that 20-year period. And so what did the study find? Well, I've got to say that the heroic effort that went into it in large part was extremely consistent with, I bet, what your stereotypes are of what thin people eat and what fat people eat. Because if we look at the food that they found consistent with weight loss, what it was is yogurt, right? And that seems pretty healthy. You see a lot of thin people eating yogurt. On the other side, the food that was most associated with weight gain was potato chips. And again, this makes a lot of sense, right? We don't think of potato chips as being especially healthy. But what astonished me about the study was the size of these effects. So people who ate one extra serving of yogurt a day over the course of four years, on average, they lost 372 grams. Whereas in contrast, people who ate an extra serving of potato chips a day over four years, they gained 768 grams, right? And that's not a lot over four years, because what it means is that if you exercise the right choice 1460 times in a row, every single day for four years, you're going to change your weight by just 1.14 kilos. And that's not very impressive. But what it's important to keep in mind is that this is an average and everyone's different. So some people are making the substitution. They're going to lose a whole lot of weight. Whereas other people are going to gain almost as much weight. And it almost all averages out when you look at a large population. So another study, the SMILES study, was published by an Australian group in 2017. And what they were interested in was not weight, which is pretty easy to observe, but mood. And so they were basing their study on a whole lot of different studies that had shown associations between diet and depression. And in particular, people who were depressed tended to eat more high sugar, high fat foods. And they tended to eat less healthy foods like the Mediterranean diet. And so what they did is they put thousands of people who had been diagnosed with major depressive disorder either on a Mediterranean diet or on a different diet intervention as a control. And what they found was that when they put them on a Mediterranean diet, what happened was that 32% of the patients recovered got better from their depressive symptoms. And this is amazing, right? Because this is comparable to what you'd get with drugs like SSRIs and other antidepressants. And one thing that they were concerned about was is it really the particular diet that's having a major impact here? Or is it just the attention that you pay to someone when you're talking to them a lot about what they're eating? You're giving them support in their diet and so forth. And so as a control, what they did is they looked at a low-fat diet where the same amount of care and attention was put into making sure people ate a low-fat diet. And in that low-fat diet group, only 8% of them recovered. So this was a huge difference, right? Two diet interventions. One of them, the Mediterranean diet, enormously effective for helping people recover from depression, whereas the other one, the low-fat diet, almost completely ineffective. So the incredible diversity in the microbiome is already being used to make highly individualized predictions in the area of blood glucose control. And so in this remarkable study from a group at the Weissman Institute in Israel, what they did is they wanted to understand why is there so much individual variation, one person to another, and how you respond to the same food. And so what they did is they took 800 people and hooked them up to continuous glucose monitors, and then they fed them a very carefully-defined sequence of diets for two weeks so they could isolate the impact of every single item those people ate on whether their blood glucose would spike after their meal or whether it would only rise a little bit. And so what they found is that when they averaged the results for all of those people, they almost perfectly recaptured the published glycemic index values. So in other words, across 800 people, they got pretty much the results you would expect for each of those food items based on decades of research. But what was amazing was the difference in individual glycemic index. So the exact same food that would send one person's blood glucose haywire would be healthy for another person. And a few of the examples that they found were really surprising. So for example, tomatoes are a pretty healthy food for most people, but they had one person in their study where every time that person ate tomatoes, their blood glucose would go up the charts. And so when they cut them out of their diet, they did much better. And then similarly, beans are another food that's generally healthy for most people, but for some people, probably because of the lectins, they make blood glucose spike. And so by being able to find these particular food items that are good or bad for you, you can really have a major impact on your blood glucose control. And one of the things that they found is that the results were so reliable that they could take another group of 100 people, and for each of those people, they could design a good diet and a bad diet where both of the diets have the same calories, they have the same macronutrients, and what was different about them was relatively small substitutions. Like, are you going to eat regular potatoes or are you going to eat sweet potatoes? Pretty small changes, but very big effects on whether your blood glucose stays under control or whether it goes all over the place. And they mentioned a whole lot of different things about the people in the study. What they found was important, what they found was important out of all of these different things was primarily the gut microbiome. And what was amazing, and so what was amazing about that was the impact of the microbes in the gut. The healthy med study is another one where, again, they compared the Mediterranean diet to a social group intervention. And so what they were trying to figure out here is if you already have major depression, if you're on the Mediterranean diet, or if you have a social intervention, which of those improves your mood more? And so what they found is that both of the groups got quite a lot better from baseline to three months. But the Mediterranean diet actually made a bigger difference than that social group support. And so what's amazing about this is that changing what you eat can have as much of an impact and even greater impact even on your mood than getting the kind of social support that's more traditional for depression. And so, again, though, individual variation was really high. What you can see is a lot of variability in each of those groups. And so trying to understand the sources of those differences for each individual is like trying to find a needle in a haystack out of all the possible different causes. So to explain individual variability, there's been a huge amount of work trying to find the key to these links between diet and health, including mental health and the human genome. And if you consider the human genome project, that project was conducted at an expense of about $3 billion over more than a decade. And so to this day, we've spent billions and billions of dollars looking for the key to the links between diet and health in the human genome. But so far the results have been very unimpressive and that leads me to wonder, are we even looking in the right place or are we like the proverbial drunk man who's looking for his keys under the streetlight, not because that's where he lost his keys, but because that's where the light is. And so what I start to wonder is it possible that the key is not in the human genome, but rather is there some other collection of keys in a completely different location that we're missing entirely when we focus on human genes? Now, what I'm thinking of is the discrepancy between what's in our so-called human genome and the rest of the genes that are associated with our body. The human genome project mapped the 20,000 human genes that we all share where we're all 99.99% identical to the person we're sitting next to in terms of those human genes. But astonishingly, it turns out that the microbes, mostly the bacteria in our gut, have a much larger gene catalogue of 2 to 20 million microbial genes. And so when we just look at 1% we're ignoring 99% of our genes. And astonishingly, the 99% that we're ignoring are the ones that we can change and take control of over our lifetime. So coming back to the Israeli study where they were looking at the microbiome sources where they were looking at the microbiome and its ability to explain who should eat what diet, one thing that they found that was just amazing to me is that they found that what you should eat is a completely different person to person and that they could explain it. So for example, you probably think of rice as a pretty healthy food, and for most people it is. But they actually found that for quite a lot of the people in this study, it was worse for them to eat rice in terms of their blood sugar than it was for them to eat ice cream. And out of all of the different things that they measured to try to explain this, what they found was that the gut microbes were the deciding factor whether you should eat rice or whether you should eat ice cream to keep your blood glucose under control. And so on learning this, people typically have two questions. The first question is, is there a test I can do that tells me am I in the category of people who should eat rice or the category of people who should eat ice cream? And in fact there is such a test and they've commercialized it through a company called D2. But more interestingly, you might ask, suppose I find I'm in the category of people who should eat rice, could I change my microbiome so that now I'm over there in the category of people who should eat ice cream instead? Or perhaps can I go the other way? Perhaps my microbes are affecting what I want to eat. Could I change my microbiome so that I no longer crave ice cream and that I crave rice instead? And that's a much more interesting question. Now, that example might seem a bit frivolous, but in the case of cancer, it's deadly serious. And I'm sure you've heard of checkpoint immunotherapy, which was one of the most exciting treatments for melanoma that's available now. And Jennifer Wago at MD Anderson Cancer Center has been able to show that with a dietary intervention, you can take people who are not going to respond to ketruder, one of the most common cancer immunotherapy drugs, so they're going to have no effect if they take it, and that's because of their microbiome. But they can change their microbiome to go from the non-responder category into the responder category where the drug will change their lives. Where the drug will save their lives, rather. And again, it turns out that the difference between the non-responders and the responders is in the particular microbes that they have in the gut. However, you have to be careful with this, although fiber is really good for reshaping your microbiome so that you can respond to cancer immunotherapy. Probiotics are not. And what they found is that the patients who were taking probiotics actually died sooner than patients who were not taking probiotics. They were able to reproduce this in an animal model by transporting both cells from the tumor and the microbes from the gut from the patients into mice, and they showed that when they gave the mice probiotics, the mice also died faster. And so this is something that's fairly generalizable and the mechanisms are being worked out why probiotics will actually not help for cancer immunotherapy, although they may be useful in other circumstances, like, for example, post-antibiotic diarrhea. So the real problem to all this is the microbiome is incredibly complicated and we need a better user interface for our microbiomes. So you could imagine, for example, when you're in the supermarket and you're faced with a thousand different kinds of products and you have no idea what you should eat, how important this need for a better way of addressing that is. And what we're imagining, I'm sure you've all heard of large language models like chat GPT, what we're imagining is something for your gut, let's call it Crap GPT, where you could ask your phone directly, scanning the barcode of a product, should I eat this to improve my health and get some sort of natural language response to it? And then you could imagine taking that a step further. What if you could also use the camera in your phone to highlight the products on your shelf according to which of the ones that you should eat based on your individual microbiome and its response instead of the products that you're looking at right now? And that's the kind of crazy thing that we're up to at the Center for Microbiome Innovation, where with over 140 faculty members and a growing list of corporate partners, we're trying to address the hard problems like this. So where we are at the moment, we just don't know if we're going to be able to do this kind of thing, but a lot of the pieces of it exist somewhere in the university and the potential is absolutely enormous that by being able to understand this huge individual source of variation, where unlike your human genes, where you're 99.99% identical to the person you're sitting next to in your microbial genes, you can be 90% different. And so taking account of all of those genes and where the variation actually is to improve your health over a lifetime, we think has an incredible possibility, not just for things like your weight and your physical condition, but also for your mood, for your memory and your ability to have a healthy brain into advanced age. So I'd like to end by just asking you to consider the potential of this being your brain on plants. If we can find out the right dietary information for each individual and have a microbiome-based test that gives you near-real-time information so that you can use it, we think there's incredible potential to use all of this information to reshape your health over a lifetime. So with that, I'll end there. I'd be delighted to answer any questions that you have. Thank you for coming along this evening. Questions, yeah? So the linkages between the microbiome and linkages to neurodegenerative conditions like Alzheimer's disease, could you talk a little bit more just about the relationship between diet, microbiome, and these types of diseases? Sure, absolutely. So the link between diet and Alzheimer's has been known for a while where things like sugar and soda tend to be associated with worse memory problems. And then on the other side, things like fish, like mushrooms, like strawberries tend to be associated with better memory. And in a study that we've been doing, the Alzheimer's gut microbiome project funded by the National Institutes of Aging in the United States, what we've been doing is we've been piggybacking on a whole lot of clinical intervention trials where essentially different diets that are known to impact blood glucose, which is itself linked to Alzheimer's, like the Mediterranean diet, the ketogenic diet, and so on are being tested on people who are at risk for Alzheimer's, and we're looking at the microbiome and metabolome. So what's been fascinating there is that we've been able to recapture by automatically taking molecules and looking at how they're linked to diet. We've been able to recapture with just a couple of days in the lab, decades and decades of studies finding these individual links between food items and disease. And so where we're going with it is to take the next step, not just at the whole population level, but at the individual level, how are those links different one person to another? And how we can give people very precise and actionable advice about what they personally should do, because maybe it's strawberries for one person, but for another person, that will have no effect and you should look at mushrooms or fish instead. Yeah. I think they're running around with a microphone, so if you could just wait for that for the recording. The microphone is an amazing presentation. Thank you. Have you been able to quantify which bacterial specimens are more important within the gut microbiome? I mean, is it the diversity or is it the Bephido kind of bacterium? I mean, is there any specific inflammatory markers which are specific to certain strains per se? That's a great question and let me just unpack it a little. So we do find particular species, strains, even higher level groups like genera that are linked to either better or worse memory deficits and also linked to responses to particular food items. It is a very complex picture that's not easy to summarize though and part of it is that the same bacteria seem to have different ecological roles and different people, so what we really need is a better way to understand when the same species of bacteria is doing a different job in one person's gut versus another, which can be dependent on things like your diet that seem to make a particular species either good or bad depending on the other conditions and also what other bacteria you have. And this is true both for beneficial microbes and for pathogens. So for example, in the Human Microbiome Project, we found that between 5% and 30% of people, depending on which pathogen it is, can be walking around completely healthy with something that we normally think of as disease-causing and so whether it actually does cause disease depends a lot on other factors like your immune system, the other bacteria that are in there and crucially your diet. Thank you. I have one last question. Is food to blame? I mean, because most of food that we eat is processed and I do have a correlation where a lot of my cousins and relatives have actually migrated to the U.S. and after 10 years of living there, they actually tend to get all the allergies in the world, not allergies, dairy, all the different allergies on earth. And when they were staying in India, they were actually completely okay and I presumably think it's because of the food they're having which is a lot more processed in the U.S. versus a lot more natural within India. Is there any studies that you've done personally which indicate that? Yes, so again... And I do have one more thing. So even from an alternative medicine perspective, food is certainly the most important. So be it Chinese medicine or the Indian Ayurveda, when they say food is what affects the entire thing, which again was... Yes, so there have recently been some quite strong links to ultra-processed foods as a category and Alzheimer's and I'm certainly sorry to hear about your family's experience moving to the U.S. which is an auto-common experience. What... There hasn't been nearly as much research for a particular food item. Is it good or bad depending on how you process it? The older questionnaires just talk about how much of each kind of food that you eat and not how it's processed. The newer questionnaires like we're using, for example, in our collaboration with Denone, where we're looking at food consumption and the impact on the microbiome across five countries. In those studies we're trying much more to capture how it's processed as well as what it is. But those studies are much less frequent than just adding up how many potatoes versus how much cauliflower versus whatever you ate. Whereas we think from a chemical perspective it does make a huge difference how you process it and especially if you're stripping out a lot of the beneficial molecules from it in that processing workflow that could have a huge impact on whether something we think of as healthy is still healthy by the time you consume it. Thanks. I have a sort of practical question. First of all, fascinating presentation. I was looking at all the colours you have. It's amazing. So, you know, there are all of these scientific studies and they're very convincing but is there any way to sort of like implement this practically in our lives in the sense that, you know, not everyone has access to places where they map out the genomes and sort of figure out which foods are particularly suited but is there a way of trial and error that people can incorporate into their daily lives to see how they react to certain foods to remove or add certain foods? Yeah, well that's a really important point that not everyone has access to the genome or the microbiome sequencing technologies, although you should be aware that they're getting cheaper dramatically and over the last 15 years there's been about a mid-and-fold decline in the cost of sequencing. It is very likely that there's going to be another very rapid rate of decline due to some changes in the IP landscape just recently. So, part of the answer is that I think that genomic sequencing is going to get cheaper and cheaper. Part of the answer is doing this kind of thing empirically where you're going through a long sequence of adding things and removing them from your diet. You can do that, but the responses tend to be pretty slow. They will often take weeks for months, weeks to months, especially if it is for something like a psychological condition. And one of the big problems with studies on weight loss is very frequently it takes months to years for a particular substitution to have an effect over the long term. And so, the process of trying to do it yourself may wind up being surprisingly expensive when you think about the time and the cost of making those substitutions over time. So, we think of the genomic technologies as actually a shortcut where especially as they become more democratized and more accessible, it will be much more convenient to do that test and get an answer right away rather than to go through this month-long sequence of trying to do elimination diets. For example, my mother who was diagnosed with IBS with irritable bowel syndrome a while ago spent a long time going through elimination diets like the FODMAP diet, for example, and just seeing that incredibly time-consuming process and the frustration of, well, what if you went out to eat and accidentally ate something that you weren't supposed to? How much of that can mess things up? Being able to eliminate that uncertainty for people, I think, would be really valuable. Thank you very much. I have a... Is it commonly available like these tests? Well, D2 was doing a direct-to-consumer model but they're not doing that anymore. Now they're trying to partner with insurance companies and large organizations that maintain health. So, right at the moment, the problem is how do you come up with a good business model where you can offer it... Can you really offer it direct to individuals? Or do you need to do it with larger organizations to get the volume up? That may change with changes in sequencing technology. So right now, you need thousands of samples to process at once to make it cost-effective because of the way the sequencing is done. But if you could sequence each sample cheaply as it came in, that constraint wouldn't exist anymore and it'd be feasible to do a lot more of this at the individual level. What our real dream is is a smart toilet where you can get some kind of instant readout, deliver it to your phone and use it right away. But that's further on the horizon and it's not going to be DNA sequencing. It's going to be some other technology like maybe mass spec, maybe Raman spectroscopy. Something that you can do very rapidly rather than something where it takes hours to do the sequencing and then takes a long time after that to do the data analysis. I have a last question if I'm allowed. You're first. I didn't. I learned something. Sorry, go ahead. You have some beautiful pictures about fresh, all kind of fresh produce. I'm deploying a global network of multi-localized food production hubs based on powders. 40% of food production is powder and you only had one photo with powder. How do you see the future of food? Because it's obvious. When there is fresh, fresh is great and it's a Mediterranean diet but the other truth is that 40% of the world's food and having a 30% food loss do you see the future combination of the right food being something for example where we have a Spanish broccoli and then we pull some like one small bag of what we have like sugar obviously of a mix of different ingredients that can be dried and powdered to personalize nutritional product that we are striving for? I'd love to talk more with you about that network and you're exactly right getting the benefits of fresh food just isn't something that's scalable to the whole world and in the past studies of things like canned food, frozen food restried food have suggested that a lot of nutrients are preserved. What we don't know is whether the active ingredients for the microbiome are preserved in that kind of process and it would be interesting to do. We have been studying centenarians in different parts of the world and when we match up the molecules in food to the molecules in the body, what we see is for example when we study centenarians in the Chiliente Peninsula versus people in the US the molecules in the Bodstream and the molecules in the stool are much more explainable if we have locally matched foods and we can explain three times as many molecules if we're using population matched diets and there's also that fascinating question why is it that the people in these blue zones like Okinawa, Sardinia, Loma Linda Nicolae and so on why are they so healthy? Because if it's something about their human genes there's nothing you can do to really export that and if it's something about the environment it might be really hard but if it's something about the food and you can understand what the active ingredients are and whether it's the same for everybody or whether it's localized to that region which I think is what you're telling me your network is about that would be amazing in terms of taking what we know and generalizing it and the American Gut Project one of the things we found is the number of different species of plants and personates has the biggest impact on their microbiome of any of the variables we looked at and so if you could assemble that from powders rather than a miserable track through the farmers market it would be incredible. And you would love by the way that we have digitized more than a million good food products so we have actually connected the lab with near infrared spectroscopy so actually to characterize the powders and ensure they're always coming in and the right quality like the last one we have been using instead of the quantitative the qualitative model that is basically the fingerprint model so we can compare ginger from India to China and so on so in line with what you said about characterization as a small let's say medium-sized company we are actually using cheaper and cheaper technology to do that. That's fascinating I'd love to talk more about that after the session. I do want to get to this one last question because you've had the microphone for a while and I'm getting the signal to wrap up the session. It's maybe a quick one and it was just to ask is the results of what you're seeing is it indicating that it's path dependent or is it kind of just a point in time because I guess a little bit linked to Vivian's question. That's a great question but we don't have enough data to answer it. In studies in mice where they've been overfed so they're obese and then they slim down again the microbiome has a memory of the obese state and so a subsequent challenge has very different impacts based not on the current state but on the path that it took but as far as I know there's absolutely no data in humans and that's especially important given the pervasiveness of binge dieting especially in the US and that is something that definitely needs to be addressed but there's been very little research so far so great question very little data though. Thank you very much. Thank you. Alright well thanks everyone I think we need to get out of the room but thank you all for coming along and for a lot of great dialogue.