January 26, 2019
[0:00:50] An All-Encompassing Platform by Onegevity Health
[0:03:33] Guest Introduction
[0:05:34] Podcast Sponsors
[0:08:15] What Onegevity Is, And What It Does?
[0:11:03] How Is Onegevity Different from Other Biome Testing Companies
[0:17:25] What A Metatranscriptome Analysis Is
[0:20:10] The Actual Testing Process at Onegevity
[0:27:13] Skepticism in Their Testing Processes
[0:32:29] Podcast Sponsors
[0:35:56] The Role of Artificial Intelligence in The Process
[0:40:02] How You Go from Stool to Shotgun Sequencing to “Don't Eat Green Beans”
[0:46:38] Dietary Recommendations after a Onegevity Test
[0:52:00] Other Technology from Onegevity
[0:54:45] Data Security with Onegevity
[0:57:23] Exercise or Lifestyle Recommendations
[1:01:20] What the Onegevity Platform Looks Like from A User Standpoint?
[1:06:07] What is Not Included in Onegevity?
[1:11:47] Onegevity Website
[1:14:31] Closing the Podcast
[1:15:45] End of Podcast
Ben: I have a master's degree in physiology, biomechanics, and human nutrition. I've spent the past two decades competing in some of the most masochistic events on the planet from SEALFit Kokoro, Spartan Agoge, and the world's toughest mudder, the 13 Ironman triathlons, brutal bow hunts, adventure races, spearfishing, plant foraging, free diving, bodybuilding and beyond. I combine this intense time in the trenches with a blend of ancestral wisdom and modern science, search the globe for the world's top experts in performance, fat loss, recovery, gut hormones, brain, beauty, and brawn to deliver you this podcast. Everything you need to know to live an adventurous, joyful, and fulfilling life. My name is Ben Greenfield. Enjoy the ride.
Hey, neighbor, I would like to talk like Mr. Rogers when I'm about to ask you to go on an imaginative adventure, but I want you to actually go on an imagination adventure with me. Imagine if somehow you had this all-encompassing platform that allows you to keep track of your blood and your stool and your saliva and your urine testing and all this different test people are getting these days along with all your self-quantified data from wearables like the Oura Ring or the WHOOP wristband or Apple Watch, or anything else that you might wear to track your heart rate or your heart rate variability or your temperature, your sleep or anything else. And then, that same platform could pull in all the health testing that you've done in the past and just kind of make that part of the dashboard too so you could have your whole health history in there.
And then, we're going to keep imagining. Imagine that the platform could basically use really, really good artificial intelligence, like highly advanced artificial intelligence to tell you exactly how to eat, what the meal should be, protein carb fat ratios, what foods to select, what foods not to select, how to supplement, how to exercise and much more, all from one single dashboard. And, you could get all the testing done from the comfort of your own home without needing to drive to an expensive lab for multiple blood draws or fill out confusing paperwork.
Well, that is exactly what this brand-new company, Onegevity Health, is putting together. They've been working on this behind the scenes for years. I've known about it for a while but I had to wait to do a podcast with these scientists before or until, really, this thing was actually ready, and they're now rolling out, this whole dashboard and this test and everything. So, I figured it was high-time I sat down with the two chief scientists for Onegevity to learn how this works because I'm pretty dang stoked. I've already started my own testing with them. I've already got access to their dashboard. I've got like six gigabytes of data from my gut alone right now, but I can see exactly which supplements, which foods, everything. They use totally different sequencing, totally different technology than anybody I've ever talked to on this podcast in the past.
And the guys I interview in today's show, which was recorded pretty much right at the Thorne research facilities in New York City, because Thorne basically owns and runs Onegevity, the guests are Dr. Joel Dudley and Dr. Chris Mason. These are cool cats. They're smart as hell. Dr. Dudley is the Professor of Genetics and Genomic Sciences at Mount Sinai, where he's also in the chair of Biomedical Data Science and the founder of something called Next Generation Healthcare. He was Director of Informatics at something called NuMedii. He was a Consulting Professor at Stanford. He's been published in over 120 different peer-reviewed publications. You'll see this dude all over the Wall Street Journal, Scientific American, CNBC. He was named one of the 100 most creative people in business by Fast Company magazine, smart cat. And, in addition to that, he also has a PhD from Stanford University, School of Medicine.
And then, Dr Christopher Mason is on the podcast along with Dr. Joel Dudley. And Dr. Christopher Mason is also a professor. He's a Professor at Weill Cornell Medicine, and he's the Director of something called the WorldQuant Initiative for Quantitative Prediction, which bridges prediction methods and finance with genomics. He has a deep history in artificial intelligence. He's spoken at TEDMED. Popular Science listed him as one of the top 10 most brilliant people in the world, which is kind of an honor. And he also has over 140 different peer-reviewed publications. He's been in Science, he's been in Nature, he's been in Cell. He got his PhD in Genetics from Yale, and then also, got his post-doctoral training in clinical genetics at Yale. And he's wicked smart along with Joel Dudley.
So, we're going to unpack this whole blood, saliva, urine, stool testing piece.
Hey, before we jump into today's show, just a few quick things. This podcast today is brought to you by Kion, K-I-O-N. That's my company, where I source and design amazing supplements for you. And in our wide range of beneficial compounds that I've personally handpicked for you, you will find many products made by the company, Thorne, including their multivitamin that I use every day, their creatine, one of the most proven sports-performance enhancing aids, just supplements in general. Creatine staves off things like sarcopenia with age and its unknown nootropic. I mean, creatine flies under the radar, it seems, aside from strength and power but it's, for so many things, so beneficial.
We've got their probiotic on there, their vitamin D, vitamin K liquid, just some of the best of the best. Thorne supplements are all over at Kion because I have toured their facilities and will vouch for everything they do. They are considered to be one of the top clean guilt-free supplement manufacturers on the face of the planet. So, that's called Thorne. And we have all the Thorne products march down at a discount over at Kion. So, you go to getK-I-O-N.com and you can get fantastic prices on all the Thorne stuff along with all of your favorite Kion supplements.
So, in addition, this podcast is also brought to you by Organifi Gold. It's like curling up with a nice piping hot cup of medicinal liquid, which sounds like it would taste like ass, but it actually tastes pretty good. They put turmeric and black pepper into this along with a bunch of insulin stabilizing herbs and spices. They've got turkey tail, lemon balm, reishi. They put ginger in there. And when you blend this up–like I like to blend it up in the evening with a little bit of almond milk or coconut milk, and I top it with some cinnamon and some coconut flakes, and then sip it while I stare longingly into the fire, wearing my cowboy hat, smoking a cigar, naked. No, I'm just kidding. I don't do all the other stuff.
But I do drink Organifi Gold and their new flavor because I'm a chocolate fiend. Their new flavor Organifi Gold Chocolate, oh, my gosh, it's amazing to die for. So, anyways, you get 20% off all of the fine, fine fare from our friends at Organifi. You just go to Bengreenfieldfitness.com/organifi. That's Organifi with an I, and use the discount code mentioned there. That's easy enough to remember. So, Bengreenfieldfitness.com/organifi and use the discount code mentioned there.
Alright, folks. Well, welcome to the show. Chris, Joel are sitting here with me. You've just heard about these guys' extensive bios, and I'm actually holding this oddly shaped object that I believe is designed to collect my stool. It looks like a cross between a rocket ship and a dildo. And apparently, this is the wave of the future. This is where personalized genomic testing, especially for the biome, is going, and you guys are at the forefront of this new thing called Onegevity, which I've already learned is not pronounced Onegevity. It's Onegevity. So, tell me, what is Onegevity? And when you begin to talk, you can introduce who you are so folks can start to keep you straight as we get–I can also make one of you talk in a falsetto tone, if that's handy.
Joel: Yeah, like T-Pain. Now, so Joel Dudley here. Onegevity, in short, is a health intelligence company that is really trying to bring the cutting-edge science of precision medicine or precision health to the consumer realm.
Chris: Yeah, and hi. This is Chris Mason. So, it really is I think a manifestation of a dream I've had for a while to bring the best and most actionable information we have in the laboratory and give it to everybody and anybody who wants it. So, this includes everything. We're starting with gut health in our gut bioproduct, actually, so people can get a complete portrait of what's inside them and basically have actionable insights from that, including what you should take, what you should eat, things you can learn about your gut health. But then to expand out everything about your genetic code, your DNA, also looking at what's in your blood. So, it really will be integrated across every layer of your biology and make it so you can live a healthier, happier life.
Ben: Okay. How did this actually come to be? Like where did Onegevity actually come from? And you guys have a pretty extensive background in everything, from artificial intelligence to genomics. So, I'm curious how this actually materialized.
Joel: Yeah. So, Chris and I do research for the last decade or so at top tier medical institutions working again in sort of personalized medicine space. And I think for me, it came from a frustration about how long it takes for that science to reach, not only the clinic but the consumer. So, it takes 15 to 17 years for a scientific innovation to reach the clinic alone, and then many years after that for that to reach the consumer. It is born out of the frustration of how long that takes and how–we're just barely scratching the surface of this technology in treating disease and we're very far off in the typical academic realm and using this technology to focus on health and wellness and health prevention. So, there's pretty much no interest in health prevention in sort of mainstream medical research.
Ben: Now, I want to start to clear up a lot of the confusion around actual testing because it seems like biome testing is a dime a dozen these days. I've interviewed folks from Viome before. People are aware of the American gut projects of uBiome. I mean, it's very, very simple to get the impression that you're simply going to poop in a tube and send it off and get results back that give you some clues as to the bacterial profile of your gut. How is this any different than anything else that exists out there right now?
Chris: Yes. There's actually a really big difference between what we do at Onegevity and what you'll get from other places. In particular, almost all the work in the field and in literature in the past really 15 years has been done with an older technology which is called 16S, whereas we do what's called Shotgun Sequencing which is, as the name implies, you fragment and Shotgun destroy all the DNA from your sample and look at the provenance or the source of where all that DNA came from. So, it could be bacteria and viruses, which you've heard about before, and people have probably maybe even ordered tests before, that might be listening. But we expand out one of the fungal species that are present or even plant or even the amount of human DNA that's left over that we see in your stool itself as an indicator of health and ensure what's happening inside of you.
And so, we use better technology, better algorithms, the AI platform is built on, all those studies that we've been publishing in the past 10 years that have understood where each molecule comes from from each species. And so the technology is better, the algorithms are better, and most importantly, it's not just a PDF that you get. You just go, “Okay. I've got this profile. I might as well use it as wallpaper in my bathroom.” That's not going to help you. You might have to–
Ben: You mean like the PDF that says, “Here are all the bacteria”?
Chris: Which is like, “Okay, it's interesting, but it's like getting a recipe in a different language that you don't know what to do.” In this case, we translate the language of the molecules inside your body and give you things you can actually take, including products actually for probiotics, prebiotic supplements that you can take into your body that will beneficially impact what's in your microbiome and improve it. And then, also, actionable insights of what else you should be eating or taking or doing.
Joel: I think that's a key distinction from the companies that are out there because obviously, as you mentioned, a lot of companies are in this space already. I would call most of those companies recreational health information companies, not really health management companies, right, because they're providing you health information that is largely recreational because it's hard to act on it and they're not providing you a solution.
Ben: Well, I don't know. I mean, when I get tested–I've been tested by companies before and I get the PDF that lists the percentage of all the bacteria, but then it says things like, “Here are the foods that you should eat and the foods that you should stay away from.”
Joel: True, but they don't deliver those foods to you, and they don't provide that solution to you, and they don't actually give you the means to test and see if those solutions are actually–if you're responding to those on an individual level, right? So, it's a shopping list and then that shopping list can change, but the key thing here is we're vertically integrated with Thorne. So, we're integrating testing with the ability to design the intervention—
Ben: Or develop the actual solutions like, “Here is the actual supplement that could address this bacterial insufficiency or excess.”
Chris: And directly basically give you something–tailored just for you that's based on goods and supplements and probiotics/prebiotics all manufactured here in the U.S. that we–you can go tour the facility and see it's the best in class production for really anything you could buy at Shelves. You'll see Mayo Clinic sells it to people at their hospital. You'll see it at Shelves, at Walgreens. You'll see it at supermarkets, specifically, the best basically products that you can buy as an intervention for what we measure inside you. And then, we can see, “Is it working? Are you part of the 10% or 20% of people who only have to take some of the probiotics once and they stick and they've been colonized and you're good? Or some of them have to be taken continually by people?” But we can measure that, quantify it, and then advise you.
Ben: Well, I definitely want to talk about that, about if you take probiotics, if it even has an impact on these measurements because I question whether or not it does. But before that, you mentioned the type of sequencing you guys would use. You said it was called 16S. The ones you have is the one called shotgun.
Chris: Yeah. 16s is considered a swear word in my laboratory. If you say it, sometimes we will tackle–
Ben: What's 16?
Chris: So, there's 16S. 16S is a fragment of what's called the ribosomal RNA. It separates out what you could say eukaryotes, so that's sort of more like animals and plants versus what are called prokaryotes or archaea, these microbes. They're literally different kingdom or domain of life. And so, this 16S, the original, that sort of fragment of a gene, was how these new class of bacteria called archaea were first discovered by Carl Woese. So, basically in the '70s, there's really an excitement in microbiology because we just used that method to find a whole dramatically different domain of life, which lived in hydrothermal vents or really extreme places.
So, it has utility in research and it's been around for decades, but it was never designed as a way to profile an entire ecosystem like in your gut or in a field of soil, in a prairie. It is just a way to count a general view of what's there, and a way to discover something dramatically different like at a kingdom level. But what we want to do is find things on at the species and strain level that are the most important for your health, and obviously, even disease. Like for example, E. coli. Almost everyone that's listening has E. coli in them or in someone next to them. But you're not sick from E. coli. When you hear about E. coli in the news, you think, “Oh, no, there's an outbreak of E. coli in lettuce.” It's a very specific strain of E. coli. It's not just E. coli.
Ben: And other strains can be beneficial.
Chris: It can be beneficial, right. So, then 16S gives you almost none of that resolution, but as we get all that resolution–
Ben: 16S would just say you have E. coli?
Chris: It would just say–and not even that, it would just give you the genus or maybe even the family and say you have something–and archaea. We didn't say E. coli; we just say something at the genus level. I mean, it would be like saying, “Are you homo sapiens?” You could say like, “Well, I don't know. I'm kind of a primate and I want to do a clinical trial on some primate and then make recommendations on that.” You want to do a clinical trial on actual human being, not just a random primate or something that has a chordate like a spine. That's a bit too vague.
Joel: And I think we're almost starting over in microbiome research. So, even though there's tons of microbiome research as the decades and there are all kinds of interesting associations with everything from depression to cardiovascular disease and metabolic health, a lot of that again has been based on 16S sequencing, which is–there's low resolution, so we're now sort of almost starting again from scratch because shotgun metagenomics is getting to a point where–we're going to have to go back and re-profile and really learn again what's happening in the microbiome with this much higher resolution technique.
Ben: What is metatranscriptome analysis? Because I see that word thrown around as something that companies are using.
Chris: Metagenome is all the DNA, and meta just means across all species. So, you're not just looking at just bacteria or only say human DNA or fungal. You're looking at all the DNA metagenome with all the genomes, whereas, metatranscriptome is everything that's made from DNA into RNA and sort of that landscape as well. So, it's all the RNA that you'll pull out of a sample. So, it could be from skin, it could be from your gut, again it could be from a prairie and you just extract all the RNA out, which is what, when DNA gets transcribed to become a more active molecule. That is the transcriptome, all of the transcriptomes of–one species is called transcriptome, and if it's all species like human, fungal, bacteria, even viral RNAs, it's called metatranscriptome.
Ben: Why wouldn't you just do that instead of shotgun sequencing?
Chris: We've tested this. We actually have led studies with the FDA, with NIST to look at basically international standards for metagenomics versus metatranscriptomics. In the literature, the idea is that, well, if it's RNA, that means it's active, so maybe that tells you what's alive or [00:18:28] ______.
Ben: Right. That's what I'm thinking is like the RNA would tell you what's actually going on.
Chris: But there are several challenges with the RNA. Most of the clinicians I know consider metatranscriptomics to be “useless.” They just don't even want to look at a deficit if it's metatranscriptomics because the RNA, while it's active, it's also then the most susceptible to degradation and it–even though the molecule is active, it is the one that is destroyed the fastest. Also, you'll only get sort of the top few percent of the species that are active because they'll be the most—that generate a lot of the RNAs.
So, one cell usually has one or two or several copies of DNA so you can quantify who's there a bit better than you can with metatranscriptomics. You'll just see who's the most active but that doesn't necessary tell you who might be present at a lower level that could be a higher risk. And so, between those two things–and then also the databases are not as good, the methods are not as good. For metatranscriptomics, it's a bit more new. Again, most of the people working in pathology or clinical labs described that my experience is useless because it's too noisy of a data and it degrades too fast.
Ben: Is shotgun sequencing more expensive?
Chris: No. They'd be about the same in price. I mean, in terms of–once you get DNA or RNA and your goal is to sequence them, they could be relatively similar, but DNA sequencing was ways to make that cheaper, and what we're putting out is going to be cheaper.
Joel: And by virtue of the way that bacteria replicate themselves and their genome, which is different in some ways than humans, there are ways to infer activity from metagenomics, actually, of the bacteria. So, just because you're doing metagenomics in DNA doesn't mean you're blind to this sort of activity that you might capture with transcriptomics or actually computational techniques you can use looking at the metatranscriptomic data to sort of infer some activity of the genes from the DNA.
Ben: Okay. So, right now, what's going on is I would get a kit from Onegevity and I would send off a stool sample. How large is the stool sample?
Joel: It fits in the–
Ben: Some of these tests, I mean, it's an ungodly amount like a hotdog tray that you got to collect in your refrigerator for three days and you'd send off in a prepaid FedEx bag. How much are we talking?
Chris: I mean, if you have a personal interest in storing a lot of your own material, you can still do that. But for our test, you don't–
Ben: I did. I like it. I like to look at it.
Chris: There's a friend of mine, he just puts it in the freezer, labels it as meatloaf and wait and see who finds it.
Ben: Halo top peanut butter cup.
Joel: I actually know a guy in Twitter–he's a microbiome researcher and there is something called fecal microbiome transplants where to do it, you can do it in autologous transplant. You say if you get a gastrointestinal issue, you could actually transplant your own stool at your bank, say, years before as a way to treat yourself. And so even though there's a guy who posts this online, he says, “I'm putting it in my freezer,” his wife–you have to make sure–in a good marriage, you communicate which stool you're putting in which freezer, of course. But he's doing it for a medical reason. If anything does go wrong, it gives you basically–it's like a bone marrow transplant.
Ben: Yeah. No, I think it's a great idea. I was recently working on a chapter in my book about FMTs and Clostridium and longevity and– Yeah, it seems to make sense that if you're at a time in your life when your poo is healthy, why not bank it? It's like the stem cells.
Chris: Yeah, exactly. So, with Onegevity, one of the things we're calling almost like a virtual stool bank, where if we–because we sequence deep enough and broadly enough, we can tell you what you look like when you're healthy, or people that have come to our company or customers that have come to us that have got problems, we can address that. But even healthy people, it's really a snapshot of where you are when you're healthy that you can go back to if anything goes wrong. But without that profile, you can't get close to it and you could even bank it as well, but the profile is the bare minimum of what you need to try and get back to where you were.
Ben: So, it's essentially the amount of fecal matter you get off of one swipe with a toilet paper?
Chris: Yeah, small spoon, yeah.
Ben: Okay. So, then that goes off to your lab, where it is shotgun sequenced, and then what happens?
Chris: And then, so we take all those fragments of DNA. So, we'll generate at this point about six gigabases of data, which is sort of millions and millions of fragments of DNA that we take each fragment and compare it to all known species, basically, all five million known species to humanity. We check and see which one of them it came from. So, at the end then you'll get–once the mapping is complete, we then give you a quantification of every species of bacteria, viruses, phages, different kind of bacterial viruses, fungi, plant DNA. We even see the amount of human DNA. So, we use this to get the first profile, and if we could walk through what's in that report, you get a report that's linked–you can still get a downloadable PDF if you want to go back to the old school way of staring at a PDF. But the site is interactive and it importantly gives you suggestions on what you should order if you want to try and say–you get an option for helping your IBS, or if you just want to–sort of the other products that might maintain your health.
Ben: Are you measuring anything with the blood as well to kind of go along with the microbiome data?
Joel: So, that's the thing that distinguished longevity really strongly going forward. It's sort of multi-omics assessment. Any one dimension is really limited in what it can tell you about your overall health because what's happening in the gut, obviously, we might think it's transferring into the blood, but unless we measure the blood directly.
Ben: Well, yeah. I mean, you get different data from different bodily fluids. I mean, a lot of people, they'll do an organic amino acids test using urine or like a DUTCH test for hormones using urine and then a complete blood count, a comprehensive metabolic profile in the blood, in the gut, and then like a genetic panel via the saliva. So, you guys are pairing blood data with the gut data?
Joel: Yeah. So, we don't know in the current product right now but that's–which was mentioned in our press release as part of the product roadmap. And for any, I think, health area that we look at, it's going to be a multi-omics sort of assessment. And really, Chris and I's background on the academic side has really been around the–
Chris: The power.
Joel: Yeah, but how do you combine multi-omics data for individuals and then how do you do inference across all of those things, those dimensions simultaneously. So, a lot of people will, say, measure genome, measure blood, measure microbiome, and then they sort of look at those dimensions one by one and say, “Well, let's try to learn something from the microbiome, the blood and genetics, but we have deep experience, and let's combine those into a single model, a single representation so we can really understand what the interplay is between all these factors and what it's telling us about what's happening in the gut and what's really driving the–in the blood.” And we use things like network modeling and things like that on the computational side. But that's really the key. No single dimension is really good enough to give you the full picture.
Ben: What do you mean network modeling?
Joel: So, instead of just analyzing the genetics alone, or again, the microbes alone, you can actually build these network models where you can actually understand how are the changes in the gut microbiome, for example, potentially propagating up to the blood. And what the network modeling will let you do is, not just look at these correlative associations, but really help you try to untangle what's potentially causally–causal is a dirty word in biology, so you don't want to say causally. But is there a directional link between what's happening in the gut microbiome and what's happening in the blood rather than just trying to guess from these correlations.
Ben: So, what would be an example of things that you would measure in the blood?
Joel: Yeah. So, to be determined right as we develop the products. And it's going to depend on the health condition, right, so what makes sense to measure in the blood, but it bacterially drive metabolites if we feel that there are some bacterially drive metabolites that are being produced by the gut.
Ben: You mean the bacteria in the gut are producing sort of metabolites that are whining from bloodstream and you're measuring those?
Joel: Right. Or you might assume, but maybe you don't know what the gut barrier could be disrupted, right? Or you might think that they shouldn't be leaking into the blood but you want to test to make sure–even though they're active in the colon, they shouldn't leak, but maybe there's a gut barrier issue, so we actually need to test in the blood.
Chris: And vitamin levels too. For example, vitamin B6, up to 85% of the vitamin B6 that's made in your body could be made by a species in your gut. So, if you think I'm a high or low in different vitamins, you might normally just look in blood, but you wouldn't be examining where the genesis of actually a lot of the vitamins are. I mean, basically, in our guts, there's a small pharmacy for all of us that both makes and processes drugs in molecules and vitamins. And so, in the absence of knowing what's in your gut, you could start to think about, “I'll just do blood work and think about what I'm missing.” But you can create some new members of the molecular factory in your gut that can make them for you. And so, part of the measurement is to see what do we predict that's being made with the vitamin levels, but also with bacteria metabolites which you can just predict based on who's there, the species. And again, it has to be the species in the strain level. You can't get it from–say, I've just looked at some broad class of bacteria because you don't know what it is.
Ben: Which is why you need to use shotgun?
Joel: Shotgun, right.
Ben: Okay. Got it. Now, I want to get back to the practical recommendations, and I also am intrigued by this idea of using artificial intelligence, and I want to learn more about how you're doing that as well. But I also know there's a lot of skepticism about this overall testing process as a whole because you're using stool. Basically, whatever happens, to be near the rectum and the anus when you test, how can that be reflective or is that reflective of what's going on higher up in the tube? I mean, is there any evidence that that really is showing what's really going on?
Chris: The best models have been done in mouse, where it's much easier to get, say, six areas of the GI tract from even what's in the food to what comes all at the end. And we have two additional studies ongoing to see–
Ben: Because mice have six assholes. Yeah, we all know that.
Chris: Interestingly, in grad school, I did make a worm that had three vulvas, which was a unique experience.
Ben: That's useful.
Chris: There's a LIN-37 mutation. It was just–
Ben: A multivulval species could be useful at some point.
Chris: And I went home at Christmas, my first-year grad student. I'm going home at Christmas and my grandma was like, “Oh, how was grad school? Are you getting your PhD? That sounds exciting. And how's Yale? It sounds great.” And I said, “Well, I made a worm that has three vulvas. And she's like, “What's a vulva?” I was like, “Well, it's a reproductive organ in the species. It's a hermaphrodite. It can make its own babies. You put it on a plate. It makes its own progeny. And she's like, “Why are you doing this again?” And I just said, “Well, I know it's important cancer signaling. It's really important.” And I realized at that moment, looking down at the floor like, “I need a better way to explain that to my grandparents.”
Ben: Well, in the biohacking sector, people are getting chlorophyll injected into their eyeballs or little magnetoreceptors in their ears. I would imagine there are probably a few hackers out there who would want an extra vulva.
Chris: It's a limited market but I think they're passionate about it.
Joel: So, despite Chris' best efforts, that's not yet on the longevity product–
Ben: So, these mice, you're able to test them and you're able to see the bacteria profile different areas of the gut.
Chris: Right. And so, it is very true that different areas of the gut–I mean, it's a digestive process, so your things are being broken down, decomposed. Ideally, we would love to get every part of everyone's esophagus, jejunum, duodenum, everything out to the colon. But from these studies we published, it's 70%, 80%, as much as 90% of the signal, what you're looking for, of what's actionable still in the stool. So, I think there are other people who can–there are ways you can even swallow devices that would take sampling all the way through GI tract. But I think it's not necessary and it's also more risky to try and collect things higher up. But not to discount, like helicobacter pylori for driving ulcers that's more in the stomach, not at the end. But you can see most of what you wanted with this tool.
Joel: And I think this is where the machine learning part of the AI, however you want to say it, comes in as there have been very large studies published where they have done this sort of spatial microbiome sequencing along the GI tract, and published these very large reference databases of, say, spatial microbiomes–
Ben: That's why it's called spatial microbiome if you're going to take it from different areas, not just like one toilet paper swipe.
Chris: Right, right. But because there are these big reference databases out there, what you can do with machine learning is take someone's sample from–just one collection site and then you could use machine learning to infer what are the likely upstream, depending on how reliable that signal is in that data, right, if there's a decay of a signal that's consistent. So, this is really the power of using large published datasets that we've pulled in into our platform. And this goes along with metabolomics as well. So, there's been large reference databases published with–where they mentioned whole metabolomics on fecal samples. And they've done, say, metagenomics and then we have a machine learning algorithm that can look at the metagenomics data and really build you a virtual metabolome using machine learning.
Ben: How much data have you pulled in? I mean, I don't know how much data there is out there on biome testing, but I was under the impression it was kind of a new thing.
Chris: There's a lot of 16S data that's out there. For 16S, there are tens of thousands samples, but for metagenomics that's paired with actual metabolomics, we've got over 2,000 samples but that's the majority of the data that's ours plus public data. There's not that much that's out there, but you think, “Well, how can you do classify our machine learning and only a few thousand samples?” And also there is a bit of AI exhaustion I think in the field. We would think like, “Oh, it's a machine learning if you're writing code. It's AI If you're just talking in a nebulous way about doing something cool.
But you can call it broadly AIs because you can use multiple kinds of machine learning on the datasets that we've curated. So, this is random forest models’ neural networks. Even in some cases, simple regression is good enough, but you can do multivariate regression. So, you look at multiple variables at once. And what that means is you're not just–the reason we have all that information is because it's bacteria viruses, it's human DNA, it's the mapping of each piece of DNA to every possible species. So, in a feature space, it's actually very, very large and many places to learn from.
And that's why I think some of our excitement about machinery and AI is built into the companies because it is to date, by far the easiest and fastest and best way to take unstructured sort of vague data of unknown importance and find signals in it and then actually act on them. And they also separate out groups of high risk/low risk, IBS/IBD, healthy controls. It's actually shockingly easy if you have enough data.
Ben: Well, hello. I want to interrupt today's show to tell you about the best grocery store that exists that you don't even have to drive and park at because it's on the internet. It's this revolutionary online marketplace where you can get dark chocolate and coconut cereal on Coconut Manna and personal care products and household cleaning chemicals that don't give you man boobs or cancer, but that are actually handpicked to be extremely clean by the folks at Thrive Market. They have non-GMO food, safe and nontoxic beauty products, kitchen staples, organic baby food, so you don't poison your baby, all sorts of amazing things.
And it's fun to shop for because you can filter everything, whether you want to filter by ketogenic or paleo or gluten-free or vegan, or even kosher. That's right, you can filter the cereal that you want to get by kosher. You get paleo kosher granola. How do you like that? Seventy percent of their catalog you can't even get on Amazon, plus their prices are rock bottom compared to like the fancy local organic–fancy pants grocery store. They're just wonderful. Even their packaging and their boxes and their inserts are made from recycled paper, 100% 0 waste. These are all the goodies from our fine, fine friends at Thrive Market. So, here's the deal, if you go to thrivemarket.com/greenfield, you get 25% off your first order to Thrive Market. That's thrivemarket.com/greenfield. Have fun shopping because you get 25% off. Then may I recommend to you their coconut cereal. That's really good, really good, just saying.
This podcast is also brought to you by PuraTHRIVE. What's PuraTHRIVE? It's actually kind of cool. They sent me some bottles of this stuff. It tastes really good. I just squirt it into my mouth but I think it's designed to be maybe blended with coffee and stuff like that. But it's called Keto Balance. So, it tastes like a chocolate smoothie. It's like this liquid but it's got DHA omega-3 fatty acids in it, grass-fed bovine collagen and medium-chain triglycerides. They get their DHA from algae oil rather than fish oil. So, technically, it's like full on, vegan-friendly. They use this patented–well, no, it's not vegan-friendly. What am I saying? It's got bovine collagen in it. I'm sorry, my vegan friends, but regardless, I still think it's an amazing tasting supplement and you should consider it either way. Like, it's just bovine collagen, that's all. The cows are really very, very happy.
So, they use this patented liposomal delivery mechanism that increases the bioavailability of the compounds in their 20-fold. So, you can get into ketosis faster, and you can avoid the keto flu, and you can achieve your keto goals in less time; keto, keto, keto. Okay. So, I didn't say that enough just now. Anyways, I, like I mentioned, have used it. I use anything I mention on this show, and it tastes really good, and it does really truly satiate your appetite. I just use like a tablespoon of it. You're supposed to use a teaspoon [00:35:30] ______ for punishment. And I think it tastes amazing and I'm going to run out of bottles really soon based on the rate at which I'm going through it. So, you get 15% off. You just go to purathrive.com/greenfieldketo. There I said it again, keto. Purathrive.com/greenfieldketo. That gets you 15% off. Again, P-U-R-Athrive.com/greenfieldketo.
So, the AI is able to predict potential for certain diseases or conditions based on the shotgun sequencing of the stool sample that you send in, but then it's also able to generate targeted recommendations for the foods or the supplements that you should be consuming. Again, this is not based on metatranscriptome or 16S; it's all based on shotgun. So, the list of recommendations you get from a test like this might be different than the list of recommendations that you get if you had tested elsewhere.
Joel: That's right. It would be different and better from our experience.
Ben: This could potentially get very confusing for people who are sending their poop off to different companies for sure.
Chris: I mean, this has happened also like with 23andMe where people going direct to consumer just genetic testing. I think every once a year, there's a New York Times reporter who says, “Well, I sent my saliva DNA to three companies and I got back different risk scores.”
Ben: Exactly. Same thing with food allergies where I sent off my blood results for Alcat in at least three different times and it came back with different foods to avoid.
Chris: And it depends. I mean, it's interesting. In that case, for the people getting different risks for even say like diabetes risk and they'll say, “Oh, how did I get three different answers from three different companies?” It would actually be a miracle if you got the exact same answer from three companies that all have three different databases that they're using as a reference. So, it's interesting. I think there's a disconnect between the consumer view of what they should see, or if you think of your DNA, you look in the mirror, you would think, “My DNA is mostly the same. I shouldn't get a different answer from three different companies.” But each company has its own database, its own set of scientists, and actually literally different risk scores that they're calculating using different data.
So, like Navigenics, 23andMe, even like Veritas, other companies where they'll give you a risk score but it's actually not at all surprising that they're different. I mean, they should be broadly consistent but they would never be identical because they're using different data. In this case, we too will be using different data and will, of course, tell you that we think our data is more comprehensive and better. But in this day–
Ben: Where is the data coming from on your end?
Chris: Its friends and family like early trials of our product. It's a clinical trial that we've run in the company and every available public dataset that we have that exists, basically. I would even argue it's the totality of all public knowledge and all company knowledge on metagenomics and metabolomics.
Ben: Now, from a longitudinal standpoint, if people are doing this test, that database is going to continue to grow.
Chris: Right. And the machine not only gets stronger and better, but basically, every day that passes in some sense, the company and its ability to make better recommendations improves. And also, we're strong believers. And if something doesn't work, if we've done something that's been hundreds of people or thousands of people, and we can quantify and literally do phenotype scores and if no one's benefiting from something, we'll, of course, stop recommending. There's no reason to give people things that don't work. I mean, it's unethical frankly. And so, even if you make money from it, our goal is not to profit from just selling things. Our goal is to sort of profit in partnership with customers so that they're healthier.
Ben: That's actually what I was going to ask you. And pharmaceutical companies waltz in and buy this data 10 years from now.
Joel: Now, they could propose a partnership, but it's on our terms, right? And as we mentioned in the press release, this has to be very transparent and equitable to the consumers who are providing the data. We're open to it because to be frank, we still need lots of new therapies for lots of diseases, right? So, I mean, look at Alzheimer's for example, I mean Alzheimer's is on the verge of bankrupting our economy and we haven't had a single drug approved. I mean, there's still work that needs to be done there, and we think this data we're collecting can be valuable for discovering new therapies. So, we do need that to partner with those groups.
Ben: Yeah. I actually thought that the curve for Alzheimer's had been discovered and involved putting MCT oil in your coffee and putting brain games on your cell phone.
Joel: Yeah. I'm pretty sure that's how it works these days, some pushups in the middle.
Ben: Yeah. To a certain extent, pushups and Wim Hof breathing. When it comes to what we can actually gather from the microbiome data, and this is kind of delve into some of these personalized recommendations that you're talking about that this dashboard that you've created is able to produce, how does that actually work? How do you go from stool to shotgun to “don't eat green beans0?”
Chris: The most was based–we're leading to a clinical trial which we're just sending off for peer review just now. Actually, the trial was extraordinary in terms of–we asked people to give a complete–big sort of–
Ben: You ran this trial internally to a population.
Chris: Mostly IBS, IBD, and healthy controls. It was a lot of quantified metrics of whether they got issues. It was at nine-point measurement scale. We had an internal IRB and an external IRB, reviewed the study, just completed it in December, and the results are really shocking because it was a relatively small trial of 96 people. But even just in that trial, we had almost perfect classification of people separating out IBD and IBS from healthy controls. Again, because we're looking not just at bacteria or just viruses but across all domains of life to see what the separation is. It was actually literally some of the best that I've ever seen as a geneticist in my entire career. In terms of a separation, it clearly shows people that are at risk and people that are healthy. That was already exciting. But then also, we did try different recommendations based on–we had two time points as we think. I wish everyone was sampled dozens or hundreds of times ideally on a minute by minute basis, but in this case, we only had two time points–
Ben: I poop a lot. You could probably test me a few hundred times a year.
Chris: We've got some interns that could follow you around. Yeah. I think this could be a good partnership. But we did do two time points, which is the bare minimum to make a line, but you have a sense of what did we recommend at Day 0 and then at Day 30, what are the species that change, and also more importantly, what was the outcome. So, I think you're alluding earlier like people talk about probiotics and maybe they can help or what did you take, what vitamins, what supplements? But are they working? In here, we could quantify, was it actually helping at Day 15 and Day 30? And so, we would just–
Ben: Just test these people again probiotic.
Chris: You, and also several supports.
Joel: Or a prebiotic. Or it all depends, depending on what we found–
Ben: You customized it?
Joel: Yeah, we customize for this person. And the customization was driven by their microbiome. A lot of these other companies customize it based on the intake survey and not actually on your microbiome data.
Ben: And you were able to use because you guys are tied to Thorne, just the Thorne broad suite of the different prebiotic and probiotic formulations they have access to. You just kind of pick the ones off the shelf that seem to work best for that person. So, you slap on those on these people with IBD and IBS, and then retest and then see what happens?
Ben: And also, looking at symptoms?
Chris: Yup, the nine-point scale for gut health, but it includes things like how often did you have diarrhea? Some things were obvious. Did you have diarrhea, constipation? But in general, if you feel gut pain or did you have gas? Did you have any bloating? What are some of the general physiological responses you had in your gut? And actually, I mean one of the best Christmas presents we had was getting some emails just from the customers who said things that were really just like–that they've got issues for five/six years and have felt awful. And suddenly for the first time, and almost a decade, some of them, they feel great, and it was really extraordinary to see that, which gives us the inspiration to really push harder and faster. But also, I think it picks the curiosity of like, “Well, why did some people have an extraordinary response?” And suddenly, you feel like that they felt the best they have almost–
Ben: I'd be curious to see what happens. They just took some random probiotic though.
Chris: Right. And so, if they just take–because what we did here was a trial. We had three groups, healthy controls as a placebo, plus two as IBS, IBD, but we're going to ramp up now to a double-blind placebo trial.
Ben: I was going to say, yeah, placebo trial would be interesting on that.
Chris: Yeah. And healthy controls serve a little bit of that purpose, but doing it as double-blind was the next step, and also expanding the scope of it. So, that's why we're launching probably towards the end of this month or next month. I mean, the results are really extraordinary to see–clearly, some people are benefiting extremely excited but now we want to tease out at a molecular level. What's driving that? What's pushing at one direction?
Ben: I was going to say that's my understanding is we still don't know why. We still don't know what's going on. There's still some questionable science in terms of how much of that probiotic even survives the acidic nature of the stomach and winds up populating the GI tract.
Chris: Great, great point. And most of them do not, not only from work and at our company but also in our labs but other–I'm on advisory boards for a variety of other companies in the microbiome space and this is a universal feature. Everyone who's trying to look at–can we get something to get into your gut and colonize? It's actually really, really hard. Actually, the number is as low as 5% or 10%, as high as maybe 30% of people. Or if you take something, it will actually stay in you. You basically transplant a little species into an ecosystem that's already pretty settled. So, it's not surprising actually that all the studies you've seen and read about [00:44:43] ______, probiotics, you have to keep taking them, which is great for a business model but it's not great if you think they want to just take it–
Joel: It's not like your gut has these like patches of soil that are waiting to be planted, right? So they have these other bacteria in various niches that they hold on to for dear life. A lot of the molecules that bacteria secrete are things to kill other bacteria to keep them off their turf, right?
Ben: Well, I mean, it's like a lot of things. Like I don't just drink one cup of coffee on a Monday and stay awake and alert the rest of the week. I have my coffee every day. I'm taking probiotic every day. Does Thorne actually make a special kind of probiotic that's supposed to seed the gut back, because you hear about soil-based probiotics? I know there are companies now enveloping the probiotic like the algae or kelp or something like that to get it to seed the stomach.
Joel: So, they do produce probiotics that have clinical evidence behind them S. boulardii. A good example where–you can go to literature and you can actually see in IBD patients that there is a positive outcome.
Ben: Benefit, yeah.
Joel: Even though it doesn't graft into the microbiome doesn't mean it's not providing benefits. So, it could provide benefits because it's transiently secreting things as it's in there or creates a stimulation. But just by exposing itself to the immune system, that ultimately results in a beneficial thing.
Ben: That's how even helminthic therapy works, the use of tapeworms or whipworms, which people do to modulate the strength of the immune system. I've done that, but you choose strains that actually don't populate the gut, their transient passengers, but they still modulate effect on the immune system.
Joel: Yeah. So, even if they don't graft, I think the important point is they still might have an effect. And transiently, that's useful. We just have to know for who and when.
Ben: Right. You just have to get past the itchy asshole effect of swallowing [00:46:26] ______.
Chris: Which some people though want that, so it's a different customer base. We're open to helping everyone, yeah.
Ben: It can be entertaining when you [00:46:33] ______ asshole. So, back to the food, you've got the stool sample, you've done the shotgun sequencing, and now you're generating the list of food recommendations. With these folks in this study, it sounds like you're using probiotic and prebiotic formulas. Did you also test out these food recommendations?
Chris: Yeah. We send the recommendations out and we ask about compliance. So, we did, for example, gluten-free, dairy-free. Also, what's actually exciting about the clinical trial, we just separated this out a couple of weeks ago looking at the data, because some people are already on gluten-free or dairy-free diets. And so, we're going to separate that well. If you're already on a diet, maybe you wouldn't have seen as much improvement. But if people who really–if we made a recommendation of what they should eat or the type of diet, did that have an impact?
And so, we could see that specific recommendation with, not only the foods but also the broad class of diet that have significant improvement in their score. So, p-value of 0.0003, so it's very significant. It's a small cohort but we could see it was different and the majority of people improved.
Joel: And I think this is where Onegevity goes in the future because again, remember, we're about not only the information but about the whole integrated experience and delivering the actual solution to the problem we find. So, I think as we go forward, we're going to see an ability to start delivering even some of the food products that we can verify.
Ben: Okay. When you're making the food recommendations, because I've toyed around with using data that I've gotten from other microbiome testing companies, there's a list of foods to eat and the list of foods to avoid, and that's what you get. When you're generating your list of food recommendations, are you saying, “Mediterranean diet, paleo diet, keto diet, food elimination diet,” or is it just like, “Eat these foods, avoid these ones”? Like, how comprehensive are these dietary recommendations we're talking about?
Joel: Both those things and also specific recipes. So, one of the things is getting ketogenic diets can be limiting in terms of ingredients. And so, we partnered with some chefs that can give a very nice actual meal.
Ben: So, on my dashboard, I can actually select what I'm going to have for breakfast, lunch and dinner based on my microbiome results?
Chris: Well, we have in the recipes. I don't know if it's all broken down by meal yet, but I mean that we could add in very pretty easily, yeah.
Ben: Can I get my food delivered to me if I don't want to make that myself?
Chris: So, we have some ongoing discussions with a variety of companies about this exact idea. And so that I think completes the circuit of everything from what we've measured in you, making sure we deliver–
Ben: Well, that would be pretty damn cool. And I realize there are a lot of people out there who are like, “Oh, you lazy bastard. You need somebody making your meals in your kitchen.”
Chris: No. They want to live healthy. I'd say they're efficient instead of lazy because they want to get things done–
Ben: Well, you could choose your own adventure too, right? You could use the recipes that are generated for you by the artificial intelligence on the dashboard, or you could theoretically, and it sounds to me like you guys are building this in, just click and sign up for meal delivery service.
Joel: Yeah, and it's not laziness. It's anything you can do to increase compliance. So, if you actually go to the healthcare system and traditional healthcare, the biggest problem we have is getting people to comply with what they know they should do, right? So, this is why you got to make it easy. That's how you're going to get the health outcomes.
Chris: And I mean, basically then the food is basically genetically tailored for what's inside you. And it's early days in this regard, so we've got promising results from our clinical trial in our early customers. But we want to actually really engage the customers. A lot of our literature and language on the website is really viewing all the customers as partners in this–improving how we do science and medicine.
Joel: And there's a lot that the world doesn't know yet about these relationships between food and microbiome. And this is something we're going to learn along the way. Getting back to the AI part of it, there's something in artificial intelligence called reinforcement learning, which is basically, you try a prediction and you see if it fails or succeeds. And if it fails, you sort of boost, say, the neural network towards the thing that succeeded. The first demonstration of reinforcement learning that really captured everybody's imagination was they had, I think it was Google, had an algorithm that looked at Super Mario Brothers, and just looked at the screen, and they would randomly press buttons on the controller, and it would see if Mario would go right in advance, or the score would go up, they would reinforce the algorithm in that direction.
It's saying, “Oh, I'm just randomly trying things I'm reading and this is reinforcing that this is the right direction.” But this thing, by randomly pressing buttons and getting this sort of feedback, was able to play Super Mario Brothers with superhuman abilities without ever being told what the rules are or what the game is, right? So, this is the type of thing you can learn with the longitudinal data aspect with putting tools in the hands of our customers and partnering with them on the science. We're going to make recommendations. Some of them aren't going to be perfect but the system is closed loop. It captures that feedback. We can apply reinforcement learning. And then, for the next person, it's getting better and better and better.
Chris: And quantify. I mean, the measurement is one of the key features, like the Super Mario Brothers. And we want to be the Super Mario Brothers controller in your gut, I guess would be the analogy.
Ben: I think using the analogy of Call of Duty would probably get a little bit more palatable, potentially sexy, marketable to the general population.
Chris: These tubes in Super Mario Brothers, I guess, but yeah, I think—
Ben: That's true, that's true. It might be a little bit too old school. Now, when it comes to the other type of recommendations that you make on this platform, you talked about using like the Thorne suite of supplements to say, “Hey, here's your microbiome and it sounds like coming down the road your blood results. Here's the actual suite of supplements that you need to take.” But where I see the world going when it comes to personalized nutrigenomics and customized supplementation would be that that is actually customized to me, that supplement, or that probiotic strain, or that prebiotic. How close are we to that kind of technology?
Joel: Well, I don't think we can talk about it on the podcast, but let's say much closer than you think.
Ben: Should I edit this out or like can you just say we're close?
Joel: We're very close.
Chris: On a variety of friends, very, very close that you could be almost perfectly genetically tailored to what you would get.
Ben: Yeah. Perfectly genetically tailored right now though on microbiome. How come you guys aren't doing saliva?
Chris: Both. So, when I say genetically tailored, I mean across domains would be microbiome–
Joel: That's it. I would say actually molecularly tailored.
Chris: Molecularly tailored.
Joel: Yeah, because the blood will come into it as well.
Ben: So, you're going to start with gut, which is launched. Then, you're going to do blood and genetics.
Chris: It was a partnership with Drawbridge Health, which their device, the one-draw device is under FDA review right now for–
Ben: I don't know what that is.
Chris: Basically, this is an FDA-approved device. And so, you can't really bring anything to the clinic or you shouldn't bring anything to the market that's not gone through the FDA in approval. So, it's in the middle of that process, and we expect it soon. So, let's say spring, hopefully, not later than summer, but then–
Ben: What are you talking about, the blood draw device?
Chris: And it's basically something that fits in the mail, so you can just pop up under your shoulder. It draws blood and it does it on a standard sort of collection. Instead of trying to do something, like make an entirely new blood technology, it's just a way to collect the blood. Then we send it to standard labs–
Ben: Just having like that girl from Theranos that didn't work out so well.
Chris: So, I published the only academic paper evaluating Theranos against LabCorp. That was on CNBC talking about it if someone wants to Google. But I can guarantee you this is very different. So, I contributed to the–
Ben: So, I wouldn't have to drive to LabCorp?
Ben: Or Quest Labs, just at the house, you put it on your arm. It draws just a little bit of blood then I send that off. And I could theoretically send that off along with my biome analysis. Will that blood do genetics as well?
Chris: Yeah. You get DNA and–
Ben: Why wouldn't you do saliva?
Chris: So, saliva you can get an oral microbiome, and we're launching with saliva just because we can use that right now. Assuming we get as much DNA as they're perceiving right now basically on the one draw device, you wouldn't have to do saliva anymore. You get everything from the blood draw. You get DNA, and also your metabolites.
Joel: Yes. We think it's going to be a game changer for the blood draw side of it, because venipuncture scares a lot of people just to even get a single blood draw measurement. But another powerful application is we're getting into this longitudinal analysis. So, replete blood draws become much more feasible over time and to build out that dynamic longitudinal profile. And people can self-administer in the comfort of their home in the future. So, again, this is all pending regulatory approval.
Ben: That's pretty cool. It's a lot of data you're going to be able to get. But it kind of reminds me of this article. I think it was just a few days ago I was reading about how 23andMe either got bought out or infused with capital by a pharmaceutical company. And that pharmaceutical company may then have access to the data. And what if I don't want my data public? What are you doing with the customer's data?
Chris: So, by default, customers are opted out and you can opt in if you want to share your data. And then, also, we've set it up so that we will basically silo what is a customer data. And if there is a moment we have a partnership with say a large pharmaceutical company or biotech, we have committed to sharing any of the revenues that would come from that 50/50 with the customers. We want to make it so that if anyone is using your data, which is your right to give away or do it as you please, if you've basically generously given it to the company for us to start to look at and learn from, you should also benefit if you didn't share that with anyone else. And so we want to make that as transparent and useful.
Ben: You mean I could make money off of my blood and my poop?
Chris: Your poop could be extremely valuable.
Ben: This could take donating sperm to the next level.
Chris: Yeah. You flush it away every day, but it could–I mean, for fecal microbiome transplant, it's very much sort of like a goldmine.
Ben: So, what you're saying is if I were to test with Onegevity and five, ten years from now this data were to be something that I had opted in and was okay with being sold, I would actually opt off it from taking part in this test?
Chris: Which builds a long view of the data and what we want to do with it. We want to really encourage people to be partners. And so the more measurements we get, whether it's blood microbiome, metabolomics, the genome doesn't change that much, but all that information has value for it to learn and improve, not only for the customer but for everyone.
Joel: And there's a reason why we're able to do this, and it really gets back to our business model because again, a lot of these companies who are getting customers to subsidize the cost of the testing upfront, they're making their sole way of making money that's through these data partnerships. So, hoarding proprietary data is really central to their business model because they have no other way of making money. In fact, they might even be losing money on the upfront test hoping to make money on the back end, these recreational health information companies because we are delivering products that people will only keep buying if they're actually fixing a health problem and we make money on products. We have the freedom in a way, besides our freedom our business model. It enables us to be open with the data and be able to share back with our customers.
Ben: That's interesting. Now, there are companies like DNAFit, for example, that are making recommendations on exercise. Are you a power responder versus an endurance responder? Should you be going high rep/low weight, high weight/low rep? Are you guys going to be working in kind of exercise or lifestyle recommendations?
Chris: Yes, but sports and human performance, will be a vertical–
Ben: How's that going to work?
Chris: Yeah. Again, it would be better, because if you just look at the DNA, you can get a general recommendation. And it's like saying, “If you sequence a baby at birth and say, ‘How tall do I think this baby will be?'” or risk of disease, risk of cancer, breast cancer, you can put up the numbers or what are sometimes called the risky grams of like, “What is the likelihood?” And it's a broad guess but–
Joel: I'm going to be a bit more forward and say most recommendations based on DNA are bullshit because–and so, I wrote a book called, “Exploring Personal Genomics.” It was one of the first books.
Ben: “Exploring Personal Genomics?”
Ben: Okay. I'll link to that in the show notes. By the way, just a quick reminder for you guys, if you forgot already, show notes are at BenGreenfieldFitness.com/onegevity. What's the name of that book, Joel?
Joel: “Exploring Personal Genomics.”
Chris: It's a great book recommended to all the first-year medical students at Cornell as well. When Joel's book came out, because we'd met a few years ago, I thought, “This is actually a great book that summarizes kind of–” If the textbook didn't exist, I would have wanted to read it, but it basically summarized everything you could know about–well, if you get your genome sequence, what could you do?
Ben: They have lots of illustrations there and very large words.
Joel: Lots of pictures, yeah.
Ben: I'm in. Please continue.
Joel: So, yeah, I'm coming at this from someone who didn't participate in this. When I was a PhD student at Stanford, I actually was lucky enough to be on the first paper that's published in The Lancet of the first clinical assessment of a whole genome sequence that was used clinically by–it's actually Steve Quake's genome. Steve Quake is a pioneer in gene sequencing technology and entrepreneur. So, this was way back in I think 2010. I think we did this study. So, I've actually been in this business of clinical use of genomics for a long time. And so, where the DNA is really useful is when you have something really wrong with you. So, if you have a rare inherited genetic mutation like a DeltaF508 cystic fibrosis mutation or any of these inborn errors of metabolism or a hard-core cancer, somatic mutation. But when you have those DNA dysfunctions, you probably know before you get the sequence done, right? It's either severe.
Chris: And to run the family off and there's heritability, but it's clear long before you were born.
Ben: That makes sense.
Joel: I think the analogy though is sheet music. So, DNA is just sheet music. It is actually not totally useless in informing health because we won't be measuring it either, but it's sort of like this scaffolding that dynamic measures of blood and microbiome and things need to be put on because it is informative, but it can only explain a very small amount of your health outcome. And it is useful, but it's just sheet music and people need to realize it doesn't tell you how the notes are being played in your body.
Ben: Got it. Okay. So, eventually, exercise and lifestyle recommendations will be rolled out but they could potentially be more comprehensive because of the shotgun-base sequencing?
Joel: And blood testing.
Ben: And the blood testing that goes along with the gut. And correct me if I'm wrong here, but did you say you're already doing the saliva?
Chris: We're launching in probably three weeks, yes. So, basically, you can then order saliva kit into your whole genome sequence as well. Today, what's launched–
Ben: Your whole genome?
Ben: That's different than 23andMe?
Chris: There's important distinction there as well. Almost every other company out there does just what's called a microarray, which means your genome is three billion letters, 3.1 billion letters. And if you're looking in a book, what 23andMe does or a lot of other companies, they only look at about one million of those letters out of the three billion. So, you're really missing a lot of the information across what genome in terms of risk for disease, what you could optimize for performance, or even just traits. And so, we do a whole genome sequencing to get as much information as possible.
Ben: So, at this point then, I've got all this data coming in. You guys are going to be rolling out the exercise, the lifestyle, the diet, the food. You've already got some of the supplement recommendations but though become more and more customized and tailored to me. What's this dashboard actually look like that I log into? Is there an app? Is it like a bunch of graphs? How does this actually work from a user interface standpoint?
Chris: I'll jump in. I mean, a lot of the dashboards that exist are really clunky or non-formative, or sometimes too informative. So, we wanted to make it as clear as possible. So, on that dashboard, you'll see what are your levels of vitamins that we can measure, both that will be in your blood as well that are being produced by your gut, what are some of your general recommendations in risk for constipation, diarrhea, general inflammation. We give centrally what's driving some of the phenotypic scores that people have been reporting and that we can predict from the microbiome. And then, again, we integrate what's present in the blood versus what's being–the sort of molecular potential of what's in your gut or in your genome.
So, you'll be able to see those. There are graphs, there are charts, but they're very clear and very simple. And every sort of chart has–if you want more information, if you want to dive deep into the dozens of papers that support this from the scientific literature, you can see that. If you want to go deeper into how does vitamin B6 are made and what else do we know about it and what is folic acid, you can look at that in details.
Joel: And I think the future roadmap has building and tools into this dashboard. So, people can start to run tests on themselves and trials on themselves because ultimately, you shouldn't believe Onegevity or any company that's giving this information because again, we sometimes are basing this on the best-published literature, and it might come from like New England Journal of Medicine from a huge trial but it doesn't mean in every case, it's always going to work for you. We're going to bring the best science possible. But the whole point of Onegevity too is learning and filling in the gaps and things we don't know, and really, bringing it down to the level of individuals.
So, on the product roadmap, we're going to start to build in the ability to do single N of 1 trial, basically. So, you get this information and there's going to be a lot of things you could do, but you're going to want to build to set your own goals and say like, “Okay. Well, of all these things, I want to improve my sleep or whatever. And here are these things I could do, and I've got this recommendation for X product, whatever. I'm going to start to conduct a trial.”
Chris: And see if it works.
Joel: And so how we do it? Actually, on the academic side, I just published a paper on all the statistics because you're doing a single trial and individual prevents statistical challenges, but there are ways to do it. So, I just published a paper with people on my team on N of 1 trial and all the mathematics behind that. But we can actually wrap that into a very user-friendly experience where you say, “Hey, okay, this is what you want to optimize. This is the recommendation you want to take. Now, we're going to guide you in a very user-friendly way for you to run this trial on yourself and you can see the results and see if it's working. And if it doesn't, our database improves, and then we give you the next recommendation, and hopefully, you can see for yourself that that works.”
Chris: And the importance is everyone has a different baseline. But at a molecular level, what you're measuring for even just your baseline body temperature, we always think that we have–you go to the doctor, you get sort of a standard blood panel, you have ranges. But if you think about it, the ranges in some metrics are very broad. But the power you get from longitudinal data is that your specific baseline for you, for your microbiome, for your metabolites, for the molecular featured in you, once you know them after two or three time points, you can start to know that this is what the baseline is. And so you look for deviations from that.
Joel: Here's a dirty secret all of medical research is nobody knows what normal is or what it means, right? So, every study you say, “We can compare it against healthy normal controls.” But anyway, nobody has a freaking clue of what that word means, how you quantify it. And it's usually, the definition is we exclude people who have the disease versus other diseases.
Ben: Yeah, lab reference ranges are always just one big giant parabolic curve.
Joel: So, there was a paper [01:05:14] ______ last year where they looked at these mental health outcomes that are often used in drug trials. They did a longitudinal analysis and I can pull up the exact paper. And we're actually submitting one from my lab that shows this is true in cardiovascular endpoints that are used. So, basically, if you profile individuals longitudinally and you look at the intra-individual variability of these measures and the inter-individual between individuals, for some of these endpoints that are used in trials within individual, intra-individual variability is bigger. That's a bigger variance spread between subjects, variability, right? So, if anytime you look–so then that calls into question to a lot of clinical trials because within individual variance is much higher. So, this is why, as Chris mentioned, we need to do this longitudinal profiling and we need to set individual baselines for that individual.
Ben: Yeah. Now, people know, if they listen to this podcast, I'm big in the self-qualification. There's got to be like 10 different lab tests that I do during the year from all four corners of the planet, so I'm juggling. I mean, you ought to see in my Dropbox folder, my lab results folder is just all over the place. Now, what I want and what I imagine is that I can get all of this done in just one place and have access to my entire dashboard and all the data in just one place. This kind of begs the question though like what are you not doing? Like, what are things that people might still–I mean, like could you test lime or mold or mycotoxins? Can you do micronutrients? I mean, how deep can you go with this stuff?
Chris: I mean, we've gone really deep in other studies from my lab and Joel's lab. So, this is the core of the platform that launches in microbiome blood and sort of genome, but we can go very, very broad. So, where we want to go next is further optimization of what we recommend for people what they get, more of the genetic tailoring, molecular tailoring of what you get. But there are lots of other things that we've published on that we know can give you measures of health, wellness, and generally even longevity.
So, this would be epigenetic age, how's your epigenome training, which is sort of the small chemical features of your DNA and how it's regulated. So, we've published extensive work on epigenetics now that changes in normal people and also in cancer. So, we want to see that as a metric. We can look at even the clonality of what cells in your blood, which is called clonal hematopoiesis that just means, what proportions of cells are in your blood, and that can also indicate risk for disease, or even sort of blood age, if you will. Basically, to do everything we've done before for astronauts. I think everyone should be able to be treated like an astronaut and everyone should have the right to query any molecule inside–
Ben: I want to, sure as hell, be treated like an astronaut. Be a hero, get a helmet, be able to do the cool–
Chris: And you could have a bus take you to Kazakhstan if you go up in the Soyuz.
Ben: I want this, yeah.
Chris: You can smuggle some things [01:08:15] ______.
Ben: Yeah. I want a monkey.
Chris: We've got a monkey actually in the office somewhere. We'll grab it.
Ben: Not like a chief hooker monkey, but like a real legit, like Russian space monkey.
Chris: Really dignified. But we have a paper just coming out shortly. That's on the NASA Twin Study. So, for those astronauts, we were sending one twin into space and his identical twin, Mark Kelly, stay on earth for a year and we want to measure them for 27 months. And one of the big question in studies, “Well, what changes when you're in space for a year for that long of a mission? What should we look for?” And because fundamentally, we don't know in a sense, we've never sent a human into space for a year, NASA has never done that so we didn't know what to expect, so we measured everything we could in the study, again, which is coming actually, it was genetics, epigenetics, looked at the sort of proteomic changes or protein changes which have small metabolites, microbiome, looked at vasculature and looked at sort of cognitive speed and abilities. And that was really just to see what happens to the body in space.
But we also pioneered new ways to do sequencing in the space station. For the first time, we could sequence DNA in space; do diagnostic assays, anywhere literally in the space station. So, the future I think is to do more of the omics because you'll get more power that way and more time points and make it more personalized so that if you want to do some of the N of 1 trial on yourself at home to test something and quantify and see if it works, we want to empower that. If you even want to do some of the metrics and measurements at home, we want to empower that, like with the one draw device where you can draw blood anytime, anywhere, and–
Ben: What about data from this ring that I'm wearing that does sleep or Apple Health Data? What about rolling that into the dashboard so you get more of the technical self-quant stuff like HRV, body temp, heart rate, et cetera?
Joel: So, a huge amount of my research at Mount Sinai has been–and the wearables, actually, it was part of the founding institutions that work with Apple on the research kit platform, and we've published studies, huge, huge studies using research kits. So, we have a huge amount of experience in that that is definitely coming in and will be pulling that information in. And I think–
Ben: That is coming in for sure.
Joel: Yeah. That's in the product roadmap.
Ben: That's pretty cool.
Chris: We're not a microbiome company, we're not a testing company; we're again a health intelligence company, right? So, we're going to bring in whatever the best information, the best testings. We're going to be on the lookout for different testing modalities because again, we're not a testing company; we're a health intelligence, consumer health solutions company. And even things like your own health records wouldn't be off the table in terms of things we could bring in. I have a huge amount of experience in my lab.
Ben: You mean importing the health records?
Chris: Yeah, from say, Oura or something like that.
Ben: It would be very useful.
Chris: And the deep patient paper came out of Joel's lab, which is on sort of using a lot of machine learning for a large-scale analysis of electronic health record data, and it was with the first paper that really have that scale and that kind of–
Ben: So, theoretically, a guy like me who is very into my health and optimizing longevity I could waltz in and take that clunky Dropbox folder and just like have it all live in one spot?
Chris: Yeah. And have a browserable, have a trackable, and also make it useful. I mean, I think the biggest problem that we've seen in the field and a lot of other companies, and even academic groups, is you make a lot of pretty pictures but there's nothing you can really do with them. So, we want to map if something's going in the right direction or wrong direction and then give you things that you can order or that you can do testing on to improve it.
Ben: This is really damn cool. I wish you guys are doing all of this right now. But you are doing the gut and the saliva pretty much right now.
Chris: Right now, yeah.
Ben: Okay. So, the name of the gut test is Gut Bio? Gut Bio, okay, by Onegevity. So, the way that people could do this is they could go to the website. It's onegevityhealth–
Ben: So, you could go to–if you're listening right now, you could go to onegevityhealth.com. And then, if you click there, you could kind of browse around. Do you guys have any pictures or screenshots there of the dashboard and what it looks like? It's because I know some people want to visualize this.
Joel: Yup. So, you can get examples. You can get a history of how we got involved in the research and our backgrounds as well.
Ben: Yeah. That would be very cool, just to be able to look at what you can't see on the podcast when it comes to kind of seeing what this dashboard looks like. So, I'll link to this website, Onegevity Health, and then these guys have been kind enough to set everybody who's listening up with a discount on the test. So, this test, to do this shotgun sequencing of your gut, is a $349 test. That's where you get the six gigabytes of gut data and the 38 trillion plus microorganisms in your gut. Just use code BEN20. That's BEN20 at onegevityhealth.com and that will save you $20 off of this test. So, use that. You get the kit to your house where the result is like two/three weeks.
Joel: So, that's a lot different. With other companies, it's like 8 to 12 weeks to get results.
Ben: It's a long time.
Joel: It's just too long when things–
Ben: Everybody's always on holiday at the labs.
Joel: Yeah. So, it shouldn't be. And in the future, we'll just do it right at your house and we'd ever customize Onegevity to just pop up results every morning.
Ben: That would be pretty cool, like a Japanese toilet that plays music and has the little heated seat. Yeah. Well, the heated seat and directed flow where if you aim it just right, man, it's a pretty cool experience. I'm just saying. I've spent a lot of time on those Japanese toilets playing around with those buttons, and I've gone some places that I can't talk about on the podcast today because there might be children listening. You go to onegevityhealth.com. The Gut Bio is the name of this test and you use code BEN15.
And as they roll out more tests, I'll keep you guys posted because I plan on doing every single test that they do to start to build my own dashboard as I go along, because I do a lot of research into what companies actually have access to the money, the labs, the technology, everything necessary to actually produce a lot of the things that we talked about on the show today, customized food delivery, eventual blood testing, eventual customized and personalized nutritional supplementation based on your data. These folks are actually positioned to do this at Onegevity, which I like.
So, you're probably going to hear possibly another podcast in the future as more tests get rolled out, but for now, this Gut Bio. But Onegevity is giving everybody who listens in a $20 discount on this test. So, you knock $20 off the test when you go to onegevityhealth.com, like O-N-Egevityhealth.com, onegevityhealth.com, and that's where you can get this gut bio test, which is the first test that they're rolling out. So, the code over there is BEN20 onegevityhealth.com. I am going to put all of this into the show notes, which again, you can get at BenGreenfieldFitness.com/onegevity.
And if you go to BenGreenfieldFitness.com/onegevity, you can also pipe in, leave your questions, leave your comments. I know some of the stuff can get kind of confusing in terms of–it seems like everybody's testing the gut, et cetera, et cetera, but hopefully, we address some of those questions today if you have more though. Go to the website, BenGreenfieldFitness.com/onegevity. Leave them there. And fellas, can I take this in the bathroom now and poop in it and try this out?
Ben: Alright, cool. I'm going to play around with this. And Chris, Joel, thanks for coming on the show, guys.
Want more? Go to BenGreenfieldFitness.com or you can subscribe to my information-packed and entertaining newsletter. Click the link up on the right-hand side of that web page that says, “Ben Recommends,” where you'll see a full list of everything I've ever recommended to enhance your body and your brain. Finally, to get your hands on all of the unique supplement formulations that I personally developed, you can visit the website of my company, Kion, at getK-I-O-N.com. That's getK-I-O-N.com.
Imagine an all-encompassing platform that allows you to keep track of your blood, stool, saliva, and urine testing results, along with self-quantified data from wearables, and even pulls in all health testing you've done in the past as part of a single dashboard.
And imagine that platform could then use highly advanced artificial intelligence to tell you exactly how to eat, how to supplement, how to exercise and much more.
All from the comfort of your home, without needing to drive to an expensive lab for multiple blood draws or fill out confusing paperwork.
That's exactly what the brand new company Onegevity Health has the money, technology and data to do, and so I decided it was high time I sat down with their two chief scientists to learn exactly how this process works. I've already sent in my own stool using their Gutbio shotgun stool sequencing test that we discuss in this episode.
My guests are Dr. Joel Dudley and Dr. Chris Mason.
Dr. Dudley is currently Associate Professor of Genetics and Genomic Sciences, Endowed Chair of Biomedical Data Science, and founding Director of the Institute for Next Generation Healthcare at the Icahn School of Medicine at Mount Sinai. Prior to Mount Sinai, he held positions as Co-founder and Director of Informatics at NuMedii, Inc. and Consulting Professor of Systems Medicine in the Department of Pediatrics at Stanford University School of Medicine. His work, published in >120 peer-review publications, is focused at the nexus of digital health, artificial intelligence (AI), scientific wellness, and healthcare delivery. His work has been featured in the Wall Street Journal, Scientific American, MIT Technology Review, CNBC, and other popular media outlets. He was named in 2014 as one of the 100 most creative people in business by Fast Company magazine. He is co-author of the book Exploring Personal Genomics from Oxford University Press. Dr. Dudley received a BS in Microbiology from Arizona State University and an MS and PhD in Biomedical Informatics from Stanford University School of Medicine.
Dr. Christopher Mason is currently an Associate Professor at Weill Cornell Medicine, with appointments at the Tri-Institutional Program in Computational Biology and Medicine between Cornell, Memorial Sloan-Kettering Cancer Center, Weill Cornell Medicine, the Sandra and Edward Meyer Cancer Center, and the Feil Family Brain and Mind Research Institute. He is also Director of the WorldQuant Initiative for Quantitative Prediction, which bridges prediction methods in finance with genomics. His work spanning >140 peer-reviewed publications has been featured on the cover of Science, Nature, and Cell journals as well as the New York Times, Wall Street Journal, CNN, Forbes, and other major media outlets. Dr. Mason was featured as a speaker at TEDMED and he was recognized in 2014 as one of the “Brilliant Ten” by Popular Science magazine. He completed his dual B.S. in Genetics and Biochemistry (2001) from University of Wisconsin-Madison, his Ph.D. in Genetics (2006) from Yale University and then completed post-doctoral training in clinical genetics (2009) at Yale Medical School while jointly a post-doctoral Fellow of Genomics, Ethics, and Law at Yale Law School (2009).
During the show, you'll discover:
-What Onegevity is, and what it does…9:45
- “Health intelligence” company
- History of the company
- Frustration with how long it takes research to reach the consumer
- 17 years to reach just the clinic; many years after that to reach the consumer
- How is it different from other biome testing companies
- Shotgun sequencing vs. 16S
- Translate the results along with actionable steps to take
- Recreational health information vs. health management companies
- Vertically integrated with Thorne, who develops the actual solutions
- Starting over in some ways in the realm of microbiome testing
-What a metatranscriptome analysis is…18:00
- “Meta” = Across all species
- DNA + RNA analysis
- Challenges with RNA analysis – some clinicians consider it useless
- Only top few % of species generate RNA in the sample
- Comparable in price to shotgun sequencing
-The actual testing process at Onegevity…20:45
- Very small stool sample required, as compared to other tests
- 6 gigabytes of data after sequencing DNA
- Each fragment compared to all known species on earth
- Interactive report on the web; suggestions on what food and supplements to order
- Network modeling:
- How are changes in the microbiome propagating to the blood
- More comprehensive than guessing based on correlations
- What is measured in the blood?
- TBD based on the individual's health
-How Joel and Chris overcome challenges and skepticism in their testing processes…28:20
- How can you measure the whole gut with a stool sample? (Spatial microbiome testing)
- Not necessary, even harmful, to sample from other areas of the gut
-The role of artificial intelligence (AI) in the testing process at Onegevity…36:40
- Predicts diseases/conditions based on the shotgun sequencing
- Recommend foods and supplements to consume
- Each company has its own variables, testing protocols, etc.; results in differing test results
- Database continues to grow with each test; data becomes more reliable
- Data is not for sale to outside parties; will partner under the right conditions
-How you go from stool, to shotgun sequencing, to “don't eat green beans”…40:35
- Internal trial; shocking results
- Can determine the efficacy of the test at certain benchmarks; 15-day, 30-day, etc.
- Very positive feedback from those who were tested
- Does Thorne produce a special probiotic that seeds the gut?
-How comprehensive the dietary recommendations are after a Onegevity test…48:52
- Not simply a particular diet, i.e. Mediterranean, Ketogenic, etc.
- Can make specific food, meal recommendations
- Possibly partner with food delivery services to provide customized options
- Very close to being perfectly genetically tailored in both microbiome and saliva tests
-How customer data is protected if Joel and Chris choose to partner with a pharmaceutical company…55:20
- Customers are opted out by default; can choose to opt in.
- Potential to benefit monetarily by opting in
-What the Onegevity platform looks like from a user standpoint…1:01:55
-And much more!
Resources from this episode:
–Click here to visit OnegevityHealth.com for the testing we discuss in this episode. Use discount code: BEN20 to receive $20 off your own test.
-Book: Exploring Personal Genomics by Joel Dudley
–Kion Organic, high-antioxidant Kion Coffee, Kion Clean Energy Bar, a complete line of supplements for total mind, body and spirit optimization and more! Kion also proudly carries Thorne products, including the multi-vitamin, which I take every day.
–Organifi Gold Wake up each day feeling rested. Relax your body after a long day. Help stiff joints feel young again. Go to Bengreenfieldfitness.com/organifi and use the discount code mentioned there to get 20% off your entire order!
–Thrive Market Your Favorite Organic Foods and Products. Up to 50% Off. Delivered to Your Door. Use my link and get an extra 25% off your first order.
–PuraThrive Keto Balance Gently push your body into Ketosis in a healthy manner so your body turns into a fat-burning machine (and stays in fat-burning mode). Get 15% off your order when you use my link.