The Melanie Avalon Biohacking Podcast Episode #292 - Leo Grady

TRANSCRIPT

(Note: This is generated by AI with 98% accuracy. However, any errors may cause unintended changes in meaning.)

Leo Grady

That's the problem that Jonah was really designed to solve, is how do you interpret 500 biomarkers that we get out of you? The AI then takes that ecosystem that's in your gut and scours literature and matches the patterns in your microbiome to every study. Microbes in your gut are not necessarily good or bad, really you kind of want the right levels and the right ecosystem all together. You know what we're really trying to do with Jonah is allow people to take control of their own health, people who want to really learn what's going on and be able to take action.

Melanie Avalon

Welcome to the Melanie Avalon biohacking podcast where we meet the world's top experts to explore the secrets of health mindset longevity and so much more. Are you ready to take charge of your existence and biohack your life? This show is for you. Please keep in mind we're not dispensing medical advice and are not responsible for any outcomes you may experience from implementing the tactics lying here in.  Are you ready? Let's do this. Welcome back to the Melanie Avalon biohacking podcast. Friends, I am so excited about today's episode. I talk all the time about the importance of the microbiome and there are so many studies out there analyzing the gut microbiome and it can be really, really confusing. How do you make sense of it all? How do you actually know what's happening in you? How do you know what you should do to change things? How do you know what your microbiome actually means? That's why I am obsessed with a company called Jonah. It was such a pleasure to meet Leo Grady and he is doing incredible things with this company. As we talk about in the show, they actually created an AI that analyzes all of the credible studies out there when it comes to the gut microbiome and then you do a gut test. They get a sample of your microbiome and they evaluate it based on all of those studies. It is so, so cool. The report you get is so deep and so detailed. It's really eye opening when it comes to the implications and meaning of your gut microbiome.  When it comes to your health, I cannot recommend it enough and you can get $50 off when you go to Melanie Avalon.com slash Jonah and use the code Melanie 50. So that's Melanie Avalon.com slash Jonah with the code Melanie 50 for $50 off. And in this episode, we talk about so many things. We talk all about gut microbiome testing, the future of personalized health, the role of digital twin studies, how to optimize your health conditions and so much more. I can't wait to hear what you guys think. Definitely let me know in my Facebook group, IF Biohackers, intermittent fasting plus real foods plus life. Comment something you learned or something that resonated with you on the pinned post to enter to win something that I love. And then check out my Instagram, finally Friday announcement post. And again, comment there to enter to win something that I love. These show notes for today's episode will be at Melanie Avalon.com slash Jonah testing. Those show notes will have a full transcript as well as links to everything that we talked about. So definitely check that out. All right. I think that's all the things as a brief reminder, you can get $50 off Jonah when you go to Melanie Avalon.com slash Jonah and use the code Melanie 50. And now without further ado, please enjoy this fabulous conversation with Leo Grady.  Hi friends. Welcome back to the show. I am so incredibly excited about the conversation I'm about to have. It is a very, very long time coming.

Melanie Avalon

And I am just overwhelmingly thrilled about this resource, which is something friends I have been searching for not my whole life because I was not aware of all of this my whole life. But as long as I can remember being in this health sphere, I've been looking for something like what we are going to talk about today. So I'm so excited.  So the backstory on today's conversation, I recently or not that long ago, aired a third episode, I think, or second or third episode with Dave Rabin at Apollo neuro, which is a company I also love and adore. And Dave introduced me a while back to this incredible human being Leo Grady, who I'm here with today for his new company, Jonah. And when I heard the elevator pitch about Jonah was I was just so excited because friends, as you know, I am obsessed with the gut microbiome. I've been talking about it for years and years. I'm always trying to optimize it. I'm haunted by how different diets affected and supplements and all the things. And there are so many studies out there on the microbiome. It's really, really difficult to know anything about it. So I've always been a little bit confused by the studies wondering how to make sense of it all. That's where Jonah comes in.  They are a company that developed and trained an AI platform to actually, well, A, you test your microbiome. So we're going to talk about that. And then B, this AI actually, well, we'll talk about what studies it looks at, but it looks at the data out there and it's able to interpret your microbiome on a very personal level with just the latest studies and all the things and hopefully correctly, which we can talk about. So yeah, it's awesome.  I did it. I like got the kit and did the test and the platform is so enlightening. So I have so many questions for Leo. Leo, first of all, thank you for what you're doing. And second of all, thank you for being here.

Leo Grady

Thanks, Melanie. It's great to be here.

Melanie Avalon

Before we jump in, I actually have some questions about your bio. For listeners, you earned a PhD in cognitive and neural systems from BU.  You've worked at a lot of different companies, so you worked for 20 years pioneering Bay Area startups like Heartflow, multinational medical companies. My question about Siemens is when you worked there, did you guys have the hookup at Epcot? Because I remember every time we would go to Epcot, projected on the ball after the firework was like something, a Siemens company.

Leo Grady

Wow, wow. Yeah, that's a deep cut.  Yeah, the Siemens, I actually don't really understand what the connection was with Epcot. But I actually, they brought a camera crew in to our division at Siemens. And they recorded a bunch of people and they recorded me. And somebody told me actually several people told me that they went to Epcot and they saw my face like on the screen like talking about science and the magic of technology. So I never saw it myself. But it's hilarious that actually made that connection.

Melanie Avalon

That's so funny. Wait, did they tell you it was like for Epcot?

Leo Grady

Yeah, they did. They did.  But you know, we had like camera crews coming through like, from time to time, because, so the division that I worked at Siemens, so, you know, my PhD is in cognitive neural systems, but it was a neural networks department. So it was really like AI. And so our division was all the AI PhDs were all together in Princeton. And they basically, this is where all of the divisions of Siemens came to get advanced technology, right? So Siemens made MRI machines, CAT scanners, ultrasound, and really any imaging equipment. And then they started branching into diagnostic testing. So when they needed AI to understand these images, to help plan surgeries, do whatever they came and came to us. So we were kind of the show pony at Siemens of like all the advanced technology, all the really fun, sexy stuff that was going on at the company. So they constantly wanted to promote us one way or another.

Melanie Avalon

That is so amazing. I'm so glad you had that answer because when I was asking it, I was like, he's probably gonna be like, what is she talking about?  Okay, that's incredible. And then my other question, I have two. You won the Edison Patent Award. What was that for?

Leo Grady

Well, as you know, Thomas Edison was one of our great innovators and the state of New Jersey, where Edison worked, has a patent award that they do every year in his honor. And so they basically pick patents in, I think it's like seven different areas, one of which is healthcare. And the patent that I'd done that year actually got picked, because there's this really new AI method for identifying tumors in radiology images and try to extract this information out of these MRIs and these CT scans. And so this got selected for the Edison Patent Award.  So it was fun. It was like a black tie affair, had to put on a tux and do an acceptance speech. And I have this bust of Thomas Edison on my mantle now.

Melanie Avalon

oh my goodness that's incredible congratulations thank you that's amazing i don't even know the patent submission process is it like just a lot of paper not just but is it a lot of paperwork is that like is that what the process is

Leo Grady

Well, you have to have an idea, right, to get patented. So you have to invent something. You know, I'm an inventor on over 300 patents, all in AI and healthcare. So I've done it a lot of times. And it's a process. I mean, you basically have to write it up and write up your idea. There are certain requirements for how you have to write it up. And then you give to the lawyers, and then they translate it into legalese and they submit it. And then the patent office comes back and says why certain things have been done before you argue with them and then eventually you get a patent. In a nutshell, that's that's the process.  So it's more than paperwork, but the part that you do as the inventor is not so much paperwork, but it is a lot of writing.

Melanie Avalon

Yeah, so I guess what I was getting at is, is it just the idea? Like, do you actually have to have anything in use? Or is it just the idea? But do you have to like prove the idea would work? Like, can you just get a patent for any idea?

Leo Grady

It's a really interesting question because in the old days, you actually had to build the thing and like bring it into the patent office, bring your contraption into the patent office. You don't have to do that anymore. You don't have to have actually even built it. And it doesn't have to be a good idea either. So the patent office doesn't care if you have like a really terrible idea. You know, they're not judging the merit of your idea. What they're judging is, is it new? And did you fully describe it so that somebody, some ordinary person that's skilled in that area of expertise could read your patent and implement it? So that's the requirement.  It has to be new. And you have to fully describe it such that somebody could go build it. But you don't have to actually build it yourself. And it doesn't have to be a good idea. It just has to be a new idea.

Melanie Avalon

That is so interesting. I want to get a patent now for something.

Leo Grady

I can help you, Melanie.

Melanie Avalon

Okay, perfect. I'm gonna like brainstorm. I remember in high school, there was some kid and he got a patent for some sort of like, I don't even remember it had to do with cars and wheels and frictionless something and we just thought he was like the coolest thing ever. Wow. Okay. In high school. Yeah. And then okay, I think that's the majority of my questions.  You were also inducted as a fellow in the American Institute for Medical and Biological Engineering. That sounds fancy.

Leo Grady

You know, we have a number of these like American College of Engineering, you know, the American Academy of Sciences, there's one for medical and biomedical engineering. And it was actually it was held at the American Academy of Sciences, the building in Washington. So you know, this is again, this is one where you have to be nominated by one of the fellows. And then again, they have an award ceremony, and then they participate in some national discussions related to medical and biomedical engineering.  So it was really great to be inducted in there. And it's an amazing group of engineers and scientists and innovators.

Melanie Avalon

So, so cool. So many things. Oh, wait, one more thing. You also received the first ever FDA approval for an AI product in pathology. Does that relate to Jonah at all? Or is that something different?

Leo Grady

No, it was from my previous company, Page AI, which was a spin out of memorial zone Keter, was building AI for tissue analysis for pathology. So, you know, if you were to get like a biopsy, or you get tissue taken out from surgery, you know, they put it on slides. Typically, a doctor looks at that under a microscope to make a diagnosis and say, there's cancer here, there's not.  What we did at Page was build an AI system that could take a digital picture of that tissue, a microscopic picture, and detect cancer. And so this FDA approval was for the detection of prostate cancer from a biopsy. And we found that when pathologists were using this AI in our clinical trial, there was a 74% reduction, I believe, in missed cancers when they use the AI, and a 26% reduction, I believe, in over called cancers, and they were faster. So basically, any doctor that was using this AI technology, just was much more accurate with prostate cancer detection, even if they were like world experts, like world experts, and offices all got better using this technology, and they were faster doing it. So that was the first and I believe still the only FDA approved AI system in pathology.

Melanie Avalon

Oh wow, prostate cancer, that's the most common cancer in men, right?

Leo Grady

And man it is yeah, that's right

Melanie Avalon

Is it over-diagnosed or under-diagnosed in general?

Leo Grady

Well, it's one of the big four, right? So there's prostate cancer, breast cancer, lung cancer and colon cancer.  And prostate cancer, typically what happens in the US is men of a certain age, they get a PSA test. And if that test is high, then they go and get a biopsy, where they basically stick a needle through the perineum into the prostate, sample multiple different cores, take those cores out, and then look at them under a microscope. And if they detect cancer, then you know, the man starts to undergo treatment. But the problem with PSA is it has a lot of false positives. And, you know, getting biopsies in your prostate is no fun, does not sound fun. It's not fun, I'm told. And so it's something that, you know, if you go and get it, though, you want to make sure you catch the cancer, if it's actually there. But don't also call it wrongly, if it's not there. 80% of those prostate biopsies are negative, which means that you have a high PSA, they do do all the biopsies, and then they don't find anything.

Melanie Avalon

Okay, that is super cool. Which actually brings me to a huge question I have. Like when did that idea, speaking of ideas, when did that idea first come to you?

Leo Grady

I've been excited about the microbiome for a long time. My father was a virologist, so I grew up around microbiology. My path led me into AI. But over time, I kept reading all these articles about how the microbiome was linked to autism, or it was linked to cancer, or it was linked to depression, obesity.  And people were getting these stool transplants, these microbiome transplants, and in some cases, seemed to be even able to cure some very difficult chronic diseases. So something that I've been intellectually interested in as a hobby, you know, I make beer, I make wine, I make yogurt, kombucha, all those sorts of things. And my ex-wife has Crohn's disease. And so, you know, we were together through her whole journey. And I really saw, you know, the microbiome plays such a key role, or we believe in the onset of that disease and some of the challenges that are associated with it. So something I've been around for a long time, and something I've been excited about for a long time. And then I took some microbiome tests myself, just out of curiosity. And I was so totally disappointed with the results that I got, which would basically, you know, tell me my guts were six out of 10, or something like that. And then tried to sell me probiotics. And it's like there was just a complete disconnect between results I was getting, and all of this amazing science that was getting published.  And I realized over time that AI was really going to be the key to bridging that gap between the incredible complexity of this data, and the complexity of the literature.

Melanie Avalon

I, okay, I love this so much. Well, not that your ex-wife had Crohn's disease. Did that get better for her?

Leo Grady

No, unfortunately, she's really kind of taken a pretty typical path of stabilizing for a while, getting worse, stabilizing for a while. But she did end up in functional medicine, and that combination of diet and supplements really made a big difference for her in a way that just one drug after another was not.  And so I did see that even with these chronic diseases, like these really intense autoimmune diseases, that there was a big role for diet, lifestyle, supplements that could make a material difference in somebody's lives, but that was not being recognized by mainstream medicine. And again, the microbiome seemed like a key link to be able to understand what was going to work for somebody on an individualized basis.

Melanie Avalon

I'm just so excited about that because I feel like the, like you mentioned, the functional world is very in tune with the role and importance of the microbiome. I feel like the conventional medical system does approach it much more from just a pill perspective or colonoscopies or procedures, and I feel like you're creating this incredible bridge where people can take agency into their own hands and do their testing and actually get like a valid interpretation, which speaking of, so the role of AI, which is clearly your background here, a general question about AI as it is right now.  I feel like in the past, past year or so, I'm trying to think when I started using chatGPT.

Leo Grady

probably two years ago.

Melanie Avalon

Yeah, yeah. So like, it's become, I feel like for the average person who's not in that field, like we weren't aware of AI, like we knew about AI, but it wasn't in our face all the time. We weren't using it as a consumer actively just on our own on the internet and things like that. I mean, I'm sure we were, I'm sure it was like involved in what we were doing, but it wasn't in your face like it is now with chat tpt. And it feels like it's just rapidly expanding and escalating.  So two questions there. One is, is that an accurate timeline perception? Like what I just painted where it was not as evolving as fast as it was. And now it's just like massively evolving. Or has it been in the background? And we're just now seeing it from a consumer perspective.

Leo Grady

Yeah, I think it's much more the latter than the former. You know, as you mentioned, I did my PhD in AI 20 years ago. And I've been in this field for a long time. And, you know, when I got my PhD was around the time that IBM Deep Blue beat Kasparov in chess, right? That was 97. I started my PhD in 99. Right. And at that time, people were like, Oh, my god, computers can do everything like the things are gods, they're going to take all our jobs was like, crazy, right? And that was, whatever, 25 years ago, almost 30 years ago. And then that died down.  And then in 2007, I think Watson beat Jeopardy. I don't know if you remember that. And basically, for the first time, an AI could be could win Jeopardy and like beat the world champion. And there was a thought that like, Oh, my god, if it can win Jeopardy, it can do all this amazing stuff. And, you know, I think people get overexcited when they see something, and then they realize, like, it's not quite as great as they think it is. But then every year, it gets better and better and better. Right. And so, you know, for example, like, one of the most shocking things to me about AI was when Apple allowed you to unlock your face, your phone with your face, right? Now, you probably didn't think of that as AI. But, you know, way back when there were whole conferences on face recognition, like people would like go to a city, they would publish there would be a whole conferences on face recognition. This was like one of the great unsolved problems of AI. Now, you know, you take a couple snaps of your face, and you can unlock lock your phone and it works really well. It works really well, different lighting, you know, different hairstyles works really, really well. And that blew me away, actually, the first time I saw it.  With these language models, you know, we see, you know, more and more evolution. I mean, chat GPT was sort of a UI interface on GPT-3, which was an improvement on GPT-2, which was an improvement on the original GPT, which was an evolution of the transformer paper and 2017. And so all this stuff has been evolving. I think what shocked me was that by making, you know, chat GPT accessible with this UI, that it just really blew up because it put a lot of these tools that have been evolving for a while in people's hands in a very simple way. And I think, you know, to some degree, there's been an overreaction to, you know, people think like, Oh, my God, chat GPT can, you know, write a recipe or write a poem or whatever. And so it's going to take my job, or it's going to do this. And I think, you know, people still need to understand like what this technology is, what it can do, and what it can't do. But there's no question that year after year after year, it gets better, and we can do more and more stuff.

Melanie Avalon

I'm just reflecting on how, because people do have those concerns about job security, and I feel like you're in the place to have job security, working with it. Has it passed the Turing test?

Leo Grady

Oh, yeah. Yeah. I mean, the Turing test got passed. I mean, there are different variants now. But I think the first time people accepted the Turing test was passed, it was like about 10 years ago or something like that.  Really? Yeah. Do people debate that? Not anymore. At the time, it was questionable, right? But there have been these systems that have been able to fool some of the people some of the time, since I think there was a system called ELISA, about 20 years ago, that was able to fool a bunch of people. And yeah, now I think people pretty much agree that the Turing test has been passed. But what people will say now is like, well, the Turing test is just not a real accurate reflection of intelligence. And these things still aren't smart. And so people get into these like philosophical debates. But the Turing test as defined, like, can you fool a person? Yeah, I think you can fool.

Melanie Avalon

Okay, gotcha. Last general question about AI.  What do you think about, didn't they recently, like this week or last week, we're talking about how maybe it's tapping into the multiverse? Do you have thoughts about that? Did you see that?

Leo Grady

No, I don't know what you're talking about

Melanie Avalon

Oh, okay. Nevermind.  I'm referencing. So apparently there's some computer that I'd have to look it up. However fast it's doing what it's doing would have taken like some number I don't even know the name of because it's like so many years. It would take a normal computer to do that and it did it real instantly. And so there's a theory that it's like tapping

Leo Grady

Are you talking about quantum computing?

Melanie Avalon

Maybe? I'm not sure.

Leo Grady

Maybe this is just Dr. Strange's computer.

Melanie Avalon

We can, I can, I can look it up and we can circle back. Okay.  So applying this all to the gut microbiome. So you were excited about the concept of Jonah. How did you just initially start the process? Did you start with training an AI model? Did you start with building the actual like consumer model of getting the testing kit? Like how did you go about creating the company?

Leo Grady

Yeah, it was really initially to focus on the AI model itself, you know, we, we work with a lab that does testing. But the the testing part is, you know, we do the most advanced testing, we do advanced metagenomic sequencing, we do really deep sequencing, that part's not so special, there are a lot of labs that could do that. The question is that after you do all the sequencing, you have this incredible amount of data. And what do you do with that?  Right? And so that's the problem that Jonah was really designed to solve is how do you interpret, you know, 500 biomarkers that we get out of view. And by the way, there's a library of about, I believe, 44,000 different organisms that are known to humanity that we might find in your gut. And so the 500 we find in your gut may be very different than the 500 we find in my gut. So there's this vast universe of things that we could find. And then the ones that we do, you get this tremendous amount of data out of it. And so the question was, how do we interpret that? And take advantage of these, all these studies are coming out, every month, there are more than 2000 studies in PubMed on the microbiome. How do we, how do we take advantage of that? So in building the company, we started from the AI model. And what I did is actually hired some people to start annotating papers to train the first version of the model and, and even learn, you know, what paper should we be including? How should we be putting these studies together? How do we rank quality of these studies? And there's a lot of nuance when you're looking at the microbiome, right? Because for example, one study might be about lactobacillus acidophilus, right? And then another one might be about lactobacillus, right? So like the genus versus the species. So you have to build into the AI that one is a subset of the other, right? And one paper may have studied Crohn's disease, but another studied inflammatory bowel disease, another said autoimmune disease, and you have to know that Crohn's is a type of inflammatory bowel and inflammatory bowel is a type of autoimmune. So there was a lot of work that needed to get it to go in to like actually building this AI model, so that we could interpret the literature and then interpret all the data that we're getting out of these tasks.

Melanie Avalon

Yeah, I mean, it's just it's really overwhelming to hear. Do you use AI to train the AI like to use programs to go through and do stuff to then train the program?

Leo Grady

Yeah, yeah. And that's exactly what I've done at all my other AI companies. So what you do is you basically, you build some initial AI model off of like a small amount of training data, right? So like, if I take my last company page, we might have had pathologists like mark cancer in tissues and tissue images. And then you train the first AI model on that. And then you run that AI model on a bunch of new images, a new new tissue models that that nobody had annotated before. And then you have the doctors go and correct those and say like, that wasn't actually cancer, this was cancer. And then you keep iterating and you build better and better models until it works.  Or at least like the amount of time it works is or doesn't work is really small, you can characterize it. And then you can really start using the model. And so we apply the same approach with the these microbiome studies where we had people go in and annotate these studies and say this is a study about Parkinson's disease, it had 500 people with Parkinson's 500 people were healthy, the people were this was, you know, done in China, it had, you know, 75% women, 25% men of these age ranges, we found that the Parkinson's population, their organisms, A, B and C were high, D, E and F were low. And it was done with this kind of sequencing, it was done in humans. And so there's all this information we have to extract from the studies. And so we need to make sure the AI can recognize that this is a microbiome study to be looking at in the first place. And there was on humans and it meets all the criteria. And then it has to pull all that information out accurately, and put it in the knowledge base. And so that was the key part from a technology build standpoint of having people annotate these papers, building the AI to extract all that information, and then being able to try that on new papers and really iterate and make sure that you know, the AI was getting it right at a broad scale.

Melanie Avalon

So you mentioned testing in humans. What is the inclusion criteria? How does it decide which papers make the cut?

Leo Grady

And so this is something we designed into the system. And what we designed in was that paper said to be peer reviewed in PubMed from credible journals, they had to be in human studies and humans. They had to be done with sequencing technologies to know like culturing studies or whatnot had to be in subjects age five and older. And I think that's it for the criteria.

Melanie Avalon

Does it check for things like if a study was powered or not to find the results that it found?

Leo Grady

Yeah, so we only look at statistically significant findings in each study. So the AI just ignores anything that's not statistically significant.  But every study that passes that inclusion criteria gets a quality score. So the quality score of all the included studies is a function of the size of the study, the cohort size, and the quality of the journal it was published in.

Melanie Avalon

Okay, awesome. And then so this is a huge question I have.  And it goes back to what I was talking about earlier with the consumer perception of AI and relating it all to these language models that we have today, like chat GPT versus all the other versions and things of AI that are happening. Because with my experience, I realize they're like literally not the same thing. But like when I when I use chat GPT, for example, to look at studies, I think it's I don't know if it has ever correctly given me information, like it just, it gives me studies, but it's just wrong. Like it reads them wrong, it makes up studies, it's like kind of shocking. So that that issue that it has, is that a common issue in training a model where it like hallucinates or makes up things? Or is that just the nature of chat GPT in that type of AI model?

Leo Grady

Yeah, it's the second. So chat GPT is, you know, building a general purpose foundation model. And it's also it's also a generative AI to write. So when you ask it a question, you say generate a summary for me. And it will generate something that it thinks is a summary of that study.  We're our AI model is designed differently. Our AI model is really tailored to find specific information in these studies, like first of all, to judge whether this is a relevant study or not. But if it is a relevant study, to really identify all the key aspects of the study, the cohort size, the demographics, the findings, the diseases that were being studied or the conditions, and then to put those into ontologies and the hierarchies of information that, you know, can be used to interpret somebody's microbiome test results. So, you know, a chat GPT is a generative model is going to be prone to hallucination, because you're asking it to create summaries, you're asking it to create, you know, responses to your prompts. For us, this AI model is really laser focused on extracting specific kinds of information from unstructured text data, and being able to do that super accurately. So it's a different kind of model. That's really fit for purpose for what we're doing. So it's not as general for you can't talk to it in the same way that you can talk to a chat GPT. But it also means it's not hallucinating. And it's very, very good at the thing that it does.

Melanie Avalon

Okay, so it's not making stuff up.

Leo Grady

It's not making anything up.

Melanie Avalon

I'm not going to get in an argument with it like I do nightly.

Leo Grady

You can yell at it, but we won't yell back.

Melanie Avalon

Okay, yeah, chat TV doesn't yell back, but it talks back. And yeah, I swear I'm on its list when it takes over.  It's like, we got to take that girl out. Okay, so, okay, so other other questions here. So when you okay, well, so first of all, the process of a person using Jonah, you mentioned earlier, like sequencing studies and things like that. So what is the actual method of like stool collection? And I mean, I went through this, oh, and my comment on it, which I think I told you at the time was I was very impressed with because I take a lot of different like at home DIY test yourself type situation scenarios having this show, as you can imagine, the process with Jonah was like so approachable, very well done, like a plus a plus I made it really, it was sometimes the instructions are a little bit confusing. And yeah, it was just it's really nicely implemented.  So when a person does that, that at home test, what is it testing? What type of how do you interpret it? And then what is the AI doing with that information?

Leo Grady

Yeah, so here's the process. So as you know, and thank you, we put effort into making that process, the sample collection processes as easy and smooth as possible. So yeah, you take a stool sample, you ship it off to the lab, do deep sequencing metagenomics on that sample, identify everything in your gut down to a strain level. So it's not just bacterias, it's fungus, it's viruses, protists, archaea, parasites, anything that's in there. Then you get this big long list of different species, probably about 500 different species that are living in your gut.  And the AI then takes that ecosystem that's in your gut and scours literature and matches the patterns in your microbiome to every study that's ever been published. So that we can say for you, Melanie, your microbiome fits the profile of everything from bloating and brain fog and fatigue, all the way to psoriasis or obesity or Parkinson's or Alzheimer's or even cancer. Then the AI goes a step further and reads all the interventional literature on how you can change your microbiome through diet, lifestyle, supplements. We know, for example, from all those studies that if you go vegan, it changes your microbiome in a certain way. And so what we do is we build a digital twin of your microbiome, knowing your microbiome today, and we apply that vegan shift to your microbiome and say, what does Melanie look like if she's vegan? So we make a virtual vegan you where we apply that shift to your gut. And then we do the same thing with a keto Melanie or a gluten-free Melanie or Melanie on some supplement. And digitally, virtually, we give you every dietary change. We give you every lifestyle change. We give you every supplement. And based on all of that analysis of your digital twin, be able to pull together what's going to work for you to really make the changes in your gut that are going to most impact your health. And then that result comes back to you through both an interactive interface on the internet and also a PDF summary.

Melanie Avalon

The online platform as well is very, very cool and it has so much information and it's very easy to navigate.  So to clarify, the recommendations that it is providing are treating the microbiome, not the condition, right? Or they're treating the microbiome to change the condition.

Leo Grady

Yeah, that's right. I mean, just to take a very simple example, let's say E. coli high was linked with some disease, right? And if we wanted to reverse that signature for E. coli high, and there was a dietary change that would lower E. coli, then we would basically put the two of them together and say, hey, if you want to lower E. coli to reduce your link to that signature, you know, take this action, right?  Like, effectively, that's what it's doing. Like, more or less reverse the different elements of your microbiome that are out of whack that are related to a variety of different diseases. But you're right, it's not treating the disease through any means other than trying to tell you how to fix these issues in your microbiome.

Melanie Avalon

And then in the results, when it's like giving you a score and things like that, is that that's comparing you to the general population or other Jonah users? I guess the general population or the studies.

Leo Grady

It's comparing it to the study, right? So like, let's say you have a pattern in your microbiome that's been linked with eczema, right? So that means that there, you know, say 20 studies on eczema that have identified a certain pattern that's fairly consistent, and then we find that pattern in your microbiome. And if we find parts of that pattern, but not other parts of the pattern, then you'll have a moderate score, right?  So if you have a score from like one to five, basically, like mild, you know, moderate, high, right? And if it's high, it means you fit that pattern that's been described consistently in the literature. If it's moderate, it means you partly fit the pattern that's been described in the literature. And if it's mild, it means that some elements of that pattern are present in your microbiome.

Melanie Avalon

Okay, awesome. And I was looking through my analysis online. I didn't realize there's an option to get the PDF. So now I'm looking at the PDF. This is really, really helpful.  We mentioned earlier, talking about the research quality and things like that. So for findings where the research level is limited versus early versus established, how much should we take that into account when we're interpreting the result?

Leo Grady

So, we tell you with the analysis how much, how well established the evidence is. And for some diseases, some conditions, there are a lot of studies. I mean, there can be hundreds of studies on the links between the microbiome and that condition of that disease. In other cases, there may be five or 10 studies.  So we're always telling you, and the same is true with the actions, with the digital twin side of things. Sometimes we don't have a whole lot of evidence on how to change your microbiome, but maybe there are a few studies on how to do it. So we're always trying to really represent the current state of science to you, which is constantly changing. You know, at the end of the day, it really depends on how you're using the study, like how you're using the test results. Yeah, there are a lot of people that come to Jonah because they're having some kind of health issue. They want to see if there's anything in their microbiome that might identify a root cause, how they might change that, and how they might reverse those signatures.  And so that's one group of people that have found Jonah really useful. There's another group of athletes, biohackers, people who don't have health issues but want to perform better, look better, feel better, that are really looking for ways that they can up their levels of health. And then there's a third group of people that are just general wellness curious, that just want to do a checkup and make sure that everything's okay, but they don't necessarily want to fix a problem. But if they see something that's really wrong, they want to know about it. So, you know, depending on which of those three groups that you fit in, like the people that have a health issue, the athletes, biohackers, or the sort of general wellness, I think it depends on how you use the information, but you're asking sort of a different question each case.

Melanie Avalon

Okay, gotcha. And some other questions. And it's just so exciting. I love getting information like this, so it's really, really exciting.  And the people's personalized reports, because you mentioned that there's like 44,000 potential things that could be down there, and we have around 500. When you include the actual list of organisms in the people's report, are those the ones that were found? Or are those the same for everybody and you put if they're there or not?

Leo Grady

Yeah, we, we allow you to look either way, right? So there are filters that you can apply. So you can say only show me the things that were present in my gut. So you can also say show me everything that could be present in my gut, you can filter it by bacteria or fungus or pathogens, all sorts of different things. So we really allow you to dig into the specific organisms if you want to. And you can see as much of that or as little of that as you want.  And the same with all of the different conditions and the action plan. With any of those, you can dive deeper and you can see the organisms that are involved in the signature. And you can see all the supporting studies and the supporting science for where that signature was developed from. And you can even link out to pub med and see the actual, you know, source papers if you want to.

Melanie Avalon

And also in your results, just as far as interpreting it, one has the sections where it has organisms and then it has the relative abundance and then it has your strongest associations. Is it that the lack of that, the absence of that is what creates the association or the presence of it?  So for example, I'll just give you an example. So like for me, first one is, I can't pronounce like any of these, like anero stipes, I guess. And it says 0.03%, which is below. And then it says your strongest associations and it lists like all these things, osteoporosis, leaky gut, stress, hypothyroidism, abdominal pain, et cetera. So is it saying that my low, like if I were to increase it, that would help those conditions? That's right.

Leo Grady

Yeah, exactly. And there's certain organisms that when they're too low, it's a problem. Other organisms that are a problem when they're too high, there's some organisms that seem to be a problem when they're too low or too high. So for example, Acromancia super low levels have been linked with things like Crohn's disease and other GI issues.  However, studies are now finding that high levels of Acromancia have been linked with Parkinson's. And so it's always important to realize that these microbes in your gut are not necessarily good or bad. It's really kind of want the right levels and the right ecosystem altogether.

Melanie Avalon

Okay, so is the chart, because now I'm seeing that pattern, I've seen that pattern, I gotcha. So like, if it says there's an average, if I'm average, it doesn't list associations, but if I'm below or above, it lists the association.  So is that chart taking that into account that for some, it's only if it's below or above, that it would have that association?

Leo Grady

Yeah, yeah, exactly. But, you know, remember, it's important to, to keep in mind that these different associations are usually not linked to a single organism, they're usually linked to a collection of organisms.  So, you know, you mentioned osteoporosis, if you're interested in whether or not you've got the signature of osteoporosis, my guess, I don't know that one offhand, but my guess, it's probably a couple dozen organisms that are involved in that. And so osteoporosis is called out at a separate area of your report. And so if you have, you know, an aggregate, all the organisms that, that fit that pattern, then that'll get called out. I wouldn't worry so much about like a single organism being linked to osteoporosis as much as, or any condition, as much as like that whole ecosystem shift.

Melanie Avalon

Gotcha. Okay. Oh, and yeah, and I'm realizing, oh, there are some where it says average, but it links to it. So that I'm just guessing the presence of that in general, links to that condition.

Leo Grady

Yeah, you can because you have to remember that a lot of these organisms like the normal amount and a normal person is zero.

Melanie Avalon

Hmm, okay. Gotcha. Oh, yeah, yeah, yeah. Okay.  Okay, major question. And you and I have talked offline about this. What should people do if their presentation of disease or condition doesn't match the results? And why might that be happening?

Leo Grady

What's important to remember that these are not diagnostic tests, right? So you should not be looking to Jonah or any microbiome tests at this point to get a diagnosis of any disease or any condition.  I think the best way with the current science to think about this is as some sort of indicator or risk factor, right? So, you know, if you think about genetics as an analogy, you know, there are certain genes that have been linked with breast cancer, like the BRCA gene, right? But if you find out that you are, you know, have a BRCA variant, you have a breast cancer gene, it doesn't mean you have breast cancer, it doesn't mean you'll ever develop breast cancer, but it is something that you should be aware of because it means that your likelihood of developing breast cancer is higher. And the exciting thing for me about the microbiome is that unlike, you know, a genetic predisposition towards something, you can't change a genetic predisposition, at least not now, whereas with your microbiome, even if you don't have a manifestation of that disease or, you know, symptoms of that disease, you know, if you have this signature in your gut, it is something that you can change, it's something you can reverse as a precaution, as a way of preventing it and reducing that predisposition.

Melanie Avalon

Awesome. How often should people, because I feel like the microbiome, it's interesting and I'm curious your thoughts on this.  I feel like on the one hand, it can change so fast. And we've seen that in studies about how just how fast it changes. But then on the other hand, people seem to they'll have some sort of incident inciting incident, and then they can't seem to get back to the way they were before. So it's not, you know, on the one hand, there's these like changes that seem to happen that you can't, you know, seem to get out of. And yet it seems like we can really rapidly change the gut microbiome. So what are your thoughts on the transiency and the changing and the fluidity of the microbiome?

Leo Grady

Yeah, it's super interesting. You know, there are people that have done studies where they sample their microbiome every single day. And they'll do it for a year, two years, I think Rob Knight's been doing his for whatever 10 years. And, you know, I think about this study, there was a study done in Boston, a few years ago, where these two guys, these two researchers sample their microbiome every day, and they have these charts in their of their microbiome composition. And what you see is it is it jitters around every day, like every day is not exactly the same, but it's pretty stable from day to day to day.  But then in one case, one of these guys goes from Boston lives in the Philippines for three weeks. And during that period, that he's away, his microbiome totally changes. And it goes into like a completely different state while he's in the Philippines. And but when he comes back to Boston, it more or less reverts to where it was before, before he left to go to the Philippines. Whereas the other guy never goes to the Philippines. But one day he gets salmonella, like food poisoning in a restaurant. And on that day, when you look at his microbiome, it just goes completely haywire. And for the next week or two, it just all over the place. It's like changing every day, just like flipping around. And eventually, it stabilizes, but it stabilizes into a different state than what it was before. Right. So in that case, in both of these cases, you know, somebody leading their life day to day, changing their diet a little bit getting a little bit less sleep, they're a little stressed, their microbiome bounces around, but it's pretty stable. But when you have these incidences, either going and living in a foreign country for a while, or, you know, effectively an invasive species in your in your gut, through food poisoning, you know, that can have a very profound change on somebody's microbiome, I can even have a permanent change on the microbiome.

Melanie Avalon

Yeah, I think, I mean, I know a lot of my listeners have probably experienced that. Have you ever had an inciting incident of sorts?

Leo Grady

You know, I have not I have not I'm actually fortunate that I've always had really good health. But I was also lucky, you know, because my father was a virologist, he was very opposed to antibiotics unless absolutely necessary. And he did it from a public health perspective, he was very worried about superbugs, and that sort of thing. So he felt that it was like being a good citizen to avoid antibiotics unless you absolutely needed it.  But we know now that, you know, the number of courses of antibiotics that you take, especially as a child, has been directly proportional to all sorts of different problems, whether it's obesity or autoimmune diseases, just pretty, pretty shocking. So I'm lucky. I think I took one course of antibiotics when I was a kid, and I had pneumonia. But no, I've never had super serious food poisoning or anything like that. Has it happened to you, Melanie?

Melanie Avalon

Yes, and wow, like only a hand, you know, like one or so courses of antibiotics when you're a kid that I'm that's so incredible. I was literally on antibiotics ongoing for my skin for a while, which I like shudder about because I ended up doing Accutane which I also kind of Shutter about although it really helped my skin So I don't regret it, but I know it has a lot of issues with it But um in order to be on Accutane I had to be on birth control for wait birth control and I had to be on birth Control to be on Accutane and to be on birth control.  I had to be the gone antibiotics I don't know. I was on antibiotics for a while for my skin which to me now just seems so so crazy to me that I

Leo Grady

Because as a child.

Melanie Avalon

Mm-hmm. Yeah and on birth control like for my skin. It's just I don't know. I would go back and do things differently.  I Yeah, I had in 2014 I feel like people can often remember exactly when it happened in 2014 I got food poisoning that just like wiped me out I was never really the same after that and then on top of that I did a round of rifaximin I don't know if you're familiar with that antibiotic It's tailored for people who have small intestinal bacterial overgrowth and it's supposed to target the small intestine in particular And it's like a game changer for a lot of people with SIBO other people not so much And for me that like did not it made me even worse. I think so Ever since that whole incident. It's been a little bit of the struggle bus. I it was nice to see that Okay, so for my results for the diversity So I actually I'm not in the normal range. I was two point six six normal range starts at two point eight It was almost in the normal range I actually I know you and I were talking about that and you know, we definitely want that square to be better I thought it would I was wondering if it was gonna be even worse because I eat I eat a really limited diet like I really eat but within you know a handful of foods Do you see with I know you can't give a specifics or anything, but you're testing, you know, so many people What do people scores tend to be and the diversity range and do you think there's a factor of? That there are a lot of like biohacker type people Taking your test like does that influence the results the population that you're testing

Leo Grady

So the normal range is based on, I believe it's about 20,000 healthy people on a global population. So it's not from Jonah users.  At some point, we may switch over as we get enough of those, but it's really using all publicly available microbiome data from the American Gut Project, the Human Microbiome Project, all these different things.

Melanie Avalon

I'm just looking at that, like the broad categories. And I would be curious in general, when people do this, if like which broad categories tend to line up the most. Like I'm curious, like I said, because skin is the only one that seems to like actually line up perfectly. So I'm wondering in general, like does skin health really typically line up for people?  Or is it super, you know, does it really just vary by person? Which categories line up the most?

Leo Grady

Yeah, I would say it varies very much by person. You know, there's some people who are kind of green across the board, like everything looks good. Some people will be generally good and then they'll have like something in the brain health signature that's really bad or in GI or metabolic. Sometimes you see people that are kind of bad across the board. But oftentimes, it's just like one or two areas where things are bad. And then some people are like really good across the board too. So it varies very much from individual to individual.

Melanie Avalon

How often do you recommend people redo it.

Leo Grady

It really depends. If you are taking Jonah because you're trying to work on something, you're trying to identify some issue in your microbiome and then make changes to it, then I would say every few months to make sure that the changes you're making are working and to make adjustments to it.  That would be true for people that are trying to work on health issues or for people who are athletes or biohackers. If you're in a general wellness category and you okay, probably once a year is okay. I think of it sort of on the same frequency as blood work, honestly. People who have chronic conditions or even athletes will do blood work multiple times a year, but if you're just generally healthy and you go for your annual checkup, you might just do it once a year.

Melanie Avalon

Okay, how often do you do it? Bye-bye, Goodbye.

Leo Grady

you know, every like six months or so, but I'm generally healthy like mine, mine's all green. So like, I'm actually like pretty okay.  And I don't have any health issues or symptoms. So I do it to keep an eye on it. But that's just me.

Melanie Avalon

I'm assuming it's like anonymous and such. Do you interpret the Jonah users data as well for things like this, like checking in to see how it's matching what people report running analysis on that?

Leo Grady

Oh, sure. I mean, we have intakes, right? And so people will report any diagnoses they have, any supplements they're taking, you know, they'll report about their diet and so on.  And so, you know, we're constantly like looking at those reports and seeing, you know, their microbiome associations and matching them up against those. And then we're constantly talking to people too. So a lot of people will get the results and want to have a call. And, you know, oftentimes we'll just get on a call with people and talk to the results or we'll talk to their provider, because a lot of people buy Jonah through like a longevity center or concierge medicine or functional medicine. So we'll end up talking to the provider as well. And so we get a lot of information about people's experience with the results and the product. And, you know, it's really trying to make all of it much easier to consume and use.

Melanie Avalon

Awesome. I had another question. Oh, do you have people on like carnivore diets and stuff? So basically like people following a really limited diet Done jonah, there's been a couple

Leo Grady

There have been a couple who have, and generally, I'm trying to think what their results were. Generally, the microbiomes were pretty okay, honestly, at least the couple that I can remember.  We get a lot of folks, and I don't monitor everybody, but there are certain individuals that we work with, more kind of influencers or whatnot, that follow some restricted diet, so we'll pay attention to those and go through their results together. But yeah, generally speaking, I would say I haven't seen any systematic problems with anyone on a carnivore diet or another restricted diet, honestly, and I think about it.

Melanie Avalon

I'm looking at the recommended food list. Red wine, yay, red meat. The thing is, I think everything I eat is probably on the rec... Maybe not everything. It's definitely a bigger list than what I actually eat though, that's for sure.  Yeah, it's really really fascinating. I would be really curious listeners to test theirs and see the results and share. You are so kind. We actually have a discount for listeners. Listeners can go to melanieavilon.com slash jona. That's J-O-N-A and you can use the coupon code Melanie50. So M-E-L-A-N-I-E five zero and that will get you $50 off. So thank you. Thank you so much for that. Oh yeah, and by the way, we talked about this, but why is it called Jonah?

Leo Grady

from the story in the Bible of Jonah and the whale. So Jonah goes in the belly of the whale and emerges enlightened and wiser. I thought this, you know, not personally religious, but I thought this was a really amazing metaphor for what we're doing. Because, you know, on one level, it's going inside the belly and, you know, emerging with wisdom and knowledge. But it's also a story of transformation.  It's a story of, you know, Jonah kind of facing up to his, his fate facing up to the things he was running away from, and like coming out transformed. And this concept of the hero getting swallowed by a great beast really appears in many different faith traditions, right? Like Hiawatha gets swallowed by a sturgeon, and, you know, the Irish hero Kukulun gets swallowed by a beast and Gilgamesh. And, you know, it's always the story of rebirth of transformation. And I thought that, you know, what we're really trying to do with Jonah is allow people to take control of their own health, people who want to really learn what's going on and be able to take action. You know, we want to help them achieve that transformation or achieve that rebirth that they're looking for, you know, in some level. And so I just thought it was really a great analogy for what we're doing as a company. But I took the age off to make it look more like a tech company, more modern.

Melanie Avalon

I love that. No, I love hearing the stories behind the names, and yeah, that perfectly matches up.  What are you most excited about with the future of Jonah, and also what's the biggest challenge that you're, like, facing?

Leo Grady

Well, I'm really excited about multiple different things. So first of all, it's just like the incredible, you know, acceleration of science and the literature here. So, you know, it's more than 2000 studies a month now, but growing, growing, growing. So that's, that's one thing I think that's just really exciting.  Another is the ability of AI to be able to do more here, right, to be able to allow somebody to do meal planning based on the AI, to interact with the AI to talk to the AI to, you know, have the AI who knows their own biology, be able to give them answer their questions and provide more personalized advice, you know, to really help integrate into the execution of their daily life and making these changes and support them. I also think that, you know, there's a hunger among, we talked, touched about this in the beginning, within conventional medicine for the microbiome too. And I've yet to meet a conventional medicine doctor who is not excited about the microbiome, but they don't have any idea how to get into it, how to use it, they don't know what to do with it. And I really think there's an opportunity, you know, starting with the biohackers and the longevity and functional medicine, but being able to actually move into changing conventional medicine too, to really be able to leverage this technology and this new organ of the microbiome. There's such an excitement about it, there's so much power to it. And then ultimately, as we understand it even further, looking further down the line, you know, we've seen through these stool transplants that you can actually, in some cases, cure some very serious diseases with microbiome transplants. So I think as we're able to understand the microbiome better, understand what's wrong with it, how to fix it, I think there's an opportunity to really, you know, someday even cure a whole range of diseases that we can't do that.

Melanie Avalon

It is so thrilling, and I just really can't thank you enough because you're at the forefront of, you know, creating all of this incredible work that is really going to push us further in that direction. Yeah, it's, it's so amazing. So thank you listeners again, cannot recommend enough.  Go to Melanie Avalon.com slash Jonah. Use the coupon code Melanie 50 to get $50 off. Was there anything else you want to touch on with, with all of this and share with the listeners?

Leo Grady

No, except to thank you, and really amazing how much research you put into all the questions and all the thought that you put into this. It's really impressive. So thank you.

Melanie Avalon

Well, thank you so much. Related to that, the last question that I ask every guest on the show, and it's just because I realize more and more each day the importance of mindset.  So what is something that you're grateful for?

Leo Grady

I'm grateful to be working in an area that's so interesting and exciting, that has the potential to really be transformative for a huge range of people. I mentioned the folks in my own life, like my ex-wife, and there are other members of my family who've had issues, and the ability to do something that can better their lives and a whole range of other folks is really exciting, and I feel very privileged to be able to do that.  Furthermore, I'm working with just an incredible set of professionals on Jonah, all of these amazing AI scientists and engineers, medical professionals, microbiome scientists. It's really incredible to be able to work on something that is so interesting and also so impactful, so I'm very grateful for that.

Melanie Avalon

I love it. Well, thank you so much. I am so grateful again, like I said to you as well. You're literally changing the world and this is just so, so incredible. So I can't wait to see the future of Jonah and all the other things that you will do. And yeah, thank you for your time.  This has been amazing.

Leo Grady

Thank you so much Melanie.

Melanie Avalon

I'll talk to you next time.

Leo Grady

Okay.

Melanie Avalon

Thank you. Bye.  Bye. Thank you so much for listening to the Melanie Avalon Biohacking Podcast. For more information, you can check out my book, What Win Wine? Lose weight and feel great with paleo-style meals, intermittent fasting, and wine, as well as my blog, melanieavalon.com. Feel free to contact me at podcast at melanieavalon.com. And always remember, you got this.