Longevity as a Service: AI & Aging Clocks | Alex Zhavoronkov, Deep Longevity & Insilico Medicine

1:34PM Jun 3, 2021

Speakers:

Allison Duettmann

Keith

Keywords:

aging

longevity

clinics

clocks

data

features

predict

interventions

younger

years

physicians

companies

alex

chronological age

understand

blood

question

driven

field

started

We're now moving more into the enabling technologies area. And in terms of technologies that we're looking at, to advance health extension, and we are moving more into machine learning. And we're more moving into aging clock areas. And we're super happy that we have Alex here was actually working on both of them, and connecting the two dots. And he's not only from insilico, but also know from D, longevity and is trying to do longevity as a service. We will hear obala from him, and he can do that much better than I can. But I'll share your bio in the chat. Super excited to have you here. Thank you very, very much for taking time. And yeah, the stage is yours, before we move on to a discussion, and I'll be in the chat in case anyone has any questions. All right, welcome, Alex.

Well, thank you, Alison, for such generous introduction, I'm very happy to see everybody lots of familiar faces on some very good friends. And very happy to be part of your series so far, attended a couple of those. And actually, I must say, I must compliment you on the upgrade that's very useful. And you're bringing a community together. So the community that I hopefully will be talking about now, as well. And I'll share my screen because I am more of a visual person. And I hope what you are to, and especially considering you know what side 2am in Beijing right now, that's where I'm dialing in from on, it's, it's probably easier to have a visual aid. Here you can see my week, chat, I use WeChat most of the time, since I'm very focused on Asia, if you see any Chinese characters in a slide side, sorry, I we are on this region, and I'm very focused on China. Today I'm going to be talking with you about longevity as a service or loss. I think I coined the term. What I'm going to speak from the position of deep longevity Officer of deep longevity, the longevity is a spin off out of in silico medicine, we spun it off in 2019. Even though we've been doing work on aging clocks, since I've raised ception. And we sold it to a company called region Pacific is publicly traded on the Hong Kong Stock Exchange 057 5am hand side the disclaimer. So not investment advice, never buy the stock based kind of presentations, especially if it's by me. I a few words about region, Pacific, the mothership of this of the longevity. So the chairman of that company, and the founder is actually Jim Mellon as well. So a very famous person in our in our field. So if you look at his bio, he actually kind of originated he started his career in Hong Kong, and has deep affinity with the region. So it's not a paper company. It's actually the entire floor in the Hanley building right next to HSBC. And as a holding company, they returned quite a bit of money to the shareholders over the years. And now they are focusing on longevity, and also male health. So they have some other products in the pipeline, focusing on male male health primarily in China. So before I start, shameless promotion of our annual meeting, as currently, I think the largest in the world in drug discovery in aging research last year, it was 2.6 1000 people, you can see that it's supported by all the major journals Lancet, a nature. frontier is, Eli is sponsoring it, and frontiers are sponsoring it. And also aging, usually republish the proceeds, this year is going to be our eighth annual, and for the first time, we'll have the longevity medicine workshop to specifically focus on not only the science of aging, but also on clinical applications because you cannot do longevity as a service without the physicians by him. And we are doing a lot of work right now to speak to to actually steer the physicians into the field of longevity, give them the ability to say speak the same language, at least introduce the very basic concepts and provide them with a toolkit that they can use in order to also engage in NF one studies and experimental medicine. So one interesting kind of factoid is that in December last year, myself, Martin Schreiber Knutson, Seema Scalia from yerba biscoff, so half MDS, how PhDs, we developed a course called the introduction to longevity medicine for physicians, and put it on Udemy. And you can see that you know, 2.6 1000, MDS to get. And it's, it's a pretty popular course now. And we decided to also make it free because you know, Udemy does not allow you to,

to put just three quarters that is more than an hour, I think online, so we moved to teachable, and you can actually find this course on longevity is a degree where I'm making a portal now for longevity education, because I think that the bottleneck, and also the central driving force are the physician, I think that the physicians are the core kind of bottleneck and the central driving force. That's why we cannot really ignore them. And we need to ensure that we engage them. And I think that I might not be the first person to say that, but our industry will move at the same speed as the physicians to. So we cannot move faster than the physicians if you are to provide longevity as a service. So you need to engage physicians. And that's what we're trying to do as much as possible. As you can see, it's actually gaining traction, there are not that many longevity physicians in the world, and many of them took this course. So now that we've moved to longevity, the degree, you can actually take a two and a half hour course it's all of them are likely to be free. This one is definitely free, we also moved it into Chinese. So I think that China is going to be a major driver for anything, including longevity. So actually, if you haven't been to China, I highly encourage you to come here because it's is just marvelous. So everything is high tech, and it's you know, outpacing Japan in terms of technological innovations, high speed trains, everything. So in longevity, there are also multiple efforts for people who really want to get more four out of the out of their life. And they understand that money is not just where it ends. So they're spending money on this, and that there are multiple, pretty substantial initiatives in longevity medicine as well. By the way, we also got CME accreditation for this course. So if you are a medical doctor, and if you take this course, you get 2.75 ama credits, and you get a proper certificate. So I think that's probably the first longevity medicine for physicians course, that is CME accredited took us about half a year to do. But now you actually come get the medical certification to when you do that. So now I'll just switch gears and talk a bit a bit about, you know, I think that you are convinced that aging is bad. Human size, I think are the only species that also understand consciously that they are aging and dying, and that the future is bleaker than today, after you know, 25 for a woman and 34 males. So everything else, everything in the future is downhill. and made sure even evolved us to ignore this problem consciously and focus on more current events, right. That's why FFF takes precedence over anything becomes addictive. So we try to avoid thinking about aging. And we are now starting to think about prevention. And prevention is possible or you can move this curve to the right slightly, but at the end of the day, if it's diet, exercise, sleep, you pretty much get to the same end, maybe a little bit healthier and a little bit longer, but still you're losing. Actually, Chinese aging is very often associated with losing that. So that's why people don't like to talk about it, but we want to talk about it and we can think about you know, possible ways to track repair and improve great to see Aubrey here he was the first one to come out and you know, turn it into a movement. So I think that we should not be shy, very many people think about this as not a credible goal. And also, if you are in pharmaceutical company, you would not show this slide. And I think that now this movement is becoming more credible, but people are still on too afraid of giving themselves false hopes. And they're still trying to plan their lives Life Life suddenly, they're in our reproductive cycles. And their their entertainment, their their their outlook on life is very similar to what it was, you know, centuries ago. I'm still

I'm still on the certificate slide. I don't know if everyone else is too, or really trying to show a different one.

Yeah, I'm sure I'm sure I'm showing a very different one. And that means that

now I got the one with a mountain

set. So we know that we have a Great Firewall here. Yes. So and I hope that you can see already track repair and prove right?

No, I'm still on the mountain. Maybe I'm the only one. But

you know, I think track repair improve. Okay, good. Okay, so that's not me. So thank you, Aubrey. great help. So telling Aubrey story in a different words. And basically, next slide. Is that how can we provide longevity as a service today? Right, because there is actually, the toolkit is very limited. And from my perspective, if we are moving at the same speed as the physicians to, it's gonna take us a very long time to be able to, you know, save the baby boomers, for example, because chances are that by the time we get rejuvenative technologies in the hands of medical doctors, it's going to be a little bit too late for the baby boomers. And that's why we're trying to define the field of longevity medicine, longevity medicine is cutting edge research, science, therapy, and diagnostics are currently it is primarily focused on participatory mount medicine and personalized science, I hope that you can see the slide longevity medicine. And why it's powered by biomarkers of aging and longevity, predictive and prognostic is data driven, individualized, and preventative. And now we have artificial intelligence to help us move in this field. And traditional preventative medicine looks at you in the context of looks at a patient in the context of their age range, and in the context of their well being for that age range. So if you don't have, you know, too many comorbidities when you are 60, the doctor would still diagnose you as healthy. However, if you compare yourself in the context of your entire lifespan, of course, you are much worse off than when you are when you compare yourself to when you're 30. So

I think we may need to hop onto the next slide. Once more. It may be the firewall. I think I'm not the only one starting to see the sides. I'm still on can we provide longevity as a service?

Oh, okay. You know what, let me pause for a second there another

preventative medicine?

Yeah, so it's basically it's freezing. So let me just kill my other screen. And if the worst comes to worst, I'll need to go to my 5g right now. So let me know if it happens. Again. And I'll have to go into my 5g. I'm apologize.

No, totally fine. You are also welcome to send me the slides by email and tell me when to advance.

Yeah, so let me try to. I hope I'm sharing my screen right now. Can you see it?

Not yet. But I'm monitoring the chat. Yeah, I see the stream.

Now. Let's so you see this traditional preventive medicine slide? Correct? Yep. Great. So that's what I was talking about. So in the context of traditional, traditional preventive medicine, the doctor looks at you currently right at this time, and looks at the age range, maybe 60 to 70 and looks at how do you do in terms of performance for that age range? Ideally, we want to look at the person in the context of the entire lifespan, and figure out how to bring the patient back to the ultimate healthy state which would be you know, 20 to 35, or 20 to 40 at least. So how do we track repair, repair and rejuvenate? So basically, it sounds in action. How do we turn sounds into clinical practice for those of you who are familiar with arborists work and AI can help with that. So now how do we do this in the context of longevity as a service? Well, that would require as I define, kind of Seven Pillars of of clinic clinical engagement. So first you need to educate and involved. So for that we've got this longevity medicine course for both physicians. And actually consumers can take it as well, it's just two and a half hours. The reason why we made that, so is that physicians are very busy. And then two and a half hours, you can just, you know, get all the content, you need to be able to navigate in this space, then you need to do really advanced anamnesis data analysis. So basically, questions and answers are with a patient. Now AI can help with that as well. Comprehensive longitudinal diagnostics. So we need to collect data over time. And all of you who are not collecting the data are losing already, because it's extremely important to have this benchmark ideal you to be able to bring your future self back to as close as to this ideal version of yourself as possible, we need to identify what I call longevity bottlenecks. So the aging clocks that are taking faster, and the aging clocks that are and the areas that are likely to age faster in the context of an ideal patient, and drive the rest of the organism with them. So I see for example, in the general population front, when you are doing agent clock analysis on different tissue, we see that you know, long liver, kidneys, bladder, they take faster than many other that many other tissue types like for example muscle. However, in the context of an individual, they might have different individual individual longevity bottlenecks, then personal risk and benefit analysis of interventions. So some people are more risk averse, some people are less risk averse. So we need to look at longevity products as financial products. Because if you look at longevity interventions, as a venture business, so you need to understand that many of those unproven and maybe less proven interventions, they bear a certain risk, but they can also give you a certain return. So the individual and their advisor or financial advisor, longevity advisor, may provide additional guidelines and guidance on how to plan their longevity interventions based on Rick's risk benefit analysis. So now actually, we're coming up with what we call know your patient strategy, and his question year to understand that risk profile and educational profile. And then of course, we can slow down the risk aging with different interventions. And that needs to be driven by aging clocks that measure every level of human organization. So who are the stakeholders in the longevity economy? So what do we need to do to get to the point where we can provide longevity as a service, and we can get the doctors engaged, doctors won't just engage without, without the clinical protocols without the proper training, the old gi for no hyperco

brothers, and they all follow the rules, they're all certified. So a lot of stakeholders need to engage in order for the doctors to be able to follow. And those are, well first of all academic academics. So what should we expect from academia today, and also in the near future, of course, we see major advances pretty much everywhere, you know, the worms left 10 times longer, we see major advances in in mice, we see new JIRA protectors. So new companies are entering the field, however, and new new academics are entering the field. And in China, for example, the Chinese government also wisely prioritized aging for the next five year plan, and more action is expected to come from those efforts. However, the academic efforts are, you know, I operate in the pharmaceutical industry and in silico. So, for the academic breakthrough to come into the pharmaceutical clinical trials. It's usually you know, five to 10 years, and then clinical trials take another five to 10. So if you look at the field of immuno oncology, the academic career which is pretty pretty pretty rapidly progressed. We see that progress over the course of 15 years from academic breakthroughs, those early PD ones video ones, start picking up right if you look Got the history of those checkpoint inhibitors that are now you know, driving progress in oncology. This is 20 years ago 2000 to 2003. And then they started to be commercialized around 2010 2008 2010. And only now you've got an avalanche of those products on the market that are now saving lives, but for longevity interventions is probably even longer because you still need to go through this. So a lot of academic efforts should be encouraged. And we should understand the basic science currently. But we also need to look at what we can do today. And if we look at what is available in the pharmaceutical industry today, and I think that that would be probably not even to us the stem station, ostentatious to say that I probably am one of the most informed people in terms of the pharmaceutical industry's efforts in aging research, I work with top 25 pharma companies in silico. And out of the top 20, at least six prioritize analytics, for example, in their early stage r&d. However, if you look at the those efforts, and how they come and go in pharma at GSK, for example, or Novartis, where they are not really prioritizing the r&d for the clinical use right as just, they need to spend a certain amount of money on r&d, they do it, they try to be prepared for what's in in the biotech sector, so to be able to acquire, so we shouldn't really expect huge progress from the pharmaceutical industry of the next 10 years. And currently, of course, there are very promising companies like for example, bio h. So Kristin fortney, Justin licensed some assets from Asia and other companies, the HIV inhibitors that might be repurposed and purposed for aging, is very similar story to restaurant bio, we understand that the practical utility of those of those interventions in the context of possible longevity gains is marginal. So we're not looking at, you know, 50 years plus. So we're really need to, of course, engage pharmaceutical industry and look at them seriously. But at the same time, we shouldn't expect too much right, because they are driven by the market, and that currently, nobody's nobody has put it or aging and longevity on the flag and decided to go into phase three clinical trials for that. So from from the clinics, we also should not expect significant progress, because the clinics are driven by the medical doctors, by the medical community, and also by scale. So in order for you to institutionalize a longevity intervention and make it big, you really need to have scale. So you have to have something like McDonald's or something like that.

You know, retail clinics, and there are some clinics like that in the states now, pretty broadly with many digital tools for diagnostics. But for treatment, you still need to go through the classical physician and, and in the hospital. So clinics will follow the clinical protocols, unless you're talking about some illegal clinics in Mexico and some exploratory and offline studies. within some small practices, currently, they lack scale. And insurance companies, of course, I think that so my big bet is on the insurance companies, because currently they need to compete and become more innovative, and many of them do. So they're trying to look for additional interesting products for both customer acquisition, and also for underwriting. And some of them are even combining the the two and allowing for some additional health procedures to be performed on really, let's say high net worth, life insurance customers. However, again, there is no major central effort. So what I'm basically where I'm leading to is that you need to figure out how to turn this into longevity ecosystem faster, and accelerate. And the way to do this, as you know, in my mind, you have to work with physicians and the physicians need to be the driving force, they need to demand those interventions they need to start experimenting. And in terms of what we can do today and tomorrow, again, there is not that much. So we are still dancing around diet, exercise, sleep, and some very basic jury protectors and basically do do what your mother told you and Get more frequent diagnosis. So currently, the the outlook on the near term is actually quite bleak. But what we can do tomorrow, and this is a slide from the longevity medicine court. So you can see that there are drugs out there, some of them are, have been on clinical trials, some of them are in clinical trials. So we should expect some of them to come to the market, some are actually already on the market, like an ID booster is that are likely to give you maybe a small edge, but also increase the risk of risk of cancer. And now we see that there are some other effects on blood biochemistry, because we have a large cohort of people who are taking in the mandem, and are in high doses, and we have their blood. So they're not only effects are positive. There is glucosamine that you can take right now, I think that's probably one of the safer euro protectors. And this is the toolkit that the longevity physician can use today. Some of those treatments will be highly quote unquote, experimental like rapamycin, I cannot confirm or deny that, you know, I take it but I do have some with me, I just four as a kind of lucky charm. And this is what you can play with what's coming up is some promising targets. Some promising interventions, I'll actually just skip through a few slides. Because many of you got lectures specifically on those subjects. Now there are multiple interventions, you can now combine them. And the first step that we can do currently, and that's available that is tangibly proven to bring value are the different measurements tool for tools for aging. And the biomarkers of aging are now becoming available and are more popular. But you know, and we are one of the kind of thought leaders in the field, published multiple papers in this area. And I must tell you that we're still scratching the surface, we do not really understand what do this markers tell you, we do not really understand how to properly combine them, and how to improve them. So they give you a really good picture of, of aging, something that we can turn into a digital twin. So of course we're trying to get there. But I think that we are very, very far from this. And actually, that's why I deeply respect the work of floaters for example, lodging, blood, the chef, and scientists in the same league court really trying to dig down and dig deep into the very basic biology of aging and even going back into you know in utero and looking at how we age, because currently we we don't understand

what those biomarkers are telling us, and what we can now predict your age. So we can predict your age where the good would end up with error. And we can now interpret those aging clocks into specific individual features that now we can even tweak whether it's going to benefit you or not. Most of the time, we do not know. But some of those biomarkers are modifiable. And I think this is the first frontier to a longevity ecosystem. And of course, you've got lectures on every one of those blocks already. The first one was so Horvath and hanham published approximately the same time. epigenetic aging clocks, epigenetic aging clocks are the most accurate, but the more but the most difficult to interpret, or turn into therapeutic intervention. At this point in time, unless you're doing reprogramming, D methylation, or re methylation. And AI can help with the methylation clocks as well. It can help in many ways. So what I specialize in is deep generatively and reinforcement learning. So generative adversarial networks allow you to create synthetic data, synthetic profiles of people here, those images are computer generated in 2017. So those are early days of GaNS. Nowadays, those people are completely indistinguishable from reality, they can basically describe what you want to see, I can say that I want to see a white male with blue eyes, and with sunglasses and with bald, and you'll get the distribution of the nose. Or you can give a network a template, let's say give it Brad Pitt and say okay, well show me Brad Pitt who is at Asian and the female, and the many features all birth bits will be retained and you'll get a distribution of those of those images where the duration conditions are confirmed. We use that technique a lot to synthesize biological data. And we also use it for biomarker discovery Target ID. So you can tweak some features in gene expression data, for example, when you're doing when you're giving a network of a template. And then you ask this network to generate a distribution of people from the current age all the way to, let's say, 120. Or you can play around with it and try to go 150 and see how those features change in time. So pretty much like aging a picture, but instead of a picture, you're aging, another biological data type. So that's how you can also identify targets and promising interventions. This same technique can be used in many other areas of human biology research and chemistry research. So we use it a lot for Molecular generation. That's what then silicone, silicone medicines primary expertise is. And that's where we managed to establish collaborations with pretty much every big pharma, which is innovative, many of them are deploying our software, or deployed our software, we can also now use it for mental health, depression, motivation, behavioral modification, research, and of course, for aging biomarker discovery. So deep longevity is. Now in all of those areas, you can see that even from synthetic data, you can now derive the most valuable biological data, which is age. And the way we do it, we use feed forward neural networks, convolutional neural networks and other machine learning techniques, where we feed a specific data type into a model, annotated by age with as many features as you want to predict just one feature age. And with

the increasing level of sophistication of those models, and also, with the increase in granularity of data, you can now try to go and understand basic biology, or even basic psychology, through those networks were experimented with a lot, also with some non traditional data types like psychological survey data, and you can try to go and understand the causality of, of the different biological processes by predicting age in different age groups. And looking at how important the features are for predicting age, for example, within a specific age group, and deriving the possible drivers. So trying to establish the cause and effect diagrams, that again, side synthetic data generation currently is one of the most valuable achievements of deep generative AI. Now, you can also train on age, and then you can add additional neuron for health status, for example, for a disease. And we've done we've done that recently at in silico. So about a couple of years ago, we identified a novel anti fibrotic subtle set of anti fibrotic targets, and discovered molecules for those targets. But the main idea for those target discovery efforts came from this diagram. So we basically trained on age and then retrained on different types of fibrosis identified valuable targets, by looking at the different features that might be causal might be might be driving the fibrotic process. And then we put them put them into pathways and established biological relevance had the disease hypothesis and identify targets. So you can do this for many other processes. And also, it provides you with a pretty good business model, because if you are targeting an age related disease, and at the same time aging itself, that provides you with a clinical pathway where where you can do a net present value on the asset and get a pretty substantial valuation for your company and fundraise for additional research. So and possibly that drug might work in aging as well. And the there are multiple biomarkers of aging that are using deep learning. So I think that the first one that was published in a peer reviewed journal was published by my group by a brilliant young scientist of gagne poojan, no relationship to the other poojan. And we got the first blood based aging clock with mean absolute error five to six years with reasonable r square. And now it's a workhorse clock for us since 2016. So a lot of people are using it and now a lot of people are replicating the work and turning from commercial practice. We also published a transcriptomic aging clockwork but published as a patent patented block which was granted microbiotic aging clocks saw different imaging clocks, facial imaging clocks, physical activity clocks, psychological aging clocks, and here is just kind of efforts. In publishing in this area, I will collaborate with some of the giants in the field, including lighting, bladder chef or M, including recruit them and others. And so far, we've developed a variety of aging clocks that can cater to this longevity as a service community, from different angles to different stakeholders. So the real workhorse clock is the deep hematological aging clock. So this is basic blood tests. And now we also regenerative approach, we can take very small number of markers, let's say 2025, and then reconstruct the rest using GaNS if we're lacking. But of course, if we get up to 70 parameters, where we do have abundant training sets, the prediction is becoming more valuable. And that and we can interpret it with with more utility, because now we can also look at specific features that we can change in order for you to look younger, to the deep neural network, even on this data type. So most of our experiments, and also most of the published work is on deep hematological, aging, clockwork colon blood age. We also of course, the most valuable block that I know of is the deep transcriptomic and proteomic aging clock. So we've built a bunch of patented, there, you can derive valuable targets from the, from the clocks. And also, they are much more interpretable than any other data type. But they're also much more variable on microbiotic aging clocks, and psychological aging clocks are also currently being deployed by our company.

Very simple idea. And the concept of a deep aging clock is I already kind of depicted it before you take a very large number of features, let's say you know, 50 markers, 40 markers, a very large number of profiles annotated with the with with age. And you train deep neural networks with the features or the for example, blood markers on the input layer. So it could be your album and glucose alkaline phosphatase urea bond going on the input layer, then you have many interconnected layers of neurons and one neuron on the output, predicting your age. You can either use one deep neural network, or you can stack them into an ensemble. And now of course, you know, this was 2015 when we did this work in 2016, republished now of course, those are much more advanced, but the basic concept is preserved. So you've got different features on the input, and just one feature on the output. And you get pretty reasonable prediction accuracy. So in our first clock, we've got 5.5 meetups with air. And at that time, we started noticing that people who are praying to be older, they also have all kinds of comorbidities, and they are less healthy. And people who are paid to be younger than their chronological age, they are usually healthier, and they have less gore abilities. And we've demonstrated another paper where we compare different populations, Canadians careers, Eastern Europeans will show that on every population if you are predicted to be five years or older than your chronological age. And by the way, we tested on an independent data set that was not using in training. annotated also with mortality data. So if you are predicted to be five years or older in any chronological age, your hazard ratio increases substantially. If you are predict to be younger than your clinical age, five years younger, your hazard ratio decreases substantially. So this is extremely important for underwriting for insurance companies, and also important for the longevity clinics, when they explain to customers why they need those blocks. So once you're once you can tell a story that those blocks are valuable because you can decrease the risk of dying of any cause. People have become more engaged actually in both communities. And this is the kind of Central basis for connecting the clinics and insurance companies together with the with the class clients because in this case, the interests of all parties are aligned. And this is how you can really provide the longevity as a service in the future. So that's what we are trying to to build this interlink between the insurance company, the clinic and the customer. lsvt skip a few slides. We have published them on multiple aging clocks using different data types. But the philosophy is the same. You take a bunch of data train a neural network to predict The chronological age and the reasonably healthy state. And then try to deconvolute those networks in the most important features and try to interpret. So currently, we developed two tools, a deep longevity that kind of try are trying to bring the ecosystem together. So one is a young AI app, which implements multiple aging clocks. And you can access it for free, right now@young.ai, I can upload your blood tests, you can upload the PDF, we can take a picture of the PDF, and it will take your blood biomarkers predict your biological age, and then tell you what features are driving you driving the prediction. And then you can work with a physician to modulate those features in order for you to become younger. And we also work with clinics. Currently, ATM clinics have deployed those clocks, including some of the really elite clinics like human longevity, in San Diego and several others like huka, for example, in London, and life club in Hong Kong, and many others are bolder longevity. So many of those clinics are experimenting with those blocks in one way or another. But now we're getting quite a bit of data before and after intervention. So for the clinics, we provide those kinds of reports, we call the geometric reports, where the different data types are being sent to the network, we do not know who is the patient, we are only using anonymized data. So we do not keep the personal data in any way.

The data hits the network, and the network just provides the automated report the predicted age, and then provides you with the different recommendations on usually we work with kind of the workhorse algorithm is optimized for 39 markers. That's blood by blood blood age optimum, it has a slightly higher price tag, those are very cheap tests are because we're already using somebody is data that was collected elsewhere. And blood age minimum 31 markers, as you can see no rocket science, all very standard markers like potassium chloride, sodium, white, white blood cell counts, red blood counts, glucose bilirubin, Bond ferritin cranium, so very, very simple markers. But they are actually telling a really interesting story. And you can narrow them down back to the individual organs, and how it works. In this part of the longevity service, you can take a simple blood test request report, as a physician, you can review the report, help make some changes, and then follow up again, and do this again. And currently we're doing it with several clinics, we hope to be able to publish soon. Because now we have quite a bit of evidence that some interventions might be moving the needle in one way or another. And those age metric reports are also quite interpretable. So some are less interpretable. Some are more interpretable. For example, psychological agent clocks are very interpretable hematological aging clocks are also interpretable by the physician and transcriptomic aging clocks, you can interpret using an algorithm but also already using the experience you got by interpreting the aging clocks. Where the features are very well known and understood. And people are trained to tell the story based on those features. So this is how those features look like. So for example, we, we tell you, okay, your white blood count is here, total cholesterol is here, maybe it's within range. But if you want to be five years younger, you need to, let's say reduce the cholesterol to this level. And the recommended value is let's say 174. And then you will shave a few years from your predicted age and get to let's say 36. And currently you are older. So currently, this is the structure of the report. Human longevity is also one of the very active users so actually highly recommend going to July and getting their comprehensive test. So you will know that your accounts are free, but at the same time they will give you the aging clocks. And I'll skip a few slides. That's the just reporting structure how it looks like so you can also see how much does individual feature like albumin, glucose, triglycerides cholesterol contribute to you looking younger or older. And what happens if we modify that feature for you, too. specific value in the reference range. So you can see that, you know, albumin provides, so in this case, is making you younger, and glucose, you're making you older, so 3.9 years, then your chronological age. So if you were to modify those features, you can come closer to your chronological age, or if you want to become younger, the system will tell you how to modify those features to get younger. And, again, I'll skip a few slides. Yes, and you also can use all kinds of techniques like self organizing maps to show where you need to be to get to your optimal healthy state, and also to your, you know, 20 to 40. So what do we need to modify over time, currently, we're doing a lot of research in this area, and we will be trying to deploy some of those tools also visual tools for people to try. And one other important aspect that we are trying to get into right now, trying to understand is the psychological age. So currently, if you think about all the features on anything you can do in the context of aging,

the effects will be very marginal and there is actually not that much you can do so you can modify sleep, you can eat modified diet, you cannot modify your exercise routine, but that is what your mother told you. Right? But can we are still trying to dig in you know, what is the optimal diet diet? What is the optimal exercise routine? In my mind? Of course, it's so it's important research, but not as important as fundamental research in biology of aging. So we thought about can we get into psychology using AI? So can we try to predict your psychological age? That's the chronological age and absence absence of mental health challenges? And can we predict your subjective age that is how old do you feel physically and how old do you feel psychologically to try to maybe play around with deep neural networks in a way so that we can as a human as humans interpret those features and see if you know they make sense so we came up with a hypothesis that biological age and psychological ation subjective age are connected. We looked for the multiple databases that are available but longitudinal data with surveys also that links the surveys to some biological data, and started working with my that's, that's the mid life study in the US data, connecting it to all kinds of other data sets like enhanced and UK Biobank. And young came up with the two features that they consider to be very important one a psychological age, which is a chronological age and a healthy state, and subjective age. It's basically how old individuals feel. And now we're even exploring ways to create feedback loops from basically explain to people how they feel, so they can record the feeling, and at the same time, correlate it with different digital biomarker, data and measurements. So they can actually feel a train to feel to feel themselves better, because we are actually the ultimate sensor. So how old do you feel is a very important question. So we are asking this question, and at the same time looking for a very large number of features from survey data, and behavioral data that are modifiable. So we are excluding non modifiable features, like for example, parental age, kids age as age of death of family members as you cannot change that. So let's not include it in the predictor. And we're looking for modifiable factors that are you know, health status, education, physical activity, longevity expectations as my favorite work by medical knowledge and other favorite social relationships, psychological support, personal beliefs, and other favorite and then we're predicting age, psychological subjective age, using neural networks, and trying to see how those predictions are correlated with difference of different factors of well being, such as mental health, physical health, resilience, distress, productivity, happiness, longevity, etc. And in our first experiment, what we did we took Midas dataset is anonymized data. So big thanks to the US government for for doing that, because it's really advanced as not only their social hero, our research to understand you know, the consumer behavior, it also gives us the opportunity to do this kind of experiments and what we do what we did, we took Originally, about 1000 features that were modifiable narrowed down to featured 50 features, and looked at the most important feature is adjusted the feature list so that we, those are interpretable modifiable, and at the same time not directly correlated to the chronological age, I am trained the predictors of psycho age with about 6.7 years mean up. So there are now we can go down to about six subjective age is about seven minutes of air, which is pretty good because I cannot predict for example, the opposite sex with the same accuracy usually predict them younger significantly.

And the deep neural network does does much, much better. Also, we demonstrated that both psychological age and subjective age prediction predictions are very correlated with the with mortality. So we cannot predict mortality just in the same way as we do with blood age. So if you are predicted to be five years or older, you hazard ratio goes up, both for psychological ages subjective age, and some data subjective age is actually a very good measure of, you know, when you're going to die. So, how old do you feel it's very important, and we deconvoluted those networks into features back into features and kind of grouped them into the different risk factor categories. And then looked at the categories like personality, psychological beliefs, personal well being demographics, health, and showed which ones contribute to the hazard ratio more and which ones contributed to hazard ratio last. And the most important outcome there was that longevity expectations and optimistic outlook on life, is the most important aspect of both psychological age and subjective age. And the easiest way to modify those is to stretch your longevity expectations. And the best way too subject to stretch those subjects, those majority expectations is to strategically deceive yourself that you are going to live longer. And the best way to justify it to yourself, because we're actually quite logical beings as well, is to learn about biomedical progress and actually attend sessions like this. And then positively reinforce this belief over and over and over. And basically, that's why for example, Audrey's teachings and Peter Diamandis, his books are so important, those are technically digital therapeutics for you to be in better mental health. So I never consulted the psychiatrist. And I actually feel great, I like feeling great. And now I basically came up with a routine for myself to strategically convince myself that I'm going to live longer. And that actually gives you great peace of mind and energy. So now, if you are to take something away from this lecture, just make sure that you also try to convince yourself that you're going to live longer, and be more optimistic about life, and try to go after more challenging projects. So we also start integrating wearables, a lot of people are doing that. And we are looking at heartrate and oxygenation of blood and many other readouts that you can take from, from your regular wearable connecting that together. And this is the tool that we envision to drive the longevity as a service. So currently, we provide aging clocks and some very basic recommendations for longevity. And we can provide you with tangible tangible measures and tangible recommendations that you can relate to, to shave some predicted age off your clock. And now we're working with insurance companies. So we've got some pilots, pilots cooking, lots of deployments with clinics already with diagnostic centuries. Currently also, trying to penetrate pharma, both for enrollment and monitoring of clinical trials. We have a few of the Jura productora is being monitored using this clocks, beauty clinics, wellness centers, beauty products, suppliers, and even longevity education. So now we try to deliver this education through the app.

And I think it will be great for employers as well, especially for those that are a little bit more innovative. So I can envision, you know, Amazon or Tesla using something like that to make their staff more optimistic and that healthier. So I wonder why people are not doing that. So the major stakeholders, as I mentioned, are clinics insurance company Here's an employer's one of the major factories and but the really amazing clinics is human longevity. Now they're run by Dr. Wei buco a really amazing Chinese American PhD from Baylor opposed Harvard, mgh Mmm, became a very successful entrepreneur. So now he runs a majority, and provides currently pre wealthy clients with a very comprehensive for health assessment. With follow up, they have a partnership with Harvard mgh to access the doctor network. And now they also offer the aging clocks, they were actually the original investor and the planned devotee one was spun off. So both HP ventures and human longevity venture fund, and actually a few other really, very high profile industries invested. And now we're trying to build the longevity network, where it starts with data collection. goes on the iPhone, ah, prediction, you've got reports going both to the customer and to the doctor. And then a pass is being passed to the clinic and also to the insurance companies. So I think that it's extremely important and possibly will be available to academics. So this is currently the vision, we're trying to build this vision into reality by partnering with clinics and insurance companies, companies separately, but in the future, we see this as a connected life meeting systems. And that's why it's extremely important to standardize those blocks and ensure that people are using just one type, or at least they know how to interpret it. And there is criteria for those blocks. And there is also a hospital path. So we of course want to, to to explore all kinds of all kinds of experimental interventions, and turn at longevity products into essentially venture capital products. Where are you? Where you create an actuarial model? And look at how much life do again? And how much risk Do you accumulate in the context of your personal longevity. So some things may have may have side effects, but those side effects might might not be as important as the life you gain. So again, we introduced a longevity education program where you can actually learn a little bit more about those concepts. And we're offering it to physicians for free. And it's also CME accredited, don't forget to register for this conference. This car is on site and Copenhagen. However, it's all online. And all the who is who is going to be there, all the startups are there. And I'll ensure that pretty substantial number of pharmaceutical executives are going to be there as well. So thank you very much. And we'd love to answer any questions whatsoever connect on WeChat.

Wow, that was a lot. Thank you. So so so much, Alex. This was fantastic. Yeah, I mean, it's, it's, it's, there's very little, I think that you don't have your hands involved in. So thank you so much. First of all, before we start, even into the q&a, how much time do you have? Because we're now at the hour.

I still have for one hour before my next call for 40 minutes before my next call. Okay,

well, let's see how quickly we get through it. And I have one, I have a bunch of questions on my end, but we have a few here in the chat. I do want to reiterate, this was really, really fantastic. Thank you. And you see that in the chat as well. And and I think first one here we have Josh, and Josh, if you want to unmute yourself, go for it. And for the rest, just connect your questions in the chat. or raise your hand here and I'll take you on.

Yeah, I was curious. I don't know if you've heard of Ford health. They're like a clinic in the bay where they collect. Okay, I was wanting you guys to partner with them. Or if you can send samples from them. Not yet. If you can help us partner with them, that would be great. Because I think that this is actually as close to longevity as a service as as it can get. And that was the clinic that I was referring to, as you know, Donald's for for diagnostics. Technically, not the not the most ideal solution. And of course, they're driven probably using by by other by other motivational factories, but we do collaborate with clinics like that, that are preparing to franchise. And there were also looking at interventions, because of course, you know, diagnostics is one thing, but then there needs to be a very clear path to an intervention. And that's the only way to make your, your application sticky. Because if there is no imagine Uber without drivers, right, so You can see where they are, and they can see where you are. But you cannot get from Point A to Point A to Point B. So you really need to ensure there are drivers. And without physicians, it's impossible to do that. And yeah, please connect me to forward. If If you notice, guys, and also very important, one of the reasons why I'm doing this talk is that we are hiring. So we are hiring a senior level, including people in business development. So if you are intertwined in the insurance industry, or if you are good in selling to clinics, we are hiring currently.

All right, thank you. Next one. We have Eduardo and Alex, one, one thing, can I make sure that in case there's an open job position, you send it to me, and I'll include it in the follow up email after this call?

For sure. So we have, it's more about people finding us we'll find a position for them, if they're very good. So again, we're currently looking for the Chief Business Officer, and business development officers. But that's primarily. So it could be very scientific, right? If you're selling to the pharmaceutical industry, and there you need to, to provide a lot of evidence that the clocks work in a very specific way. And they can measure the outcome measures that the pharma companies interested in AI. So that's one type of people that were looking for, of course, if they're more focused on selling to clinics, that's that that's another kind of angle. And for that very often, you don't even need to have a medical degree. And the insurance companies, that's the most important part. So if you have experience selling into insurance, and specifically working at the kind of innovation, sourcing level of that organization, what would we like to talk to you,

Natalie, and we'll always finish the talk with, you know, what can this group do to you. But before that, we have a bunch of questions. Next up, we have Eduardo Eduardo, if you can say maybe one sentence about yourself, just that Alex has some context and who you are.

Cool. So I'm a graduate student at Caltech in the bio engineering program. And recently, in the past couple months, I started learning about the field of aging, before that my expertise is all like single cell, RNA seek and bioinformatics. So all this stuff is relatively new for me. And one of the things that, you know, caught my attention, when I started reading about the field was the epigenetic clock clocks, obviously. And as I started reading about them, I was like, holy crap, the just doing linear regression works really well. Like this, this seems a little too good to be true. And then I saw a bunch of papers on not only genetic, but also transcriptomic clocks, that for longevity in C. elegans worked really well. And we're also just using linear regression. And I was not aware of all the modalities that you that you show today, I'm definitely going to look up your papers. But my question is, you talked a lot about deep learning. In the issue of having interpretable models. I think that the biggest surprise for me in learning about this field of clogs of aging, was how well just plain old linear regression

works. And my question is, in your experience, like, what is the breakdown, between like situations where like, linear regression does the trick and situations where it's like, linear regression is not? It's not up to the job, but then you were able to make it work with a deep learning model. So, you know, I actually I will, probably, because I'm not sure if other people are as good in data science, as we are. Here, especially coming from Caltech, so I just referred to UCLA to a paper you can probably see my screen right. So machine learning and human muscle transcriptomic data. We published it in 2018. But we did the work in 2017. And here, in this paper, you can see a comparison of the different type algorithms and specifically for for ranking. So we can basically look at SVM revenue or plastic net DFS, right. And then you will also look at how they pick the most important features and some of them have different accuracies. Some of them, they have different robustness. But what's important is that how they, which features they prioritize, right. And actually, by combining multiple machine learning techniques, and also linear regression, some just very basic stats, you might be able to sometimes, again, get some biological relevance, right? Because you understand how they rank how they work, right? What what what is usually important for a specific algorithm, you understand the amount of data you've got, so number of samples, and, and the number of features in each sample. And then, you know, once you get the experience of predicting different tissues like this, you see that it's actually sometimes valuable to combine some of those blocks, right, and those methods and also compare them. And I'm in here, we actually did just very simple board to rank or to count to do the final rank, and we identify some draggable, possibly druggable targets for sarcopenia. But that was a young kind of a very exploratory work that we're not exactly proud of, but kind of explains that answers your question. Yeah, definitely. Thank you so much. I'm gonna look at that paper. That's awesome.

Next up, we have Lee.

Hey, Alex, hopefully you can hear me. Yeah, yes, it can be heard. Okay. Hey, Alex. Good talk. Thank you very much. So, you know, I think it's a bit naive to reset. Some blood markers show younger age like cholesterol, because, for example, middle aged women onwards are more protected by higher cholesterol levels. So I think you'll agree with that by yourself. I a medical doctor as well or not? No, no, no, it's the driver. You know, I've been measuring hundreds of biomarkers for many years. Oh, yeah. Yeah. Okay. Yeah, the geeky guy in Slovenia, right, the Scotsman. So Alex, the question I wanted to ask is the what you quickly said,

you were not a believer in epigenetics. Yeah. I'll say believer in epigenetic clocks. But you passed over it quickly. Could you? Could you clarify that, please. So again, I actually have a way doesn't say that I'm not a believer in aging clocks. I said that that's the most accurate biological clock you've got today is just not very interpretable. And it's not super actionable. From my standpoint, and I might be wrong. And by the way, I don't think you're wrong. I just like a little clarification. What do you mean by it's not what you're saying? It's it's not responsive to treatment and lifestyle change? It doesn't respond quick enough. Yeah. So that's one thing. And another thing is that there is always a trade off between accuracy and sensitivity to interventions, right. So some clocks, so are extremely variable. So one, one person can be, you know, younger and older, the same day. But that's exactly what you want. Right? You want to understand what makes you younger, or older on the same day as well. And what what what what what is it that makes you older and younger? On over a long term over a long, long time, right? So I actually first answer your first question like cholesterol modification actually agree with you. So it might be a defensive mechanism in the in a later life, right? We just, we don't know, where I just showing what needs to be changed in order for the deep neural network to predict your younger, right, there's just there are not that many things that it's kind of like think of it as wrinkles, right? So as wrinkles on the face. So we actually do have a photographic aging clock, and I'm the advisor and the early investor in a company called how.ai. You should get them on the program. They're amazing. they cater to all the big consumer companies, pretty much everyone that deals with cosmetics. And they they're run by really wonderful scientists. female scientist, Mr. Car yesterday actually published a paper with me called photo age, where we took a very large number of Germans from 17 to 70. The data set was consented and provided by one of the largest consumer companies. And we've demonstrated that you can build an age predictor with which is 2.2 years accurate. So very accurate, right. And then you can use on different techniques, for example, covering different parts of the face to see where are the major features that make you older or younger. And that algorithm structure that that study showed that those features were around the corners of the eyes. So if you fix the wrinkles, they are, you are predicted to be younger. Right? Of course, that doesn't necessarily mean those are going to be healthier or you know, younger, etc. But you are going to be predicted to be younger, by the network. So we follow the same kind of philosophy with with blood and other markers. Again, I'm not saying that it's wrong, it's that it's right, it's actually possibly wrong and likely wrong, because when you are talking about blood biomarkers, many of those systems are interconnected. And this cholesterol, again, might be cardioprotective. Currently, there are multiple schools of thoughts. But we can at least show some of those, this is just an experiment, right? So we at least can show that you would be predicted to younger if you change some features, and it comes from an algorithm. So we work on a biological system. So that's that, that's, that's a great thing to see. I'm saying that it's not, I cannot say that it actually will make you live longer yet. So for that, we need to have longitudinal data, and nobody has done that before. But the thing is that if we don't try if we don't test, and if we don't study this area, in humans, we won't go that far. It would actually be nice to do this experiment on animals, but it's just not something that I work with.

Thank you. Next up.

Hey, Alex, great talk, very comprehensive and really exciting stuff. Quick question here. So you had mentioned that you're partnering, obviously, with, you know, some companies in pharma and maybe cosmetics industries? Are you currently partnering with any, like other types of industries or companies there? Or do you have any plans for other ways, you know, some of these technologies have potentially broader applications that could serve as a way to bring more people into being aware of this kind of work. So I'm just curious if there's anything like officially on the on the roadmap or still kind of thinking about it.

So of course, constantly thinking about it, you know, I come from a semiconductor industry originally, right. So when I was still in my early 20s, I made some money in that industry. And that enabled me to go into biotech, right. And, of course, I'm very, very interested in partnering with pretty much anything tech, right? The ideal partner for us would be like, my dream job would be with Amazon, right? Because these guys don't provide longevity as a service in the very true nature and the true meaning of this word, right, because currently, they help you waste time. So all the consumer goods you're buying, that you don't need, those, you know, extra pair of shoes that you will never wear, you spend a lot of time on, on their platform searching and playing around with. And they also provide you now with content, like amazon prime video, that basically wastes a lot of your time, and sometimes in your optimal state of health and well being. So it's only natural for them, you know, to follow the esgr, some, you know, environmental, societal, governmental, whatever, guidelines to give this time back to people. And they accumulate a lot of data about you much more than we can ever collect ourselves. And people trust Amazon as the most amazing platform. So, you know, if I could partner with AWS, or Amazon itself, that would be a dream, right? So of course, we're looking at those big cloud providers, but currently they lack this vision and my job partially through those talks, and hopefully I can maybe ignite somebody else to you know, bite those guys, so they get this longevity virus. So somebody who, you know, at the very top thinks that okay, well, you know, aging is a universal equalizer, 1 billion, 2 billion, 3 billion, 5 billion, doesn't make a difference, you know, don't don't fly business class to arrive earlier. So it's just to make your life a little bit more comfortable, but it's not necessarily better. So I think that the most impactful thing for this guy's to do would be to get into longevity and really explore longevity as a service, I think they tried to do something like that with Berkshire Hathaway and try to do this insurance slash health care partnership, but then it's likely, you know, probably it ended up with some really high profile management consultants that turned it into just a very simple plain business model, it maximizes profit. So here, you just need to maximize research and maximize human life. So for me, you know, partnering with an organization that has a lot of data that helps you waste time, I, in order for us through this partnership to make time, that would be a dream. So of course, I'm looking at that, and, you know, any possibility to work at the higher level of those organizations would be a dream.

Thank you. And I guess this is a good tag along to a follow up question that I had just, I mean, you mentioned a variety of different specific services that one can already access, right? Even like young AI and so forth. What are like next steps for those individual efforts? And is there anything that people in this group could do to add to help those along?

For sure, well, I can come work for the longevity if they're qualified, right, because of course, we definitely were, we're currently not super well funded. So we do have the funding. However, however, it's not super well funded, yet. It's a public company. However, the stock is not trading at, you know, anywhere near the levels where it would be possible to fundraise much more effectively. Even though again, to me, I think that's usually promising company an area. So we are interested in people who can really turn it into a bigger business, specifically in those verticals, the insurance industry and, and the clinics. And, of course, we have the consumer facing app, we are working on it, to be able to connect the user to those two industries, both insurance and how and clinics. But that will take time, right? Because those industries also need to grow up in order to be able to partner at this level. So we are a little bit ahead of our time. And we were we always were ahead of our time of work in silico. If you remember 2016, I was presenting this generative chemistry, nobody believed me. And now we're getting those mega orders from Big Pharma and also progressing the drugs through the, through the, through the through the various stages of clinical 3d printing, preclinical development. So I think that we will, we will get there sooner, sooner or later, too. So we'll get this longevity of the service going and connect the insurance and clinics, I hope that it will happen sooner than later, the way to help us is to come work for us or help us connect with those players more efficiently. Because I don't think that there are so there are a few groups that are also very dedicated to longevity in a very credit with a very credible business model. And actually deeply admire those groups. But there are very few right on where one of them. So we need to stick together and try to promote those businesses. so that they become more profitable, more, more More, more, more, more active. And we need to embed those businesses deeper into the major stakeholders that are currently driving the world. Right? Because changing just one stakeholder insurance will not change without medicine, medicine will not will not change without insurance, pharmaceutical companies will not change without medicine without insurance, academia is not going to accelerate without. So it's not a rat race, like in semiconductors yet right. So without a huge consumer business. So in order for us to drive for the longevity as a service, vision, we need to ensure that all of those areas are are progressing very quickly. And we cannot progress faster than medicine. So that's why we need to ensure that medical doctors are engaged involved and educated. So you know, if you want to contribute, you know, take the longevity medicine course and give it to a friend. It's free. You get a CME credit certificate you can put on a wall.

Yeah, I love how you went ran through every different different sector and you're like, this sector is here and it could do this. And so yeah, well, thanks for that bird's eye overview, I guess in particular on either end or machine learning and aging clocks. Do you have any predictions where, you know, there are interesting fields for folks that are excited about either using machine learning for advancing progress on aging, or really, of trying to do something? You know, either by really standardizing a bunch of aging clocks, or something in that regard, like, where would you see, in particular, those fields headed? Are there any particular future challenges that are interesting to take on, because many people here in this group have an interest in this and either already working in this or are currently putting their mind stood?

So for sure, well, first of all, I think that the best way to operate in this field is to join one of the companies or academic groups that is already working in this area. I think that there are many more data types that we could work with in order for us to predict age, and also try to understand it better. So I think that if you are just in machine learning currently, and if you do not understand biology very well, in all go into imaging, specifically, and go deeper into imaging, start working with the MRI, CTS and, and other data types on ultrasound, I think that there is huge opportunity for advancement and ultrasound, because that's where you can build better aging clocks, try to try to get this very accessible data type into the clinic. And again, I'm a huge fall fan of ultrasound, I really if I had more time and resources, that's where I would go, because that's one of the ways to to get into non invasive diagnostics so quickly, and actually gives you a pretty good resolution of your, you know, tissues and organs. And then tried to combine ultrasound with something that is more exotic. So like gene expression, or mineralization levels, or cross links. And I think that combinations of different, very accessible and very excited biomarkers is the is the way to go, because that's the way to make the exotic biomarkers cheaper, because some of the data types are very expensive, right? So and for some of them, you really need require to have very advanced equipment. And if you can substitute some of them are basically you have the same answer to your question to a medical question from another data type that is extremely valuable. So just my advice is that ultrasound write down another really cool, man, I just asked a question about ultrasound, have you seen that it works well with other species than just humans? So, actually, this is one of the very few data types where there is more human data than animal data. So the to be honest with you, I don't know. Another another really cool advice I'm, I've been thinking about this a lot is if you if you are starting your career currently in this area, or if you are thinking about switching it or going into business, I would recommend going into monkey business. So specifically looking at monkeys. So because I'm talking to a lot of contract research organizations currently, especially in China, you know, that experiment on our monkeys? And it turns out that on the supply monkeys, turns out that most of the monkeys that are being supplied to Farmer are exactly like three years old, because if you are below three years, they are considered to be in mature if they are above three years, they are too expensive, right? Because you need to feed the monkey for one more year. So technically all of the experiments by him like most of the experiments are done in this kind of age range, right? So so nobody is trying to mature them and and and look at them legitimately am about races or marmoset doesn't matter something that is closer to humans than than mice, right. I am something that is intelligent and can jump on trees, an exercise so I think that having a really cool collection of data coming from monkey colonies, that would help us much more because then we would be able to actually test those. Well answer those questions that were also raised today. Well, what about cholesterol? What What if we, you know, turn it back? If you want to reduce it, is it good or bad right is it gonna be fact your longevity or not? And those are the animals where we can actually go and, you know, sample the tissue a little bit more. The easier way, right.

Interesting, I think really mirrored also what we had bronica kransky on here from the NA, n Exactly. It's exactly what he said, exactly. Okay, walk me what species can we find so that we're moving away from mice? And we're getting a little bit more closer to humans?

Yeah, because my students are just not level long enough to accumulate the same forms of damage that we have. And it's just sad.

Yeah. Okay, one last question. If I'm now a potential customer, what which website re go on. So you know, that's the longevity degree, but there's also young AI, right. So you know, because at the beginning, when you mentioned, okay, if you're not already tracking your health data, you're already losing out, right? Like, let's say, I'm convinced, what what are the easiest next steps to do for not just someone that may want to work for you guys, but also someone you know, for whom it's just really excited. And once you these use these products.

So currently, in terms of customers, we only want to come from a customer perspective, whether you b2b as the major segment, right? So it's so if you are running an insurance company, or if you are running a clinic, please do come to us, we'll partner with you. If you're a consumer, currently, you can download the free app and start using it. And also report to us anything that you like, or you don't like, because currently, we're trying to, you know, play mini Bezos, and basically put the customer in a chair, and try to focus on user experience, because we have a weekly user experience call, we try to understand what people like what people don't like, and what would make the product more sticky. Currently, it's a very experimental system. So we can, you know, swing it in multiple directions, but we haven't pushed the trigger currently, in terms of you know, where we want to go. So it's just as an app, which allows you to track biological age, and allows you to, you know, tell us as much as you like, so you control how much data you give, you own the data, you can delete it, you can delete it, you can get out of it anytime we want, keep it. And in this context, we want to ensure that, you know, we just get user feedback to make a better product. And that's, that's what we're looking for.

Okay, so to summarize, if you want to work for Apple, if you want to work for the boundary, go contact or go on the website, or read the email that I'll be sending out afterwards. If you work for Amazon, then also please immediately, if you're on the consumer side, you know, you may want to be joining their consumer test. And you know, apart from that, I think you know, you mentioned a really few fantastic ways for either and, you know, really getting longevity education out there with a longer degree, or by just downloading AI and participate is starting to participate. Thank you so much also for noting a few of the areas that machine learning researchers could go into and look into super, super valuable, is there anything I'm missing that could be useful for, for the positivity or in silico?

I think that we are actually doing quite well. Now. You'll hear some really cool news. So we're not in dire need of anything. We're we're short on really highly qualified and competent staff it actually in both both cases. And, yeah, I just think that it's a very good time to be in longevity. So if you have a friend, you know, have a dinner with with a friend, I'm convinced a friend to get into longevity, because everything else they're doing is pretty much you know, waste of time.

Yeah, thanks. No, it definitely sounds like you're on a roll there. And I'm super excited. I mean, you know, I guess if I only follow up in like half a year or something, then the slide deck is even, you know, two hours long or something. So I'm super, super, super grateful that you shared so much with us. And yeah, thank you for your work. Thank you for even taking time, so freaking late in the night to share this work with us. Really, really, really appreciated by everyone in this group. And thank you everyone for staying on so long. And we are now finishing with the main talk, Alex, I'm hoping that I can reach out to you maybe get my hands on your slide to share it out with folks. And we are now doing a few community updates. In particular, we are going to be starting with Jonathan Carlson who has a similar I think announcement to make and if anyone else afterwards has a community update and wants to go for it. Please drop me a note in the chat. And otherwise we'll be seeing you for our keynote meeting and next week Thursday.

Is ageing pharma that orgy please register to for this conference. Again, that's University of Copenhagen is getting the funds. But it's a very important conference. And we are, I'm sponsoring I'm supporting it. And we're building a community there. So you can see that it's for the first time. I know Eli from frontiers are actually sponsoring it. So usually you sponsor the journal here as the other way around, and Lancet nature aging, they are all supporting it. So it's a high profile conference, please do come there. And this year is going to be at the Grand Hall of the University of Copenhagen. So you know, it's the Viking guys the guys who are quite popular on Netflix. Everybody's talking about Vikings, so come to the to the, to the to the home of Vikings, right and Morton Schreiber. Knutson is the you know, most amazing researcher and longevity medicine practitioner. So he is MD PhD. He is the main Chair of the conference. And you know, it's it's a great community. So we'll try to be there on site. So I'll get vaccinated and get to Copenhagen.

Thank you so much. Okay, Copenhagen's calling. Okay, thank you so much, Alex. And next up, we have Jonathan with the community on that announcement. If anyone has another one to share, then please just let me know. Otherwise, there'll be a next time again next Thursday, Jonathan.

Thanks a lot, Alex. I mean, a good shout out for the Vikings. I'm not sure I can claim that myself. But Copenhagen is a great place. It was great meeting your way, way back, right, four years now in Berlin or so. So excellent to see this development. Really, really nice. Three quick things, I think one, really appreciate the holistic view that you're having there and the time wasted, and so on, I think it can very much agree with that. I do have a couple of contacts into high level executives and dammuz on so I will talk to them in the next couple of days. And then hopefully connect you if there's any anyone else that I can connect you to super happy as well. Also hire two great commercial guys, just this week to my own venture, got a couple of really good examples where we've been super happy with, you know, sourcing people. So if anyone else in the call as well could be happy about like some way to source both commercial guys, and particularly really technical guys. That's also been very valuable for me, happy to share that one. And lastly, not what we're trying to give back is, you know, from our organization, we're hiring two people, ourselves to CEO, and also channel strategist. why we're doing that is because surely I cannot compete with you in any sense when it comes to the lab work, excellent work. But we believe that I worked myself a lot within effective altruism, for example, in sustainability. So I'm used to completely failing at communicating something extremely interesting and valuable and important to people who simply don't want to listen. And I think we all had this a little bit of longevity as well. So what we're trying to do is a be testing throughout certain keywords and so on towards the market, how we can much better communicate around longevity. And hopefully, we'll building then a more commercial platform that can be shared within community as well, to kind of, you know, put this onto a broader, broader base out there in society. So if that's of any use to yourself, Alex or anyone else out there, then we're happy to share. So we'll come back.

Well, thank you, everyone. Thank you so much. Thanks for staying on too long. Alex, I hope you get to see Oh, one last question. And do you already You said you connected to fit, but do you connect it to the roving already? Do you track that we're in? No, don't, okay.

But it's a it's a good thing to do, right. And that's becoming much more popular. So we try to get the wearable data, and of course, the sleep data as well. But currently, we just don't have the capacity to, to do a lot of research on this data. So we hope that we'll be able to do that in the future. But currently, we are looking at other data types that are, you know, consumer friendly. And I was I must say that I was extremely surprised how people are interested in a psychology of aging. So when we launched our test, more than 300,000 people took it in, basically three weeks. So without any advertising without anything, so it was a major driver. So people are very interested in how to do this. And I think that this is a very engaging area. So I think psychology of aging is underappreciated. And I myself I thought that only you know, psycho psychiatrists are not going to say that word but not exactly very credible, and necessary, and a waste of time. But now you can actually, you know, looking at longevity psychology, I think that this area needs to evolve. Get into the hands of those in our 1000s of psychiatrists in the States, because I think that we can actually fix people by making them more optimistic by giving them actually longevity education.

Well, and that's what you're doing. And then again, you know, I think it's a it's a positive feedback loop. So I guess the, the more likely it is that the more younger we look, and so forth, and so on and so forth. It is a more positive feedback group to get into. So thanks for doing all of them. And yeah, I'm super excited for what's next. And I am hoping to see many of you next Thursday, where we have a very, very full house, and we're going to be talking about an agent clocks again, and this time with buddy Morgan and a few others. So super, super excited for that. Thank you so much, Alex. It's been really a pleasure. Bye, everyone.