Latest / Lab to Lives / How Much Does Your Life Cost w/ Stuart Berkowitz

Transcript

How Much Does Your Life Cost w/ Stuart Berkowitz · Lab to Lives

• 0:09
  SPEAKER_04: Welcome to Lab2 Lives, the podcast where we ask how do we get great ideas out of the lab and into people's lives.
• 0:20
  SPEAKER_04: Your hosts are Alexander Buddh, Chiara Kianiz, and Ivana Rosenberg Ucida.
• 0:28
  SPEAKER_04: We are improv comedians who also work in life science.
• 0:33
  SPEAKER_04: On this show, we invite guests from the industry to give us their bold take on how we could get more healthy with less waste.
• 0:44
  SPEAKER_04: So whether you're developing a next generation drug, innovating in MedTech, or just curious about how healthcare moves from discovery to impact, welcome to Lab2Lives.
• 0:59
  SPEAKER_04: Welcome everybody to another episode of Lab2 Lives.
• 1:03
  SPEAKER_04: Today on the host side with me I have Alex Booth and I have Kiara Kinnis.
• 1:09
  SPEAKER_04: Hi! And our guest today is Stuart Berkowitz.
• 1:13
  SPEAKER_04: Welcome, Stuart.
• 1:14
  SPEAKER_00: Thank you.
• 1:15
  SPEAKER_00: It's a pleasure to be here.
• 1:16
  SPEAKER_04: Alright, well, nice to hear your hello.
• 1:19
  SPEAKER_04: We ask our guests on the show to introduce themselves by placing them on the five value streams in pharma.
• 1:26
  SPEAKER_04: And on the show, we usually categorize pharma into five different buckets or streams or pillars.
• 1:33
  SPEAKER_04: One is the innovation stream, where the magic happens of discovering new treatments.
• 1:37
  SPEAKER_04: There's the manufacturing stream where we figure out how do we make many of those and actually distribute them.
• 1:43
  SPEAKER_04: Then there's the compliance stream, meaning how do we ensure that all of the above is safe and actually works for the intended purpose.
• 1:49
  SPEAKER_04: Then we have the commercial stream, which is all about getting the drug to the actual patients in the different countries where we want to promote it.
• 1:57
  SPEAKER_04: And the corporate stream that binds everything together and makes sure that everyone also gets their salaries.
• 2:02
  SPEAKER_04: Where would you place yourself among these five value streams?
• 2:06
  SPEAKER_00: I think my closest tier is innovation.
• 2:08
  SPEAKER_00: I worked in the in a lot of different areas, primarily in research math, and entrepreneurialism, but I also was a consultant for many years for the Ontario Genomics Institute.
• 2:19
  SPEAKER_00: And we gave half a million dollars or less research grants to researchers in genomics and genomic medicine.
• 2:26
  SPEAKER_00: And I was also a judge on the International Genetically Engineered Machines competition that universities compete with, where they try to create fungus and bacteria, which are edited by programmers, to perform certain functions, like to create drugs or to create sensors in the environment.
• 2:43
  SPEAKER_00: And so I have some familiarity with drug creation through the use of these organisms and also other applications of these organisms, like finding gold, which is kind of weird.
• 2:53
  SPEAKER_00: Just an interesting story is that we actually had the Delph Tika bacteria named of the University of Delph, which likes to eat gold, and we introduced you to drop the gold into bucket so we can mine gold out of the mines.
• 3:07
  SPEAKER_00: So I have some experience.
• 3:09
  SPEAKER_04: That is fascinating.
• 3:11
  SPEAKER_04: You can have a bacteria that eats gold.
• 3:14
  SPEAKER_04: How I have questions.
• 3:15
  SPEAKER_04: How does it even digest it?
• 3:17
  SPEAKER_00: Well, it it I I think we think that that things are either plants or animals, but there are certain organisms that suck electrons off of metals, and these are the delft bacteria.
• 3:28
  SPEAKER_00: D L F T I C A bacteria is this bacteria, and then it can be modified.
• 3:34
  SPEAKER_00: So we shine a blue light on it, it will drop its goal into our bucket.
• 3:38
  SPEAKER_04: That's amazing, and also just the picture of it.
• 3:40
  SPEAKER_04: Wow.
• 3:41
  SPEAKER_04: Stuart, I am also curious, genomic medicine.
• 3:43
  SPEAKER_04: To our listeners who may not know what that is, could you explain a couple words?
• 3:47
  SPEAKER_00: Yes.
• 3:47
  SPEAKER_00: Okay, so all diseases can probably be categorized into three categories: genomic, somatic, which means it happens after a fertilized egg, and then a third is environmental or a combination of all three.
• 4:00
  SPEAKER_00: And there are many mechanisms in the translation of the genome into proteins or non-coding RNA, and these can have alterations that can be neutral, deleterious, detrimental, or beneficial with various probabilities.
• 4:17
  SPEAKER_00: And so these can be a source of disease.
• 4:20
  SPEAKER_00: They're not necessarily an obligatory disease, but they they add to the probability that a disease will happen.
• 4:25
  SPEAKER_00: Contingens, for instance, which is a over 30 repeat.
• 4:29
  SPEAKER_00: I think of the HTG, I may have that wrong gene.
• 4:32
  SPEAKER_00: That's a genetic disease.
• 4:33
  SPEAKER_00: And if you only have one copy of it, you get it.
• 4:35
  SPEAKER_00: But that's a rare situation.
• 4:37
  SPEAKER_00: Most diseases, you only increase the probability of getting a disease.
• 4:40
  SPEAKER_00: You have a compulsory disease.
• 4:42
  SPEAKER_04: Yeah.
• 4:42
  SPEAKER_04: And Stuart, I am curious.
• 4:44
  SPEAKER_04: Research math.
• 4:45
  SPEAKER_04: What kind of environment does that happen in?
• 4:47
  SPEAKER_04: Does it happen in a lab?
• 4:48
  SPEAKER_04: Does it happen in an office?
• 4:50
  SPEAKER_00: Well, I was a PhD student in theoretical computer science, which is math.
• 4:54
  SPEAKER_00: And I invented the world's fastest method for inverting a matrix.
• 4:57
  SPEAKER_00: Matrices are used all over the place in engineering and in finance, even in Google, even in large language models.
• 5:04
  SPEAKER_00: If you if you look up a result by me and uh Paul Samuelson who won the Nobel Prize, it's in Wikipedia.
• 5:10
  SPEAKER_00: It's called The Fastest Method for Inverting a Matrix.
• 5:12
  SPEAKER_00: I also wrote some other academic papers in monotone circuit complexity.
• 5:17
  SPEAKER_00: And I had a lot of fans in Germany who were very interested in this because they had assumed the opposite case was true.
• 5:23
  SPEAKER_00: So they were very surprised.
• 5:24
  SPEAKER_00: And so I did a lot of research in math, and that helped me achieve a little bit of fame, and it helped me when I started a company after my PhD.
• 5:32
  SPEAKER_04: Very cool.
• 5:33
  SPEAKER_04: Well, Alex and Kiara, you have both had a similar journey through academia.
• 5:37
  SPEAKER_04: I am curious what did your environments look like when you were doing research?
• 5:41
  SPEAKER_04: Was it in a lab?
• 5:41
  SPEAKER_04: Was it in an office?
• 5:42
  SPEAKER_03: I was in a lab.
• 5:43
  SPEAKER_03: I was in perhaps one of the saddest labs, which was a converted storage cupboard.
• 5:50
  SPEAKER_03: No.
• 5:50
  SPEAKER_03: And I was also in a research group just to myself.
• 5:53
  SPEAKER_03: So it a lot of my PhD was sitting alone in this converted storage cupboard.
• 5:58
  SPEAKER_03: No German fans, unfortunately, either.
• 6:00
  SPEAKER_03: I didn't get the David Hasselhoff effect.
• 6:03
  SPEAKER_03: So that it was still fun.
• 6:04
  SPEAKER_03: I was in there tinkering with subsystems and things.
• 6:07
  SPEAKER_03: It was a good time.
• 6:08
  SPEAKER_03: Where was it?
• 6:08
  SPEAKER_03: Where was this?
• 6:10
  SPEAKER_03: This was in this was at the University of Leeds.
• 6:12
  SPEAKER_03: Big shout out, University of Leeds.
• 6:14
  SPEAKER_03: Absolutely great book for both studying and research.
• 6:18
  SPEAKER_03: Yeah.
• 6:19
  SPEAKER_02: Well, my environment was actually a wet lab.
• 6:21
  SPEAKER_02: Hey Stuart, I did my PhD mix of mailing for Italy.
• 6:24
  SPEAKER_02: So I spent a lot of hours, like a lot of hours among tubes.
• 6:29
  SPEAKER_02: I did my PhD in the stage where we went from doing everything manual to actually being able to have the next generation sequencing.
• 6:36
  SPEAKER_02: So I was also part of the in our lab.
• 6:39
  SPEAKER_02: You know, how do we go from doing everything manual to fishing sequence phishing sequencing?
• 6:46
  SPEAKER_02: So from phishing sequencing to next generation things, the whole genome sequencing.
• 6:50
  SPEAKER_02: I was also part of the cancer genomics lab where we had some of the first robots that they prepared.
• 6:57
  SPEAKER_02: So the robots prepared the plaques.
• 6:59
  SPEAKER_02: So we needed to train not train, like set this robot up and see how that would work.
• 7:06
  SPEAKER_02: I was also confined in one of those uh sterile rooms where there's only you and a thousand tubes.
• 7:12
  SPEAKER_02: I would often also so the m the most exciting time of my week was when I had to go and get samples from like a cryo-freezing room.
• 7:20
  SPEAKER_02: So that was the highlight of the week because we had a change of scenery.
• 7:25
  SPEAKER_02: It felt a bit dangerous.
• 7:27
  SPEAKER_00: So when I was working in in the genomics medicine area, we had this theory that biology was transitioning from wet end to dry end, and that many more analytics-based techniques were seen to be useful for the promotion of medicine.
• 7:42
  SPEAKER_00: And so we were obviously interested in epigenetics, genetics, genome splicing, and all sorts of other things that could be exploited to understand how the body works and understand how disease can be prevented or treated.
• 7:55
  SPEAKER_00: I think biology is becoming a computer industry and a computer application as opposed to a test tube operation.
• 8:02
  SPEAKER_02: Yeah, already when I I was working in this cancer genomics lab, it was almost 10 years ago.
• 8:07
  SPEAKER_02: Oh my God.
• 8:08
  SPEAKER_02: But it was a lot of bioinformatics.
• 8:10
  SPEAKER_02: And then the big challenge was when we changed from single gene analysis to single gene sequencing.
• 8:16
  SPEAKER_02: We went to this whole exome, all hexome sequencing, all genome sequencing, and the output that we got from the machines was so high, so heavy.
• 8:25
  SPEAKER_02: So there was a lot of work into understanding and setting up the bioinformatic analysis of it, which was not so direct and so immediate because you have to set up all the software, so the statistical.
• 8:36
  SPEAKER_02: So I definitely agree with you that there's a lot more on that end because we have developed the humanity has developed these tools that can provide such a huge output from a little sample of DNA.
• 8:47
  SPEAKER_02: And then we have access to the whole genome.
• 8:50
  SPEAKER_02: So of course, there has to be a lot of computer bioinformatics tools.
• 8:54
  SPEAKER_02: And now with AI, I can imagine that could be really, really a huge opportunity to understand all these, all this output that can come out of these analyses.
• 9:02
  SPEAKER_02: So yeah, I totally agree with you.
• 9:04
  SPEAKER_03: Yeah, it's an interesting transition.
• 9:05
  SPEAKER_03: This and I was thinking a little when Stuart was talking earlier.
• 9:08
  SPEAKER_03: I wonder whether all of this is just some of the progression of our advancement in how we address challenges these days, and particularly our increasing ability to use biology not as something that just exists and we can influence but not control, but something that we can actually use and adapt as tools.
• 9:26
  SPEAKER_03: And that used to be simple.
• 9:28
  SPEAKER_03: We used to take wolves from the wild and train them to be dogs through a personal training and experience.
• 9:34
  SPEAKER_03: But as we move into the detail of leveraging biology to hunt gold or produce specific compounds or replace mRNA in our cells to treat genomic diseases, then that whole space is so complex that, yeah, we're moving into this highly computational end as well to give us the power to do that.
• 9:53
  SPEAKER_03: But is that your experience, Stuart?
• 9:54
  SPEAKER_03: Is that how you see things?
• 9:56
  SPEAKER_00: Well, I think that that uh these non-biological applications of hunting gold allow you to perfect your techniques in manipulating and programming bacteria.
• 10:05
  SPEAKER_00: One of the observations we learned is that bacteria don't have enough ATP.
• 10:10
  SPEAKER_00: They don't have mitochondria, and they can't power computation as readily.
• 10:14
  SPEAKER_00: So they tire out.
• 10:15
  SPEAKER_00: And we can overwhelm a bacteria by programming it with too many decisions.
• 10:20
  SPEAKER_00: And so we need a better model for drug development.
• 10:23
  SPEAKER_00: If you think about drug development and a bacteria or fungus, we need mitochondria to power it.
• 10:28
  SPEAKER_00: We also need protein folding and protein creation facilities in order to have it properly being folded, especially for large molecules.
• 10:36
  SPEAKER_00: You're probably familiar with the constructor theory of molecules that molecules that require more than 15 steps are more likely to be biological molecules.
• 10:44
  SPEAKER_00: Well, those have protein folding problems.
• 10:47
  SPEAKER_00: So in a bacteria or fungus vehicle, you need protein folding, and that's absent in these models.
• 10:54
  SPEAKER_00: So we need a better model for creating drugs in protein.
• 10:58
  SPEAKER_00: But I one of the things I wanted to talk about, if I get a chance today, is that I have a new website based on my mathematical research that uses Bayesian analysis, but it has a direct application to drug development.
• 11:08
  SPEAKER_00: Here's how.
• 11:09
  SPEAKER_00: When you're looking at which area of the world you want to focus a drug discovery in, you should look at that area and say, how well is it served by a collection of interventions that if it's a 98% served, 98% disease reduction served by some collection of interventions, then it's probably not worthwhile to develop a new drug there unless there's some side effects that you can avoid.
• 11:32
  SPEAKER_00: Okay.
• 11:32
  SPEAKER_00: But if you have a disease where it's not well served, then you have an innovative opportunity.
• 11:38
  SPEAKER_00: So that's a lead into what we do, if I may go there.
• 11:41
  SPEAKER_04: Stuart, maybe I can ask you based on just the short intro, would you say that that is one of the key pain points in our industry, that drug development is directed into areas that are already overcrowded and not enough drugs are developed into areas where no current treatment exists?
• 11:55
  SPEAKER_00: Yes, I would say that there's a lot of Me Too drugs that don't improve the treatment greatly over the previous generation of drugs.
• 12:04
  SPEAKER_00: And that limits profitability, that limits growth, and it's frustrating for a researcher to spend 10 years of his life developing a drug and then find out that it's not really a huge success.
• 12:14
  SPEAKER_00: And I think that using our website to study disease areas, to look at what the Pareto optimal treatment plan is as the alternative to a new would-be drug is crucial to finding the opportunities to target.
• 12:29
  SPEAKER_04: I completely agree.
• 12:30
  SPEAKER_04: There's a lot of waste in our industry with a lot of focus on areas that are already overcrowded and not enough spreading out into solving actual problems that patients face.
• 12:39
  SPEAKER_04: Stuart, I am curious, how did you come upon this solution?
• 12:43
  SPEAKER_04: Like how did the idea of aggregating all this information in one place, how did that come to you?
• 12:47
  SPEAKER_00: Okay, well, as a math guy, I knew that the Bayesian theorem, which basically gives you an update rule for your credence or your degree of belief in a certain theory.
• 12:58
  SPEAKER_00: So the basic Bayesian formula is basically that the probability of a theory given some evidence is equal to a factor times the previous probability of theory without the evidence.
• 13:07
  SPEAKER_00: And that that factor is the technical part, which can be given as the probability of the evidence given the theory over the probability of the evidence.
• 13:14
  SPEAKER_00: That's the factor.
• 13:15
  SPEAKER_00: And that factor helps you adjust your credence in the new theory, in a theory given new evidence.
• 13:22
  SPEAKER_00: Now, we do this all the time.
• 13:23
  SPEAKER_00: You and I, let's say you have a favorite sports team, like a soccer team, and there's 25 soccer teams going for the championship.
• 13:30
  SPEAKER_00: Your probability is 4%.
• 13:31
  SPEAKER_00: But you may assume that because it's a local team and you like them a lot and you know the players' names and you know their the player's stories, you may assume that that's a 20% chance of winning the championship.
• 13:43
  SPEAKER_00: But as you go along, you notice that the opposing team that is most likely to challenge you, their player got in a car accident and they can't play for eight weeks.
• 13:52
  SPEAKER_00: So you raise your probability of belief from 20% to 30%, but you have no reason for why you do that.
• 13:56
  SPEAKER_00: It comes out of your head as a feeling that you should raise it 30%.
• 13:59
  SPEAKER_00: And then maybe your team wins the first three games.
• 14:03
  SPEAKER_00: So you raise it to 45%, but you don't have any reason why you do that.
• 14:06
  SPEAKER_00: You just assume that's true.
• 14:08
  SPEAKER_00: But those factors are amateur hour factors, which a Bayesian formula would give you the exact different uh the exact way to adjust your probability belief.
• 14:18
  SPEAKER_00: And the the effect of that is that you will get to the wrong answer sometimes, or you'll get there much slower.
• 14:24
  SPEAKER_00: Now, if you're using a computer, that doesn't matter because the computers are very fast.
• 14:28
  SPEAKER_00: But Bayesian analysis is used by humans, but it's also the whole basis for large language models is based on one little problem.
• 14:37
  SPEAKER_00: It's called completing the sentence.
• 14:39
  SPEAKER_00: You have two words, what's the third word likely to be?
• 14:41
  SPEAKER_00: That can be translated into a Bayesian formula.
• 14:44
  SPEAKER_00: And so large, but it's imperfect.
• 14:46
  SPEAKER_00: It's an approximation.
• 14:48
  SPEAKER_04: May I ask, Stuart, is what you're trying to say is that our whole industry is like this math problem that needs to be solved, and the problem of solving it is not having sufficient information to figure out what is actually the probability of which drug to develop next.
• 15:02
  SPEAKER_00: I'm not saying that per se.
• 15:04
  SPEAKER_00: I'm saying that this Bayesian math that I'm using in this website called longevityresearch.ca is a precise use of Bayesian math, but Bayesian math is all over the place.
• 15:14
  SPEAKER_00: Like you're you're using it when you think about sports teams and who do you think is going to win.
• 15:18
  SPEAKER_00: You're using it in large language models, but you don't know it because it's an imperfect form of Bayesian math.
• 15:23
  SPEAKER_00: And what we do with this is that we derive new information that doctors don't know, that research don't know.
• 15:30
  SPEAKER_00: We derive new information from old medical research papers or recently published medical research papers where we create a causal model and we remove cross-correlations.
• 15:39
  SPEAKER_00: There's a few PhDs here, so you'll know what I'm saying.
• 15:42
  SPEAKER_00: This we convert single point probabilities in the form of hazard ratios into joint probabilities.
• 15:48
  SPEAKER_00: And once we have a joint probability, we can then search an exponential space of interventions and find the one that is optimal.
• 15:55
  SPEAKER_00: It's called the predo-optimal.
• 15:56
  SPEAKER_00: So it's optimal in the sense it achieves the maximum benefit, but it reduces the number of interventions to a small number, so you minimize side effects.
• 16:04
  SPEAKER_00: So that's the bogey that a drug researcher has to be better than.
• 16:09
  SPEAKER_00: Okay, so if I create a and we look at not just over drug space, but over the space of exercise and supplements and psychological interventions and neurological training devices.
• 16:22
  SPEAKER_00: And so when I look at all that from the point of view of medical research, I can derive the true causal effect of any combination of those interventions, and therefore I can find the optimal and I can search through the exponential space.
• 16:36
  SPEAKER_00: When I say exponential space, I want to point out that if I have 10 interventions, there's a thousand twenty-four different ways to subsets of ten interventions.
• 16:44
  SPEAKER_00: If I have 20 interventions, it's a million.
• 16:46
  SPEAKER_00: If I have 30, it's a billion.
• 16:47
  SPEAKER_00: Okay, the average number of interventions in our 93 oracles is about 25.
• 16:52
  SPEAKER_00: So it's somewhere between a million and a billion.
• 16:54
  SPEAKER_00: Doctors can't do this, researchers can't do this, only a computer can do this.
• 16:58
  SPEAKER_04: So you're taking information that is available about research that has been done, and you're essentially creating it into a math problem that you can solve, and you can show the results on the website that you have created.
• 17:10
  SPEAKER_00: Right.
• 17:10
  SPEAKER_00: So that's why it's new.
• 17:12
  SPEAKER_00: There's no other website like ours.
• 17:14
  SPEAKER_00: You know, I'm not trying to make any money.
• 17:15
  SPEAKER_00: We don't collect any personal data, and we're just trying to help people save lives.
• 17:21
  SPEAKER_00: And so it's just something I want to do to give back to the world and say it save life years.
• 17:26
  SPEAKER_00: We we save an average of 4.1 life years over the doctor's protocol.
• 17:31
  SPEAKER_00: So the way we do this is we have the NHANES data set from the CDC, which is uh free.
• 17:36
  SPEAKER_00: It has 101,000 anonymized medical records, and then we have the drug treatment plan that the doctor gave them with the disease classification, and then we look at what the doctor did, and we compare that to the Bayesian Pareto Optimal Plan, and we save an average of 4.1 years over all different kinds of diseases that lead to an early death.
• 17:57
  SPEAKER_00: So there's about 11 diseases like brain diseases and pulmonary diseases and heart disease and cancer and liver and kidney, and there's some other ones, and we save an average of 4.1.
• 18:08
  SPEAKER_00: But if you look on our website, there's a more detailed characterization for every disease, every type, every subtype, what the potential savings would have been over the CDC data with where they actually have doctors' recommendations.
• 18:21
  SPEAKER_00: So this the reason why doctors underperform is not because of malice or anything like that.
• 18:26
  SPEAKER_00: It's because the computation is too difficult for a doctor to do in his head.
• 18:30
  SPEAKER_00: It takes too much time, and there's an exponential number of things to evaluate, and it's just way too difficult for a doctor to do.
• 18:36
  SPEAKER_00: So why is this important is that we are basically showing the best in class treatment from existing resources that a drug researcher can look at as the bogey that he has to beat.
• 18:48
  SPEAKER_00: And if he can't beat that, then he should go somewhere else.
• 18:51
  SPEAKER_04: It's fascinating.
• 18:52
  SPEAKER_04: It does address one of the problems that we have in our industry, and that is, as you're saying, the uncertainty of where exactly to focus.
• 18:59
  SPEAKER_04: And just because something is a good idea from a drug developer's standpoint, often what happens is that we fail in the market because it is already too crowded or better solutions actually exist.
• 19:11
  SPEAKER_04: So this is giving that feedback at the point in time where a different decision can actually be made.
• 19:16
  SPEAKER_00: Yes, I agree with that.
• 19:17
  SPEAKER_00: This could help people.
• 19:18
  SPEAKER_00: If they look around my website, they will see something called, I think, the Pareto-optimal treatment plan for every disease.
• 19:26
  SPEAKER_00: I think we have over a thousand diseases, I think.
• 19:29
  SPEAKER_00: We have 45,000 academic references.
• 19:32
  SPEAKER_00: It's all documented in painstaking detail for any researcher to validate and verify it.
• 19:38
  SPEAKER_00: One thing that we could develop in the future, but we haven't done yet, is let's say you had a drug called X and you knew what its characteristics would be in terms of what its mechanism of action would be, what its hazard ratio would speculate it to be.
• 19:52
  SPEAKER_00: You could then look at it in combination with all the other interventions and see whether or not it would be chosen as an optimal intervention.
• 19:59
  SPEAKER_00: Okay, if it doesn't move the needle, then it won't be optimal.
• 20:02
  SPEAKER_02: I just want to give Kara and Alex a opportunity to ask some questions before you proceed.
• 20:06
  SPEAKER_02: Yeah.
• 20:07
  SPEAKER_02: While you were talking, I'd have a thought.
• 20:09
  SPEAKER_02: You mentioned often that this is for the drug developers.
• 20:12
  SPEAKER_02: So my question is you build this very comprehensive website.
• 20:15
  SPEAKER_02: How do you see it being used?
• 20:18
  SPEAKER_02: Imagine I am Kara, I'm a researcher.
• 20:20
  SPEAKER_02: Is that the type of person you are targeting?
• 20:23
  SPEAKER_02: Or are you also targeting doctors in general?
• 20:25
  SPEAKER_02: And if, for example, is researchers, at what stage of the drug development this tool can be used?
• 20:31
  SPEAKER_02: Is that early discovery?
• 20:32
  SPEAKER_02: Is it clinical development?
• 20:34
  SPEAKER_02: So, where do you see this tool being applied and by whom?
• 20:37
  SPEAKER_02: So if someone is listening to this and like, oh, I can actually use this.
• 20:41
  SPEAKER_00: Well, I think the three types of users are drug researchers.
• 20:44
  SPEAKER_00: This is a research tool.
• 20:46
  SPEAKER_00: This doesn't give prescriptions for people to take, but it's a research tool.
• 20:50
  SPEAKER_00: So I think drug researchers, epidemiologists, researchers, uh doctors for sure, and specialists, but also sophisticated patients, like patients with, you know, science background.
• 21:01
  SPEAKER_00: And the the interesting thing is that in the overall population, 19% of people have a chronic illness resulting in an early death.
• 21:07
  SPEAKER_00: But it uh but as you go up in age, like say to 70, I think the number at 70 is 43% of people who are 70 have a chronic illness resulting in an early death.
• 21:17
  SPEAKER_00: So if you're a person that has a skill to read this website, you have lots of friends your age and relatives, your siblings, your cousins who have a chronic illness leading to an early death.
• 21:27
  SPEAKER_00: And if you care about them, this is a free website that you can act as a surrogate proxy, and you could input their information as much as you know, not perfect, but you probably know their disease, you probably know the subtype, you probably know the stage, like in cancer, and you may not know the treatment that the person is taking, but you could figure out what the optimal is, and you could send it to the person who is suffering from an illness, and you could say to them, Look, I've done this, I don't know your exact medicine you're taking, so I can't compare it.
• 21:56
  SPEAKER_00: But here's what is achievable as an optimal result.
• 21:59
  SPEAKER_00: And if that's That's useful, or you want me to input what your treatment is, I can do that, and then I can show you how many life years I can translate the risk reduction into the number of life years gained by looking at the mortality tables and looking at the the risk of death in any year and seeing how that changes and that adds life years to your lifespan.
• 22:17
  SPEAKER_00: So we think it can be used by those groups of people, and each one could study this as a pre useful tool.
• 22:23
  SPEAKER_00: But I think primarily it's gonna be doctors because doctors, this is a big, open, unexplored problem for doctors' expertise, and it becomes too computationally difficult.
• 22:32
  SPEAKER_00: The Bayesian update formulas of backdoor and front door and eigenvector analysis and eigenvalue analysis is just too complicated.
• 22:39
  SPEAKER_00: It's all matrix techniques and complicated techniques, and it's iterative.
• 22:43
  SPEAKER_00: So a doctor's not going to do that in his office.
• 22:46
  SPEAKER_00: Also, as I said, there's an exponential number of bundles of interventions that have to be individually analyzed.
• 22:52
  SPEAKER_00: If you were to do this with an RCT, a randomly controlled trial that costs$50 million, you'd need an exponential number of$50 million trials, and no funding agency, no government, no world government would pay that kind of money to do this as an RCT.
• 23:07
  SPEAKER_00: You can only do this using Bayesian causal analysis.
• 23:10
  SPEAKER_00: By the way, the FDA, Federal Drug Administration, just approved Bayesian causal analysis in January for the approval of new drugs.
• 23:17
  SPEAKER_00: There was a giant article in March discussing this.
• 23:19
  SPEAKER_00: And so you should look into this as a valid alternative to RCT, randomly controlled trials.
• 23:26
  SPEAKER_00: That is interesting.
• 23:27
  SPEAKER_00: I was advising a company, and I don't want to say which company, but they made a drug.
• 23:31
  SPEAKER_00: They're making a drug that hasn't been approved yet.
• 23:33
  SPEAKER_00: They're using an RCT, but I showed them how if they switched to a Bayesian causal method, because the FTA just approved it, they would save about six million dollars right now in the middle of their work because they could convert to an RCT, to an association study.
• 23:48
  SPEAKER_00: But the the best part of it was that they save eight months earlier drug availability on the market, and they'd make eighty million dollars more.
• 23:56
  SPEAKER_00: So they'd make a total of eighty-six million dollars more because they got the drug earlier, and they use this approved process that the FDA has already announced.
• 24:04
  SPEAKER_00: And it's all because of Bayesian causal analysis.
• 24:07
  SPEAKER_02: Yeah.
• 24:07
  SPEAKER_02: It sounds like from what you're describing, can hold a lot of the answers to the problems that we're facing in the research field.
• 24:14
  SPEAKER_02: I as you talk, I found it all fascinating, and I'm all up for the you know democratization of information.
• 24:21
  SPEAKER_02: I just wonder, and I'm maybe I can get here just a very quick answer.
• 24:24
  SPEAKER_02: What do you see as potential risks of making the tool like this available, for example, as you said, to people who can be scientifically literate?
• 24:33
  SPEAKER_02: Is there any risk?
• 24:34
  SPEAKER_02: Maybe a quick question, like one sentence answer.
• 24:37
  SPEAKER_00: Okay, so some of my friends and family have said that I have to dumb it down and to make it really easy to read.
• 24:43
  SPEAKER_00: But I said to them, well, I have to talk the language that doctors talk and serious people talk, which is hazard ratios, positive primitive value, D values, and all these things.
• 24:53
  SPEAKER_00: Otherwise, they're not gonna believe me.
• 24:55
  SPEAKER_00: Otherwise, I'm just telling a story without any scientific evidence.
• 24:58
  SPEAKER_00: And the reason why I think we will be safe and the control on it is that doctors are the gatekeepers of specifying which drugs a person can take.
• 25:06
  SPEAKER_00: And doctors are the responsible parties.
• 25:08
  SPEAKER_00: So nobody could take what we do and administer drugs to themselves without the doctor first specifying that those drugs and giving a prescription for those drugs.
• 25:17
  SPEAKER_00: So that's a safety protocol that society has developed, which I'm riding on.
• 25:23
  SPEAKER_00: So I think that I'm just giving information.
• 25:25
  SPEAKER_00: It may be confusing to neophytes, but hopefully they will send it to their doctor and the doctor will understand and say, yeah, this is something we could we could explore.
• 25:34
  SPEAKER_00: Let's try it out.
• 25:35
  SPEAKER_03: I mean, a lot of this makes me think do we also need to think about some of the other factors here in terms of how we go about processing and analyzing and even choosing the information sources that we're using to make decisions as well.
• 25:47
  SPEAKER_03: And I say that because I like what you were saying earlier about this giving you a way to identify what is the best, most effective drug area to target.
• 25:55
  SPEAKER_03: And what happens is we tend in the industry to go for these incremental improvements versus the big hit improvements.
• 26:01
  SPEAKER_03: But I like to think about organizations as organisms a little bit as well.
• 26:05
  unknown: Okay.
• 26:05
  SPEAKER_03: And I wonder whether I wonder whether part of this is you have this sort of research organism over here, which is very much interested in data and numbers and what's going to be the biggest effect.
• 26:14
  SPEAKER_03: I'm not sure that the industry organism has those inbuilt same incentives, right?
• 26:20
  SPEAKER_03: And I think part of the reason that incremental improvement happens a lot is because we have a tendency to overvalue the familiar but small against the unfamiliar but big.
• 26:32
  SPEAKER_03: And so alongside this, do you think there's also a need for us as doctors or as an industry or as even individual people with a sickness to also think about the types of information we take in and how we're making decisions as well to sort of open up this as a possibility.
• 26:46
  SPEAKER_00: Yeah, you said a lot of things there, but I'll just try to address the the ones that I can remember.
• 26:52
  SPEAKER_00: As I said, I think we have about 45,000 academic references in a list on the website.
• 26:57
  SPEAKER_00: And what we rate those A, B, C, or D using a standard protocol for the measurement of the quality of academic research.
• 27:05
  SPEAKER_00: D is animal studies, which we can, you know, you you animal studies are good because I hate to say this, but you can take a liver out of an animal, you can kill an animal, which is immoral.
• 27:15
  SPEAKER_00: And I'm a vegetarian, so I don't like to kill anything.
• 27:18
  SPEAKER_00: But uh, animals are beneficial, but they're also not necessarily translatable to humans.
• 27:23
  SPEAKER_00: That's D.
• 27:23
  SPEAKER_00: And there's some research which is in the D category, but generally we're using A, B, C.
• 27:28
  SPEAKER_00: What I'm trying to do is I get a result, I do A, B, C, D, and then I say, okay, now rebuild this using evidence quality ABC, and if see there's any difference.
• 27:36
  SPEAKER_00: If there's no difference, then I don't mind using A, B, C, D.
• 27:40
  SPEAKER_00: I think that was your opening comment about what's the quality difference.
• 27:44
  SPEAKER_00: How do I know if you've done a complete and correct evidence?
• 27:47
  SPEAKER_00: I think that was related to your comment.
• 27:48
  SPEAKER_00: Did you want me to address any other aspects of the other?
• 27:50
  SPEAKER_03: No, no, I was more thinking about the weight that we give to different types of information as well.
• 27:56
  SPEAKER_03: So, yeah, and different types of scientific evidence have different weights and importance.
• 28:01
  SPEAKER_03: That's fine.
• 28:02
  SPEAKER_03: I wonder more about stepping up a level and say, are we as an industry and are we as say individual patients making decisions for ourselves and what we need to pursue based on scientific information?
• 28:14
  SPEAKER_03: So, industry, for instance, I think tends towards commercial information, right?
• 28:19
  SPEAKER_03: And I think we've all seen cases where high impact drugs lose out because while the patient case is there, the commercial case is perceived as weaker.
• 28:29
  SPEAKER_03: And so, do we also need to try and have a bit of a revolution in the industry about how we prioritize things, even with tools like yours that are giving us the information?
• 28:38
  SPEAKER_03: Are we able to use that information in a way that that gets the best outcomes even?
• 28:42
  SPEAKER_00: Yes, I think what you're talking about is the compliance problem in the industry.
• 28:45
  SPEAKER_00: And so one interesting statistic that I came across is that when a doctor prescribes medication, only 50% of the people comply essentially, and 30% don't comply so badly that they don't achieve the benefit of the medication.
• 28:59
  SPEAKER_00: So this is the compliance problem.
• 29:01
  SPEAKER_00: You probably know this yourself.
• 29:02
  SPEAKER_00: Why are people overweight?
• 29:03
  SPEAKER_00: Why do people smoke?
• 29:04
  SPEAKER_00: Why do they pe why do they drink alcohol?
• 29:06
  SPEAKER_00: Why do they take drugs?
• 29:07
  SPEAKER_00: Why don't they exercise enough?
• 29:09
  SPEAKER_00: These are all well-known things that help people live long, but for some psychological reasons, people don't want to be bothered.
• 29:16
  SPEAKER_00: And this could be your spouse, could be your children.
• 29:19
  SPEAKER_00: Even though you talk incessantly with them, that your life is so valuable.
• 29:22
  SPEAKER_00: I often pose this question to people.
• 29:24
  SPEAKER_00: I say to them, the most valuable thing that you ever will get or ever got was your own life.
• 29:30
  SPEAKER_00: And on your birthday, people shouldn't be congratulating you.
• 29:33
  SPEAKER_00: You should be thinking about the gratitude you have towards your parents and all your ancestors.
• 29:38
  SPEAKER_00: And I then I also say to them, Lauren Buffett is 95 years old and he's worth$150 billion.
• 29:43
  SPEAKER_00: And let's say you are 45 years old, so there's a 50-year difference.
• 29:47
  SPEAKER_00: Would you trade places with Warren Buffett?
• 29:50
  SPEAKER_00: Would you change your age from 45 to 95 but get 150 billion dollars?
• 29:54
  SPEAKER_00: I've never met a person that said yes to that.
• 29:56
  SPEAKER_00: And what they're really saying is that their life is worth 50 years divided into 150 billion is worth 30 billion uh$3 billion a year.
• 30:06
  SPEAKER_00: Now, if you had a fancy car, you'd wash it, you'd clean the garbage out of it, you'd take it in for servicing, but you have this life, which most people would admit is worth some number of billion dollars a year.
• 30:18
  SPEAKER_00: Why wouldn't you do the maximum in terms of compliance with the doctor's recommendations, with the scientific evidence which says what you should do?
• 30:25
  SPEAKER_00: So, for instance, on my website, not only do I look at diseases, I look at aging as a disease, which is an innovative way to look at things.
• 30:32
  SPEAKER_00: And so, how do I become the longest living person on earth?
• 30:35
  SPEAKER_00: Well, you can look at the Social Security mortality curves and you can see what your individual based on your age, what's your prior, that means what is your age cohort's risk of death, average risk of death per year.
• 30:47
  SPEAKER_00: And it accelerates.
• 30:48
  SPEAKER_00: And there's something called the Gompertz equation, which basically says something like this your risk of death per year doubles every 15 years when you're 15 years old.
• 30:57
  SPEAKER_00: But when you get to be, let's say, 75, it's only seven years.
• 31:00
  SPEAKER_00: So your risk of death accelerates as you age.
• 31:03
  SPEAKER_00: Now, what are all-cause mortality oracle?
• 31:06
  SPEAKER_00: That an oracle is what we had in the Odyssey as godlike answer machines, but that was tongue-in-cheek, name it or an oracle.
• 31:13
  SPEAKER_00: But the the all-cause mortality oracle says the following.
• 31:16
  SPEAKER_00: Okay, if you're in the hundredth percentile of your VO2 max, which is your aerobic fitness, it's the integral of all the exercise you've ever done.
• 31:24
  SPEAKER_00: If you're in a hundred percentile, you can lower your all-cause mortality by 92%.
• 31:28
  SPEAKER_00: If you take a sauna, which the Norwegians and the North Europeans love to do four times a week, you lower your all-cause mortality by 40%, but it sort of interacts with exercise, but I can figure that out.
• 31:40
  SPEAKER_00: And if you do weight training one hour a week or t or a hundred hours a week, you get the same benefit, only 28% reduction in all-cause male.
• 31:48
  SPEAKER_00: And when you combine these things in a Bayesian causal analysis, you can reduce your all-cause mortality by 95%.
• 31:55
  SPEAKER_00: And these are all free, almost free.
• 31:57
  SPEAKER_04: That is fascinating.
• 31:58
  SPEAKER_04: Both that you have figured out what a year of life is worth.
• 32:02
  SPEAKER_04: That's uh mind-blowing, but also how you use math to figure out how you can essentially extend your life and exactly at which point you have achieved that life extension.
• 32:12
  SPEAKER_04: That's a fascinating way.
• 32:13
  SPEAKER_04: Why I really like that.
• 32:14
  SPEAKER_00: So just I just let me extend this a little bit more.
• 32:17
  SPEAKER_00: So I realized this about 30 years ago.
• 32:20
  SPEAKER_00: I got a treadmill desk at work.
• 32:22
  SPEAKER_00: So I'm always walking.
• 32:23
  SPEAKER_00: I'm on a treadmill desk right now, but it's I'm not walking because I don't want it to be bumping up and down.
• 32:27
  SPEAKER_00: Okay.
• 32:27
  SPEAKER_00: But I also bike, now that I'm retired, I bike four to five hours a day.
• 32:32
  SPEAKER_00: Okay.
• 32:33
  SPEAKER_00: Then I I go in a sauna, I weight train, and I'm maybe doing five to six hours a day of exercise.
• 32:39
  SPEAKER_00: So my VO2 max is 61.2, which puts me on the Olympic bicycling team for women.
• 32:46
  SPEAKER_03: That's a very high VO2 max.
• 32:47
  SPEAKER_00: Okay, they're about 20 years old, but I am maybe 50 years older than that.
• 32:51
  SPEAKER_00: So how did I do that?
• 32:53
  SPEAKER_00: Well, it was exercise.
• 32:54
  SPEAKER_00: It's what your mother and father told you when you were a little kid.
• 32:57
  SPEAKER_00: Go out and exercise.
• 32:58
  SPEAKER_00: Okay, but I'm studying the math side of it, and I'm figuring out how to be extremely long-lived.
• 33:04
  SPEAKER_00: Like the range on my lifespan is put into the Bayesian causal analysis, and I have a lifespan of 111 to 135.
• 33:13
  SPEAKER_00: Okay, because I'm an extremophile on VO2 Max.
• 33:16
  SPEAKER_00: I'm religiously doing sauna for four to seven times a week.
• 33:21
  SPEAKER_00: I'm always doing weight training.
• 33:22
  SPEAKER_00: And I have about what what's interesting, here's another interesting little side to that, because I I there's a bunch of supplements.
• 33:28
  SPEAKER_00: I take 11 supplements, but they only improve my my all-cause mortality reduction by 2%.
• 33:33
  SPEAKER_00: So, in other words, these three things reduce it by 95%.
• 33:36
  SPEAKER_00: So, supplements, a lot of people want to promote supplements and say supplements are wonderful.
• 33:41
  SPEAKER_00: Uh, I have a list of every disease on the website, what its pareto-optimal intervention set is, and what percentage of that benefit is achieved by supplements.
• 33:50
  SPEAKER_00: Almost everything has no benefit or very little benefit with supplements, but there are certain disease conditions where supplements help.
• 33:58
  SPEAKER_00: For example, osteoarthritis or pre-eclampsia or all-cause mortality have a 10-15% benefit from supplements, but most of it has nothing.
• 34:08
  SPEAKER_00: So this can disabuse a lot of biohackers from getting thinking that supplements are going to save themselves, but but it's in some cases it does, but in 99% of the cases it doesn't.
• 34:20
  SPEAKER_00: Let me give you an example of Steve Jobs.
• 34:22
  SPEAKER_00: He had pancreatic cancer, which was treatable with the Whipple procedure, but instead he chose supplements.
• 34:28
  SPEAKER_00: Now, if you would have looked at my website, he'd be alive today, maybe.
• 34:31
  SPEAKER_00: Okay.
• 34:32
  SPEAKER_03: I love that board claim, Stuart.
• 34:34
  SPEAKER_03: Well, what I love about that is the tangibility.
• 34:36
  SPEAKER_03: And I mean, you're you're treading on my area of interest now as well, because as someone who recently turned 40, I'm having these thoughts, right?
• 34:44
  SPEAKER_03: I want to age well into later life, then what are the best things to do?
• 34:48
  SPEAKER_03: And you know, exercise and weight training, and I think our VO2 maxes are approximately the same, but which is amazing that you kept yours so good.
• 34:55
  SPEAKER_03: And I was struck by your Warren Buffett example, right?
• 34:58
  SPEAKER_03: Which is we often don't think about this, which is well, what is our life worth?
• 35:02
  SPEAKER_03: And what are we willing to give up to extend our life?
• 35:06
  SPEAKER_03: And as you said, you know, people would clearly give up three billion a year.
• 35:09
  SPEAKER_00: Right.
• 35:10
  SPEAKER_00: A famous Greek philosopher, a poet who lived 2,600 years ago, named Archer Locus, said essentially this man doesn't rise to the level of his aspiration, he falls to the level of his habits.
• 35:22
  SPEAKER_00: If you have poor habits, you're gonna have poor results.
• 35:25
  SPEAKER_00: And a habit, once you habitualize something, it's like brushing your teeth.
• 35:28
  SPEAKER_00: You don't even think about it, or like combing your hair or putting on your clothes.
• 35:32
  SPEAKER_00: You know, you just do it and you get on with your day.
• 35:35
  SPEAKER_00: It's just business, okay?
• 35:37
  SPEAKER_00: It's no big deal.
• 35:37
  SPEAKER_00: So me going biking every day for four hours, I don't even think about it anymore.
• 35:41
  SPEAKER_00: It's just part of my habitual plan.
• 35:43
  SPEAKER_00: And that, and I think it's with anything.
• 35:46
  SPEAKER_00: So we have to understand why people, you know, this goes back to the earlier discussion before about compliance.
• 35:51
  SPEAKER_00: Why people don't do the right thing, why why they are not conscientious.
• 35:55
  SPEAKER_00: It turns out that conscientious people live about 11 years longer than non-conscientious people, because conscious people do the right thing.
• 36:03
  SPEAKER_00: What your teachers and your parents told you to do, they do.
• 36:06
  SPEAKER_00: Okay, but people who are unconscientious who always do the wrong things, they give into their non-science-based ideas and they get into trouble.
• 36:14
  SPEAKER_00: And it's sad to see that happen.
• 36:16
  SPEAKER_03: I like this build-in habits analogy as well.
• 36:17
  SPEAKER_03: And I mean, to take it back to what we like to talk about a lot, which is how to get treatments from lab into people's lives, I think on an industry level, this habit point is a great one because I think this is a lot of the incremental drugs come about because we have this habit of, well, we want to say stay safe, we want to fit, you know, stay in this area that we feel comfortable in.
• 36:35
  SPEAKER_03: And we haven't necessarily developed the habit of, well, actually, this is the one that gives the best patient benefit by Bayesian analysis or anything else.
• 36:43
  SPEAKER_03: And so we need to step out into that less secure place and have the trust that, well, if you deliver for the patients, then the money and the profit and everything sort of follows on.
• 36:53
  SPEAKER_03: And so I like that.
• 36:54
  SPEAKER_03: I think there's habits all levels, right?
• 36:56
  SPEAKER_03: The personal, the the sort of organizational, the health system that having this access to data can support.
• 37:02
  SPEAKER_04: And I'll say that this is what we have time for on this episode.
• 37:06
  SPEAKER_04: So unfortunately, we are going to skip the quiz so that I can make my next meeting.
• 37:11
  SPEAKER_04: And instead, I'm going to ask us to do a round of reflections on the content of this episode.
• 37:18
  SPEAKER_04: So I'll just give everyone an opportunity to mention what your favorite thing was based on this conversation and what you might still be thinking about tomorrow in the shower or on your way to work on your bicycle.
• 37:30
  SPEAKER_04: I will share.
• 37:30
  SPEAKER_04: That's for me.
• 37:31
  SPEAKER_04: I was really fascinated by this ability that you can in fact hire out a bacteria.
• 37:36
  SPEAKER_04: I think it's adorable that bacteria need nap time too.
• 37:40
  SPEAKER_04: And also, I really enjoyed, Stuart, your insights into how by applying the proper math, we can find definite answers to many of our health problems.
• 37:50
  SPEAKER_04: That is fascinating and definitely something I will be thinking about both today and tomorrow.
• 37:55
  SPEAKER_03: Yeah, I think for me, I uh I like the idea of this feedback loop.
• 37:59
  SPEAKER_03: And partly because I've been doing it in my my own life.
• 38:01
  SPEAKER_03: I mean, maybe Stuart, we're we're similar in that, in terms of, well, I value my life, I want to live longer, what's the best way of doing that?
• 38:06
  SPEAKER_03: Well, what let's look into the most effective things, and so exercise is is a big one, obviously, and I don't take many supplements, but I do do a lot of exercise.
• 38:14
  SPEAKER_03: Using information as a reinforcement to good habit forming, again, both at the individual level and the organizational level.
• 38:21
  SPEAKER_03: I love that as an idea of a process, actually.
• 38:24
  SPEAKER_02: We takeaway, so I was also like Ivana fascinated by your ability to use as a means to find solutions.
• 38:33
  SPEAKER_02: For a long time in my life when I was studying, I thought that I had to learn math just because it was part of the subject that I had to learn in school.
• 38:42
  SPEAKER_02: But now it seems I mean it's definitely so central.
• 38:45
  SPEAKER_02: And I don't think we give it too much credit.
• 38:48
  SPEAKER_02: And then I was also fascinated by the the way you really showed how you value your life.
• 38:54
  SPEAKER_02: And you kind of are an example for everyone, not just about how to live longer, but how to live better longer.
• 39:02
  SPEAKER_02: So that's also what I takeaway from me.
• 39:05
  SPEAKER_02: It's not just about living longer, it's also making living your life in a better way so that you're healthy for longer.
• 39:12
  SPEAKER_00: Great.
• 39:12
  SPEAKER_00: Thank you very much.
• 39:13
  SPEAKER_00: Yeah, I got a lot of interesting takeaways, and an idea popped into my head while I was speaking with you, which is can I modify my website to allow researchers to put in a pseudo-drug, a prototype drug with prototype characteristics and study its effect in the world of all the other interventions?
• 39:31
  SPEAKER_00: Uh and I'm gonna think about that problem and see if that is something I can solve.
• 39:35
  SPEAKER_00: Uh, because that could be very interesting for drug researchers and people looking at candidate drugs.
• 39:41
  SPEAKER_00: But, you know, I'm gonna play with that idea.
• 39:43
  SPEAKER_00: It's sometimes it'll take me a couple of weeks, a couple of months to figure out the right way to do things, but I'm gonna look at that idea, and it came about by having an open, interesting conversation with you guys about how to address this particular need in a scientific and easy way for researchers.
• 39:58
  SPEAKER_04: Oh, that's great, Stuart.
• 39:59
  SPEAKER_04: And I must ask you now, where can our listeners find this website?
• 40:03
  SPEAKER_00: It's called longevityresearch.ca.
• 40:07
  SPEAKER_00: I'm in Toronto, so we use CA for Canada.
• 40:10
  SPEAKER_00: And here's a little trivia.
• 40:12
  SPEAKER_00: Do you know that Canada and Denmark share a border, a land border?
• 40:16
  SPEAKER_03: Yes, yes.
• 40:17
  SPEAKER_03: They leave uh whiskey and snaps for each other, right?
• 40:20
  SPEAKER_03: I've heard this.
• 40:21
  SPEAKER_00: There's an island in between Greenland and Canada called Block Island, and there's a line down the middle, and we share a border.
• 40:29
  SPEAKER_00: The only other country we share a border with is the United States, but we share a border with Denmark.
• 40:34
  SPEAKER_02: Yeah.
• 40:34
  SPEAKER_02: I did not know that.
• 40:35
  SPEAKER_02: Wow.
• 40:36
  SPEAKER_02: Yeah, yeah.
• 40:36
  SPEAKER_04: It's explosion in my brain.
• 40:39
  SPEAKER_04: It's the most polite war that ever was Yeah, I'm not surprised in the Dane.
• 40:46
  SPEAKER_04: Well, Stuart, if our listeners have follow-up questions to you or want to ask you more questions about this really cool project that you have created, how can they reach out to you?
• 40:55
  SPEAKER_00: They can reach me on the website.
• 40:57
  SPEAKER_00: There's a contact me form, and I'll answer those questions that come in that way.
• 41:01
  SPEAKER_00: I don't know the exact web link, but on the top, there is a contact me, and they can ask me any question they want.
• 41:08
  SPEAKER_00: And I welcome questions, especially ones that I don't know how to do, because it might spark a new idea for improving the website.
• 41:14
  SPEAKER_04: That's awesome.
• 41:15
  SPEAKER_04: Thank you so much for coming on the show, Stuart.
• 41:17
  SPEAKER_00: Thank you.
• 41:17
  SPEAKER_00: Thank you for having me.
• 41:18
  SPEAKER_00: And I found it a very engaging and interesting discussion.
• 41:21
  SPEAKER_02: Lovely.
• 41:22
  SPEAKER_02: Thank you, Stuart.
• 41:23
  SPEAKER_00: Thank you.
• 41:28
  SPEAKER_04: This was Lab2 Lives.
• 41:32
  SPEAKER_04: Follow us on YouTube, Spotify, Apple Podcasts, or wherever you get your content.
• 41:39
  SPEAKER_04: What is your take on how we get from the lab to making a difference in patients' lives?
• 41:44
  SPEAKER_04: Send us a note with the link in the show notes or via LinkedIn.
• 41:49
  SPEAKER_04: If you want to be a guest on the show or book us for a comedy event in your company, go to lab2lives.com