Heath Fletcher (00:13)
Hi there, welcome to the Healthy Enterprise podcast. My name is Heath Fletcher and thank you for joining me again today if you've been here before. And if it's your first time, I hope you enjoy this episode. I'm going to be speaking with Adityo Prakash. He is a founder and CEO of Verseon, where they are redefining drug discovery through deep quantum modeling and AI. The mission is to create new medicines once beyond reach and raise the standard of care for major diseases. Prior to Verseon, he founded
Pulsent a company that pioneered the technologies that helped shape modern video streaming. I'd like to welcome you to the show and introduce you to Adityo Prakash.
Adityo thank you for joining me today for this episode. I really appreciate your time and looking forward to learning more about you and about Verseon. So why don't you start there? Introduce yourself to our listeners and tell us a bit about yourself and how you got to where you are today.
Adityo Prakash (01:18)
Well, ⁓ we've been building this company, Verseon, over two decades now to change how the world finds new medicines. We set out to solve this problem back in 2002. This was our second adventure. Before this, we had built another company. Every piece of video you use on the internet today, from Netflix streaming to this call, key parts of the transmission rely on technology that we built at our previous company.
All of that technology is with Intel now. It makes Intel billions of dollars in profits. And what we like finding are opportunities where different markets are coming together. In our previous company's case, it was computers and networking and traditional media and the rise of the internet. And in those kinds of situations, the incumbent players in each field, they don't know where things are going. And we believe that's when you have the opportunity to build.
an ex-big platform company based on fundamental advancements in science or technology, and over time, convert those technology barriers into market barriers. In the case of Verseon, it was about high-tech and biotech convergence. People have been talking about this problem of ⁓ being able to design drugs atom by atom on the computer for a very, very long time, since the mid-80s or so. But the word hadn't...
turned the corner when we started the company and 20 plus years later the rest of the world still hasn't turned the corner. So ⁓ it has been a very very long journey. All of the science and technology that we use for our platform they've been all built in-house which is a complete anomaly in the current world. You know a typical VC backed company would not be allowed to do an 18-year science chase right?
Heath Fletcher (03:06)
Mm-hmm.
Adityo Prakash (03:14)
So, and this is a problem that you couldn't solve in academia either because the many of the problems that we had to solve are so hard that you might not publish for seven, eight, 10 years. That would be career suicide as an academic, all your rents. Okay. And Big Pharma couldn't do this because they're worried about quality results. They don't even know who to hire and how to together an effort like this. So.
Heath Fletcher (03:30)
That's right.
Adityo Prakash (03:42)
We feel fortunate that we have had the team, talent, timing, inspiration, luck, all of those things come together. We are now at a stage where we can actually do this. It took a long time to build this platform, but now, using it, we can systematically build completely novel drugs that were previously beyond humanity's reach. Drugs with uniquely desirable therapeutic profiles, drugs that promise to change the standard of care for the diseases that they address across a whole range of major human diseases.
It's going to be really amazing for what we hope it can do for humanity. All the diseases we still can't treat or the diseases we treat, but poorly, being able to actually change this care, the treatment for all of those diseases. ⁓ as we develop this,
It'll be one of those hopefully overnight successes 20 years in the making.
Heath Fletcher (04:43)
Amazing. then, know, yeah, I was on your website and you're, right on the website, there's this really compelling and engaging video that I really like. I wanted to just say something about that because it was, I played it, I've watched it like four times because I just, I really like it. And I really like how it tells the story behind what you're doing. And for a lot of people that are listening, they're probably thinking, well, aren't we already doing drug testing and then learning at this, but there's just something about what's going on that wasn't, I didn't know, you know, like it talks about
you know, the 30 years of drug testing that is using the same formulas, the same data, the same information. And it's, you know, the very low odds was 45 to one, which is not good. so no, very good odds. And so what I, you know, learning about what Verseon is doing is it's taking that model and really kind of disrupting it entirely because it's old, it's outdated, and it's really slow.
Adityo Prakash (05:28)
not good odds.
It's not serving the needs of humanity, really. We all live in the modern world, rely on modern medicine to keep us living healthier, longer lives, of the 10,000 diseases that affect humanity, do you know how many we can treat so far? 500 only. Yeah. Isn't that pretty crazy? And even most of those 500, we treat poorly with three mile long side effects lists.
Heath Fletcher (06:03)
in all this time.
Adityo Prakash (06:12)
We should be able to do better. This has a direct impact on your and my life and everybody else, our loved ones, everybody we know across the globe.
Heath Fletcher (06:24)
And on average it takes almost a decade to get a drug, over a decade to get a drug even approved.
Adityo Prakash (06:28)
Over and ahead.
Indeed, ⁓ all of these drugs, start at the beginning with a trial and error process to find them, okay? Though to set context for a moment, our bodies are made of proteins, all our cell structures are made of proteins. Our DNA is just a code book or a recipe book, if you will, to make these different proteins. These fit together to form all our cell structures. But some of them sometimes are associated with diseases. ⁓ Let's say there is some enzyme.
whose job inside my body is to go attach to other proteins and cut them up and form plaque that over time could cause some downstream disease. And what I want a drug to do when I take that pill, I want the drug molecules to dissolve in my gut, go through my bloodstream, find copies of that particular protein, that enzyme that's causing some disease, et cetera, and prevent it from doing that damage. Once it's bound there, it can't cut up the other ones. And this is how...
Heath Fletcher (07:25)
Right.
Adityo Prakash (07:30)
Every drug works from traditional remedies, herbal remedies, other stuff that have been handed down to us, to today's modern pharmaceutical miracles. When we take them, they go bind to some proteins in our bodies and achieve their functions. Now, sitting in the 21st century, you can ask, why does that drug molecule bind to that protein? Turns out at that level, it's actually a physics problem. The atoms on the drug and the protein, they push and pull on each other. The two things flex and twist, and they form this perfect lock and key fit. The liquid.
the water in our bodies, ⁓ mediate interaction. This is very, very complicated problem. People, once they understood what's really happening, have been talking since the mid-80s, as I said, that, we should be able to go design these things atom by atom, right? That's how we design just about everything else. Buildings, airplanes, computer chips, we design them on the computer, right? Why not drugs? But it's a really hard problem.
And despite billions of dollars spent on the problem over the last three decades, the world's still stuck. We simply don't have the accuracy. So how do we find drugs in the pharmaceutical industry today? You take the disease-causing protein involved, you purify it, you put it in these pipettes, and you stamp it down into a whole bunch of different chemicals made up and stored in tiny little test tubes in the lab. In the old days, used to be ⁓ doing this with lab operators, with pipettes.
And now we have automated that with robotics. So you stamp down all these, you know, ⁓ proteins using a bunch of pipettes, you know, that automatically move. We have a fancy name for it called high throughput screening. But the reality is it's still just root force trial and error. anything you want to test must be made before, right? And that is a huge bottleneck. So humanity has managed to make in the last 150 or so years, about 7 million distinct
types of chemicals that are drug-like, okay? If you look at all the vendor catalogs and whatever's out there, turns out if you might pull down a hundred million or a hundred and fifty million, two hundred million entries, but when you actually look at which of these are truly distinct, not Joe versus Joseph, know, relabeling to the same thing, you know, or tiny little changes around the edges of the same backbone of the chemical structure, turns out it's fewer than 10 million.
Heath Fletcher (09:57)
Wow. No.
Adityo Prakash (09:57)
Seven million. Is that a good enough number?
To answer that question, we need to ask how many possible drug-like chemical structures, backbones, can be made based on the current rules of chemistry. That number turns out to be 10 to the 33, one with 33 zeros. Wow. When you compare that to seven million, you realize we're not even... You know, to use the ocean analogy, we're not even fishing in a...
Heath Fletcher (10:21)
We're scratching the surface.
Adityo Prakash (10:26)
We're fishing in a droplet. Right? And no matter what disease you're going after, we're testing the same little pool of chemicals that we've made so far. And we are slowly trying to grow it, but this is a very, very slow process. So if you wanted to go find great new ⁓ fantastic drugs that nobody has ever made before, the kinds of structures, the brute force process isn't going to get us there.
Heath Fletcher (10:55)
No, we've already proved.
Adityo Prakash (10:58)
Now, yeah, that's why we have this dismal odds for finding new drugs. The long time that it takes to come up with anything. ⁓ Obviously, we all as society deserve better, right? And nowadays people are saying, ⁓ I'll use AI to solve all problems. ⁓ It's on everybody's mind is the biggest.
Heath Fletcher (11:21)
Yeah, yes. Because
AI will solve all the problems.
Adityo Prakash (11:28)
Yes. Every problem in society, why not medicine, ⁓ You know, AI has applications in many aspects of healthcare, ⁓ making your diagnostics quicker, cheaper perhaps, ⁓ doing treatment recommendations, ingesting all of the data that you feed it, everything that we have ⁓ so far collected. But right there is also the Achilles heel of AI.
Yeah. As we know today, AI needs a lot of data for training. Then when you ask for something similar, it knows what to protect.
Heath Fletcher (12:06)
Right.
Adityo Prakash (12:09)
Train AI on the existing data of the seven million or so distinct types of chemicals we have made so far and how they work, what they bind to, which proteins, et cetera. What it will give you back are small tweaks on those same molecules. Right?
Heath Fletcher (12:23)
Right.
It's only as good as the information it gets.
Adityo Prakash (12:27)
Yes, correct. You know, if you go ask, type it into Google search bar, why do donuts have holes? And nowadays you get the AI summary, because you can ⁓ bake it more uniformly, blah, blah. Did AI figure that out? No, some human had to do this and they posted it on some websites that this is what happens and the AI is spinning it back at you. Right?
Heath Fletcher (12:39)
Yeah.
Right.
Which is great, and it's great for research that already exists. If you want to find an answer to a question...
Adityo Prakash (13:01)
Nobody is discounting the fact that it can improve productivity in certain areas, to expect it to do magic, that's not happening. It's going to give you back molecules that are similar to what it has seen, even with so-called generative AI and whatever else. It'll just skirt the edges of what it has seen at best. And that's exactly what we're seeing with all the current generation of
quote unquote, AI and medicine companies. All they're producing are tweaks on old known drug molecules, getting around J &J's patent, getting around Vivorion's patent, getting around various other companies' with small little changes, okay? ⁓ And by the way, most of those drugs that this set of AI companies have been putting into phase two have been failing so far. So that's kind of a sobering realization.
Heath Fletcher (13:54)
Yeah, it is so boring, isn't it? Yeah. Because really at that point, all you really don't all we're really looking at is a bunch of different drugs that are slightly different from the other. And it's only a comp. It's basically a competition on who has the best branding and marketing is going to sell the drug. You can get at that point. Yeah.
Adityo Prakash (14:12)
You can get it through there.
Assuming that you're just tweaking something that somebody else has already tried and put it out there and maybe already has a marketed drug, right? ⁓ Yes. Is it a useful thing? Some of them, if they come to market, will they make some contribution? Sure. But the great new medicines of the future that we all want and deserve, they're not going to be found this way.
Heath Fletcher (14:20)
Right.
They're out there.
Adityo Prakash (14:40)
Yes, correct. They're out there in that uncharted ocean and no way to get there by just training AI on the existing data. You need something else. And this is where being able to design atom by atom, completely new drugs using the physics of, you know, how these things will form chemical bonds with the protein, where those will form, how it'll flex and twist comes in. You know, let's say you're trying to build a building, a big skyscraper in Vancouver or whatever.
Heath Fletcher (14:43)
Yeah.
Adityo Prakash (15:10)
in that ⁓ general part of the world, let's say. What do you do? You hire an architect and the architect comes, looks at the plot of land and ⁓ everything else, all the data about it that they have. They design something to your requirements ⁓ and they show you how every room will look, where the load bearing structures need to be. By the time you build it, it works like that because they are actually using all these
physics-based models to say, the load-bearing structure for this building with this much weight, you know, and the balcony sticking out this way needs to be like this, okay, for it to hold. And it works. Same thing here, but you have to build it at the atomic level, these kinds of CAD-CAM tools at the atomic level, and those don't exist. That's what took us 17, 18 years of the company's life to build.
Heath Fletcher (16:07)
That was what you were focused on for all that time.
Adityo Prakash (16:10)
Yes. So once you can do that, now you can design completely new drug structures that have never been made before, but that are actually perfect binders for any protein that you're going after. And you don't just design one such thing. You design hundreds of these things. This so far, when you have designed it on the computer, there are a couple of other problems ⁓ that come up. If all you did is design something that kind of makes sense where these
carbon atoms should be connectable by these bonds and you should be able to hang off this nitrogen or whatever else. That's not entirely useful unless you can also tell this drug that you just designed how to make it. So you need to have your system be able to tell your chemist, here are the five, six, seven steps you have to go through in the lab to make it. We have to solve that problem too. So far now you actually have a viable, actually makeable.
Heath Fletcher (17:04)
Wow.
Adityo Prakash (17:08)
or in chemistry parlance you can say synthesizable molecule that'll do the job. That's great. That's a, when you have designed these things on the computer and you know they're makeable, this is what you'd consider synthetic data in the AI parlance. try not to use the word synthetic data because since we operate in the ⁓ pharmaceutical or biotechnology world, the moment you say synthesis, mean synthetic data, they think.
Heath Fletcher (17:27)
Okay.
Adityo Prakash (17:38)
you're actually synthesizing a molecule in the lab. But in the AI world, we call that synthetic data or virtual production of data. So that's of completely novel structures that nobody has ever made before. AI has never been trained on such things. But you don't want to train your AI with just the synthetic data. It's time now to make them in the lab. All the different ones that you've designed for a particular protein. It's time to make them in the lab.
Heath Fletcher (17:47)
Right.
Adityo Prakash (18:07)
carry them forward to all the standard batteries of tests. Pick the best ones, but also show the ones that didn't work. What happened to them? What kind of lab data did they generate? You're generating completely novel chemical and biological data that the rest of the world doesn't have. Now you can feed all of this to your own AI. And your own AI now can learn from it and create variants, which is what AI is good for. Says, ⁓ mix and match these things, take the head of this molecule and actually the tail of this guy.
You should have even better properties that you're looking for itself. But you don't trust the AI just giving you that ⁓ suggestion. It's just there, as you were saying, as a support tool to improve your productivity.
Heath Fletcher (18:49)
moving the data around that you're providing it and giving it lots of ⁓
Adityo Prakash (18:52)
Now that it has suggested something, it's time to go make them and test them again in the lab.
Heath Fletcher (18:57)
Right.
Adityo Prakash (18:59)
It starts not with AI, but fundamental quantum physics breakthroughs we call deep quantum modeling. You do that, you come up with these completely novel drug structures. Now you collect actual lab data around them, use AI to create further variants and the best ones you advance into clinical trials. And this changes the process. Today, you were mentioning the odds of success in the pharma industry. If you started 45 different drug programs,
By drug programs, we mean going after different protein targets for different diseases, right? After four or five years of all the trial and error and AI and everything else, two thirds of those programs, 30 of them, will produce no viable clinical candidates that you can put into clinical trials. The 15 that survive with one drug candidate each, you put them to clinical trials, and only one of them reaches market as a new entity. No other industry would survive with these kinds of odds.
The pharma industry, on the other hand, makes over a trillion and a half dollars in patent protected drug revenue each year. What keeps pharma CEOs up at night is that even the one drug going through phase three trials, et cetera, it's not a sure thing. It's a highly erratic outcome. If they have a sudden failure of a high profile drug in phase three, they could see a hundred billion dollars wiped off their market cap overnight. our ⁓ process, our whole objective,
Heath Fletcher (20:11)
No.
Adityo Prakash (20:26)
our whole last two decades of work is to change this whole trial and error ⁓ with really, really low odds into something that's systematic. That whenever you start a program, should always be able to great new candidates that are ready to get their trials. And not just one, but multiple for every program. That changes the odds of this entire business. But even setting the business problem aside for a moment, that changes.
how well we treat human disease, what we can offer to people who need the tuneups. As we go through life, we all need little tuneups, right? We all suffer from diseases. Wouldn't it be nice to have ⁓ much better options coming at a steady stream to treat all of these things, to take sort of...
Heath Fletcher (21:06)
Yeah.
Adityo Prakash (21:23)
the stage of current medicine to whole new level. We all read about biotechnology breakthroughs in newspapers, right? Or we hear about them in podcasts. Oh, we figured out how some specific drug or disease process works, okay? Wow, this would be super exciting. But then when you actually read the fine print, it says it might become a therapy in 15 or 20 years.
still early days, right? Does it take that long? Because it's a completely erratic trial and error based process. And we want to be able to change it for the benefit of everybody.
Heath Fletcher (21:54)
Yeah, very early.
So that's where the deep quantum science comes in, right?
Adityo Prakash (22:16)
that allows us to find things that you can't find by just trolling through existing experimental data.
Heath Fletcher (22:23)
Did you know that the quantum physics was something that you were you were going to be utilizing like 17 years ago when you first? ⁓
Adityo Prakash (22:31)
Absolutely,
we set out originally to solve that problem.
Heath Fletcher (22:35)
That was the problem you wanted to solve.
Adityo Prakash (22:37)
that
was the problem, to actually come up with completely novel drug structures that would bind to the protein of interest. And you should be able to tell that on the computer without having to make those things and do the experiment in the lab. Replace that experiment with pure computer simulation, but with high enough fidelity, high enough accuracy, that it actually replicates what would happen if we actually did the experiment. That's where everything's stuck.
Today, there are companies that build tools for these physics model, but they simply don't have the accuracy. Okay? In this business, even if you are, nowadays they're not even close to that, you they're about 70 % accurate on some simpler systems, let's say. But even if you're 99 % accurate, okay, it's not good enough because you're gonna run through billions of different possibilities. If even one out of every 100,000 or one out of million, that's a good fit.
Heath Fletcher (23:12)
Right.
Adityo Prakash (23:36)
the false positives and the false negatives are going to kill you if you don't have ⁓ accuracy that really ⁓ is at a level where you can truly be able to replace the expression.
Heath Fletcher (23:47)
And so now you actually have drugs that have hit various stages in the clinical process, ⁓
Adityo Prakash (23:55)
And that's the exciting part. This would be just talk, unless
you have results to prove what you can do. Okay. And our drugs in our fast growing pipeline show what a system like this can do. Have you seen older people with bruises on their hands and faces? Yeah, we all have, right? We have friends and family members who are at risk of getting strokes or heart attack. They have a slight arrhythmia in their heart as they age.
Heath Fletcher (24:12)
Yes.
Adityo Prakash (24:24)
They're given these so-called blood thinner drugs, these anticoagulants to prevent clots from forming that lead to heart attacks and strokes. Now, all these drugs simply can't strike the balance of preventing those clots without turning you into really prone to bleeding, right? You bump your arm on the way out the door and it becomes an internal bruise that lasts for a year. And that's what we see in these older people.
Heath Fletcher (24:45)
Right.
Yeah.
Adityo Prakash (24:54)
And it's not just cosmetic. They bleed when they brush their teeth. They have internal GI bleeds. Their doctors tell them, you just have to live with it. Because the alternative is you stop the drug and you end up with a clot forming in your heart that gets pumped out into your brain and you're dead. We have developed drugs that prevent these clots from forming without increasing the bleeding risk in any meaningful way. Now that would be considered almost like science fiction, but we the results to prove it.
Heath Fletcher (25:23)
Yeah.
Adityo Prakash (25:24)
And
you cannot find drugs like these by just, you know, iterating through the existing.
Heath Fletcher (25:30)
Using the same recipes.
Adityo Prakash (25:33)
Correct. So these would potentially change the treatment paradigm for a half a billion people around the planet who are at risk of these strokes and heart attacks. know, most of us as we age past a certain point, we are at risk of these things. The joke among doctors is that you either die from heart disease or die with heart disease. Right? So... ⁓
That being the case, these are really, really important. So here are some examples. And we didn't just find one such thing. We have multiple candidates going forward into clinical trials. We have another set of drugs for the damage that diabetes causes.
You know, if somebody has diabetes long enough, their blood vessels become leaky after a while. And this causes damage in the back of their eye, they lose their eyesight. One third of all diabetics today are suffering from what's called diabetic retinopathy. It's a huge problem. Over 200 million patients around the globe. Okay. You want to guess what the treatment is for these people today?
Heath Fletcher (26:30)
Right.
I can't even guess.
Adityo Prakash (26:55)
You wait while the disease progresses. Initially when you find it, you say, can't do anything. You're still too early. Once you're almost blind, then I can convince you that you need to be willing to take these monthly injections in the eye. Okay. With a repurposed cancer drug to only treat the symptoms. Can you imagine that getting jabbed in the eye?
Heath Fletcher (27:13)
was a freak!
Adityo Prakash (27:21)
The big innovation that has come along in the meantime is that they have moved it from a monthly to once every two months injection. But still, that's ridiculous.
Heath Fletcher (27:29)
Yeah, that's ridiculous. That's what we've come to.
Adityo Prakash (27:32)
And
not to speak of the potential side effects of this treatment. ⁓ You can end up with infections in the injection site. You have all kinds of other things. You even have things like corneal melting. Cornea and melting should never be next to each other. So that's the kind of treatment we have today. My mom was getting these things.
Heath Fletcher (27:49)
in same sentence.
Adityo Prakash (28:01)
And 50 % of patients that get them see no improvement, but they're desperate. They're going.
We have developed drugs that can take rats with raging diabetes, treat them with this drug, completely stop the leakage and reverse the disease. The best part is an oral pill. You can give it to people at the earliest signs of disease. We're the only company that's ever done this.
Heath Fletcher (28:23)
Even
even things like that, even with drugs that are ⁓ that are treating certain things right now, even eliminating the side effects or the risks that go.
Adityo Prakash (28:35)
Of course, but you know today's drugs as I said, they don't even address the root cause all the leakage they they leave it alone. They can't do anything about it. They just simply treat the symptoms the downstream symptoms with these eye injections. We're saying hey, why not tighten up those junctions and can we develop something that will do that and we did you know, this will treat what's called diabetic retinopathy of the earliest signs, you know, so you don't
Heath Fletcher (28:56)
Easy.
Adityo Prakash (29:05)
slide down that path of losing your eyesight. That's just for starters. Now, the same fluid leakage causes all sorts of other end organ damage in diabetes. This is a technical term for it. It's called diabetic end organ damage. This is the same fluid leakage, why diabetics end up losing their kidneys and going on dialysis, then damage their liver. We can stop all of that. Now, my target is not just 200 million people
Heath Fletcher (29:07)
Right.
Adityo Prakash (29:33)
suffering from diabetic retinopathy, but almost 600 million diabetics around the planet, stopping all their downstream damage in their bodies. So are these useful? Yeah, it can change people's lives in very meaningful ways, right? We are developing three different cancer programs. One that prevents cancer from metastasizing, goes after one of the major metastatic pathways for cancer. And another
is an immune oncology drug that cranks up your immune system against a whole ring of major cancers, lung, pancreas, colon, et cetera. But we're also developing a novel chemotherapy agent because chemo is still the baseline therapy for most cancers. And the worst news cancer patients' families hear is that the drug has stopped working because the tumor mutates and the drug no longer works. We've developed novel chemo agents that even when the tumors mutate, they keep working.
Heath Fletcher (30:21)
Great.
Adityo Prakash (30:31)
Zildem oncology is a fantastic new tool in the arsenal.
Heath Fletcher (30:35)
Amazing.
Adityo Prakash (30:37)
And program after program, what I'm telling you about, these drugs, the results we are showing to the world, these are drugs that you cannot find unless you can go out in the uncharted chemical ocean. We'll find things that ⁓ people haven't made and tested so far.
Except for when you can do it our way and then actually make them in the lab and show what the...
Heath Fletcher (31:03)
So these drugs now are actually in human trials now.
Adityo Prakash (31:08)
Actually, we have one drug pass phase one, we have four more ready to enter clinical trials, we have many more, six more behind that that will enter trials within a year. We have a fast growing pipeline that's moving forward.
Heath Fletcher (31:22)
mean even like so you spent the first 17 years like you say developing the computational Intelligence and now even in just in those last three in the last three years you've already Created these many drugs that
Adityo Prakash (31:41)
span
of time we've got 16 different drug addicts across eight major disease areas and we're just getting started.
Heath Fletcher (31:47)
That's incredible. Wow. Wow. Okay. Take me back a bit because when you, when you first came up with, mean, you had the vision for this company and you were the, what, what was it that sort of inspired you to, know, I guess there was probably a group of you that, actually founders. Yeah. You're one of them. You're one of them. Yeah.
Adityo Prakash (32:09)
three co-founders we got three co-founders
Yeah, I'm one of them. And we asked ourselves, can we make a difference in how the world finds new medicines? Because this is a problem that has affected all of us. We have all had friends and family members with diseases that, you know, ⁓ don't get treated very well. They suffer. when something you care about is suffering from some disease, getting them better cares and getting them better matters far more than any toy I can buy.
Heath Fletcher (32:41)
Yeah. Yeah. How many toasters do we need?
Adityo Prakash (32:47)
Correct. So we sat down, we looked at this problem. One of our co-founders, he had built one of the industry's first bioinformatics platforms for early gene discovery. Oh, wow. And during that time, in the 90s, during that time, what he saw was that the companies that survived were not just the companies that came up with a whole bunch of information about new genes. That's clearly very valuable, okay? But...
The companies that actually did well were the ones that reinvented themselves as other drug company or a diagnostics company, because that's what people need and pay for, right? Something that touches them. The gene information is extremely valuable and useful, but ultimately needs to be translated into something actionable, right? So we said, we want to really look at where things are today. ⁓ People have been talking about
designing drugs on the computer by then already almost, you know, 18, 19, 20 years, you know, ⁓ what has changed? How have people ⁓ advanced so far? And what we realized is that not much has happened other than some toy solutions that don't really work. ⁓ And ⁓ we are not afraid of taking on big challenges.
Heath Fletcher (34:05)
All
Adityo Prakash (34:13)
But at the same time, we're realistic about what we can deliver. We're not going to promise you that I'm going to put you on Mars tomorrow, right? We said, OK, let's, we think we have some good ideas on how to do this. But this is going to be a long haul. And we sort of decided among the three of us that we wanted to actually jump into that. That's the beginning of the.
Heath Fletcher (34:39)
take this on. That wasn't
big. So the three of you came up with this idea. And now are you are you where you wanted to be at this point? And had you wish you would be? Would you like to have been further along? Or is this kind of where you expected you kind of assumed this was going to take this long?
Adityo Prakash (34:54)
You
know, there, if you always, if you look back, you always wish you could do certain things. Et cetera. And yes, we would have loved to be further along by now. But for instance, there are outside events, geopolitical things, other things like those that have affected our world. 2008 crash really slowed us down. Okay. I mean, at that time, I think.
Heath Fletcher (35:02)
Sure, why not.
Adityo Prakash (35:24)
40 % of all biotechs in the country went out of business. And we saw a complete shutdown of any kind of funding, which is critical for ⁓ moving things forward. when you are operating on a shoestring budget with not enough funding, things slow down. those kinds of things have slowed us down. COVID slowed us down. yeah, those things happen, but you keep going because for...
Everyone that works at Verseon, this is not a job, it's a mission. We really want to change how the world treats human disease.
Heath Fletcher (35:57)
Right.
And now you're here. Now you're at that tipping point, right? You're at the, you've got the boulder at the top of the mountain, right? And yeah, I bet you're.
Adityo Prakash (36:12)
There is still a lot
of work to do. We're developing these drugs. We still have to take them through clinical trials. We have to get them to market. And there's a lot of work in building a full-fledged, fully-integrated pharma company.
Heath Fletcher (36:26)
What kind of pushback have you experienced in this? I mean, you're talking about some pretty, ⁓ know, quantum physics is something that is being adopted now as a mainstream sort of thing, but it hasn't always been. It was kind of like one of those sort of things, like you said, science fiction. was like something you saw in the movies.
Adityo Prakash (36:49)
Today
you hear a lot about quantum computing, ⁓ making hardware that follows the rules of quantum mechanics ⁓ for ⁓ general purpose computation. That's fine, that's great. As true general purpose quantum computers come along, we will be happy as some of the first people to try to use it. But what we're talking about is actually on classical computers, ⁓ simulating.
Heath Fletcher (36:52)
Sure.
Adityo Prakash (37:17)
how a quantum system behaves, okay? And this is a very, very, very hard problem. There are many, things, even if you had a quantum computer, you'd still have to actually think about and solve many of these problems that are completely non-trivial, okay? And we have to do that. once you create all of this, lot of things change along the way. It's not just one thing, right?
The quantum physics modeling is important, but you also have to solve when you're coming up with these novel drug molecules, how do you synthesize them in the lab? Can you always guarantee synthesizability? That's a whole another set of problems ⁓ related to our knowledge of chemistry, how to convert that into something that's actionable, realizable every time, Non-trivial, how do you integrate all of the biological testing and characterization in as efficient a process as possible?
Heath Fletcher (38:02)
Right.
Adityo Prakash (38:16)
Again, not easy. All these things, these breakthroughs in physics. ⁓
chemistry, ⁓ biology, et cetera, they all have to come together ⁓ in combination, working with AI to create further variants as one of the many tools to create this change and this transformation and how you come up with new medicines.
Heath Fletcher (38:43)
What was that criteria for you for which diseases to go after first? How did you make that choice?
Adityo Prakash (38:49)
⁓
So once you build a general purpose platform that can go out for almost any disease, now the only smart thing to do is to make your decision on what you're going after based on the business side of things or ⁓ the sort of large medical need side of things.
Heath Fletcher (39:13)
Yeah,
that makes sense.
Adityo Prakash (39:15)
Those
the programs in our current pipeline and we'll continue to grow them, right? When we chose the programs based on very large unmet medical need, right? Well understood ⁓ biology of how the disease progresses. clinical endpoints. So if you look at our current drug pipeline, we do not have any ⁓ central nervous system disorder drugs, meaning drugs for the brain. Do you know why?
Heath Fletcher (39:44)
No.
Adityo Prakash (39:46)
because the clinical pathway for those drugs is really, really poorly understood. the diseases themselves are poorly understood. We don't even know what protein is involved in what action, know, what to go after. Designing the perfect drug for the wrong protein doesn't help you, okay? For that specific disease at least. You know, it might turn out to be it's useful for something else and we're happy to do those kinds of things and take those, you know, poke those shots as we grow as a company.
Heath Fletcher (39:52)
Okay.
It's my shot.
Adityo Prakash (40:15)
There
is something far worse though. Even when we kind of think we know how a particular central nervous system disorder or brain disorder works, it is not a drug. But who do you give it to? Let's say it's Alzheimer's, right? Either time you find that somebody has Alzheimer's, it's way too late. It's like scrambling in length. Everything inside their brain is already scrambled. You needed to give that drug to them 20 years earlier.
Heath Fletcher (40:31)
Right.
Right.
Adityo Prakash (40:44)
But how do you find them? Only now people are improving the diagnostic. Find early should be treated. Well, those are still just in their infancy, they're evolving. So that helps with the process. there is more. How do you figure out ⁓ whether you're actually helping with Alzheimer's? Do you follow this person for 20 years? That's a non-starter as a clinical process, right?
Heath Fletcher (41:01)
That makes sense.
Adityo Prakash (41:13)
You know, as a company, would be suicide. So you don't want to have your clinical trials take 20 years. So you need to have some sort of surrogate biomarkers that you can measure to say, hey, look, I'm improving these things.
Heath Fletcher (41:28)
You
need a high impact win that's going to show.
Adityo Prakash (41:33)
Just staying with the same disorders. What I'm trying to tell you is that even if you figure out, this is a guy that's already showing early signs of Alzheimer's, which we till very recently had no idea how to tell. And we're going to treat it. But now you have to be able to submit some data to the FDA or the European Medicines Agency and say, hey, guess what?
I treated it and I made changes in these specific biomarkers in their body, in their blood ⁓ tests or some other kind of tests, I see these changes. They need to accept that that's a good surrogate for actually improving Alzheimer's. Instead of coming back at you and saying, no, you gotta just keep monitoring the guy for the next 20 years. Tell me if he develops Alzheimer's. Full blown, right? So.
Heath Fletcher (42:14)
Right.
Right.
Adityo Prakash (42:25)
This is a problem with certain disease areas, especially brain disorders.
Heath Fletcher (42:29)
Right.
Which is why you chose that.
Adityo Prakash (42:32)
or
you take that risk as a business, you ask how we choose these programs. People often ask us, why don't you have an Alzheimer's program, some other drug program? ⁓ And we say later, we need to grow up as a company to be able to handle that kind of ⁓ uncertainty and risk in developing these drugs.
Heath Fletcher (42:58)
Right. And something like diabetes, which affects millions of millions of people, you can have an impact, have a win and have validation that this works.
Adityo Prakash (43:10)
Absolutely. Diabetes, heart disease, so many other metabolic disorders, cancers, you know, these are all tangible where you can have wins, you can change billions of people's lives.
Heath Fletcher (43:23)
And it affects people at all ages, not just, right? Not just in the later stage. Yeah, makes sense. So tell me how has been your evolution ⁓ as a leader ⁓ in this company? You you're the CEO. You know, how has the evolution of leading this company been for you, you know, from where you were at, you know, 17 years ago? How do you how have you developed yourself?
Adityo Prakash (43:50)
You grow as a person through our experiences, right? You learn, you adapt, you change. Hard to tell how that has been, ⁓ except for the fact that we hope that we always keep learning ⁓ as a company, every single person within our company. We believe that even when we're trying something, remember, we're doing something completely new. It's okay to make mistakes. It's not a problem at all.
People are encouraged to take risks, try something new, et cetera. What we say within the company is that we always want to learn when something doesn't work, so we don't make the same mistake twice. Okay? That's all you can do. And keep learning, evolving, making yourselves better. That's what we try to do. ⁓ Every single person within the company.
Heath Fletcher (44:33)
Right.
What do you do for yourself to inspire yourself every day? Do you ⁓ mentor with somebody? you ⁓ listen to podcasts? What do you do to kind of keep your leadership skills sharp?
Adityo Prakash (45:06)
Actually, if there are interesting articles talking about ⁓ some new insights on corporate organization or psychology or whatever else, you read them, you learn, but you also learn from your own experiences, you know, always analyze, figure out what works, what doesn't work, continue to evolve these things. ⁓ And ⁓ it's something that
unless you are willing to take as an integral part of who you are as a person, not just leadership. It's about everything we do. We should be learning every day.
Heath Fletcher (45:46)
Right. Makes sense. like that. ⁓ What do you, ⁓ with the company where it's at now, how do you market? How do you find people to support you in your cause and to ⁓ get on board? mean, investors are probably a large part of the process. How do you reach them? What is it about what you're doing that attracts them?
Adityo Prakash (46:16)
Well, the right investors, they see the kind of transformation we bring to the table, right? Bringing to the world, really. And ⁓ we're now, as a company, as we're growing, we're coming up to the stage where we want to reach out and tell this message to people who are looking for technology-driven complete transformations of entire industries, not people who say they want to support innovation.
But what they really mean is they want to support iteration. Too many people just want to do that. ⁓ Despite whatever lip service they pay to innovation, they don't really understand what true innovation looks like. It doesn't look like anything that you've already seen. It doesn't fit an existing bucket. Everything that we are doing as a company, we don't look like a typical biotech or biopharma company or a typical pharma company. And we're not supposed to.
Heath Fletcher (46:48)
Right. It's very different.
No.
Adityo Prakash (47:16)
So, but we want to get the message out and not just to the investor community. We want to get the message out broadly to the world because this is something that matters in all of our lives. We all live in the modern world. We rely on modern medicine to keep us longer, like I said, okay? And changing what we treat, how well we treat them, how extensively we can treat all the things that happen to our bodies as we go through life.
has an incredible impact on our lives, right? Health plus time is the most precious combination we can have in our lives.
Heath Fletcher (47:58)
That's right. Yeah.
Adityo Prakash (48:01)
We can make more money, we can buy more toys, but we cannot create more time. If we can have more healthy years on this planet, it's surprising how short that time so far is, right? If we can extend that and not just ⁓ extending the time, but actual healthy, productive time. ⁓ think of anything more valuable than that.
Heath Fletcher (48:07)
Right.
Yeah, I think I've heard the difference between calling it a health span as opposed to a lifespan, where your your span of your life is more is living in health as opposed to just living life, right? So an important distinction there for sure. Well, Adityo, this has been a fantastic conversation. It could go on probably for a couple of days. There's so much like scratching the surface. We barely scratched the surface today.
what you're doing at Verseon. It's very exciting and it's ⁓ fun to listen to for me. I really enjoy this conversation.
Adityo Prakash (49:08)
Hit, it was a pleasure. And I would love to talk more in the future sometime about things that evolve as we see them evolve within the medicine landscape. Have a great afternoon.
Heath Fletcher (49:27)
We'll be talking again, absolutely. Now, if somebody wants to get more information about Verseon, where should they go? Is the website the best place to start?
Adityo Prakash (49:35)
Right?
You know, www.verseon.com is a good place to start. And we're slowly building a community site and various other places where they can request more information. And we want to build a whole community that wants to support this new kind of transformation.
Heath Fletcher (49:54)
I could see that. Yeah. So a community site that kind of allows people to kind of engage and communicate with each other.
Adityo Prakash (50:00)
It's just that it's Infancy.
Heath Fletcher (50:03)
is that a great idea? That's a wonderful idea. Well, again, thank you so much for your time today and I really appreciate you ⁓ chatting with me and I look forward to doing it again. So we'll do it again. I promise.
Adityo Prakash (50:16)
Thank you, Eid. ⁓ Enjoy beautiful British Columbia. We'll talk again soon. Okay, bye.
Heath Fletcher (50:25)
That was a fascinating conversation, with Adityo, wow, they're not just trying to make drug discovery faster. They're actually rethinking it from the ground up. Deep quantum modeling, not three words I've ever thought I'd say together, but ⁓ sounds pretty cool. And where physics meets biology and computer intelligence meets creativity.
Instead of relying solely on historical data and trial and error, they're creating new information and designing new drugs atom by atom, unlocking therapies that could address some diseases we once thought were untreatable. So it's really a glimpse into the future where we're not just improving existing medicines, but fundamentally changing how we discover them. Wow. I hope you enjoyed this episode as much as I did. ⁓
Be sure to follow us, subscribe, so you don't miss the next one and ⁓ share it with your friends. I hope you ⁓ enjoyed today's chat and ⁓ until next time, thanks for listening.