Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Ep. 28 Aneil Mallavarapu on Building and Monetizing TechBio Platforms, Storytelling and Why Diagnostics Won’t Save a System Designed to Fail
0 Plays2 seconds ago
In this expansive episode, Aneil Mallavarapu delivers a masterclass on what TechBio is really about and why most investors and founders are still getting it wrong. We dig into the communication gap between technical genius and venture capital, the brutal economics of research tools, the data dilemmas in wet AI, and why the future belongs to founders who can both build deep and speak simply.
Aneil shares hard truths about platform strategy, revenue (yes, revenue), and the myth of “cool diagnostics” inside a broken healthcare system. But this isn’t just about critique. It’s a hopeful, strategic call for a new generation of purposeful companies, built not only for returns but for impact. If you're working at the intersection of biology, tech, and meaning, this is the one to listen to.
Recommended
Transcript
Introduction to Sparktime Podcast
00:00:01
Speaker
Welcome to Sparktime, where biotech's thought leaders, investors, CEOs, and industry experts break down the evolving story of life sciences. Hosted by Danny Stoltzfus and Will Riedel, two scientists and strategic communicators, we dive deep into how biotech leaders can shape the narrative, win investor confidence, and communicate breakthrough science in ways that truly resonate.
00:00:21
Speaker
From emerging trends and cutting-edge technologies to what investors and partners really want to hear, we go beyond the usual echo chamber, bringing you fresh insights, unexpected perspectives, and the strategies that set biotech's top players apart.
00:00:34
Speaker
If you want to sharpen your corporate messaging, decode industry shifts, hear from voices shaping the future of biotech, and get inspired, then you're in the right place. Let's get into it. Okay, welcome in
Guest Introduction: Anil Malavarapu
00:00:46
Speaker
into our listeners. Today, we were fortunate to be joined by Anil Malavarapu of Humane Ventures.
00:00:53
Speaker
This conversation meandered, I think, in all of the best ways. I was able to put Anil into the hot seat for his comments about biotech almost immediately.
Tech Bio Breakthroughs and AI's Role in Therapeutics
00:01:02
Speaker
But we quickly went deep into the breakthroughs that Anil sees happening in tech bio, how AI is actually changing the game in therapeutics and beyond, and and spoiler, that's not only letting us analyze more data more quickly.
00:01:16
Speaker
We also got to the social and ethical implications that tech bio leaders should really deeply consider. Okay, let this episode spin.
00:01:27
Speaker
Well, today on Sparktime, we have a remarkable guest whose career sits at the crossroads of science, technology, and investment. Anil Malvarapu is a computer scientist turned entrepreneur, PhD from UCSF, with deep roots in systems biology, biochemistry, and AI-driven biotech.
00:01:45
Speaker
Anil was instrumental in pioneering the Harvard Medical School Virtual Cell Program and helped build industrial genomics platforms used by several leading biopharma companies. After executive roles and founding ventures in the Bay Area, Anil now co-leads Humane Ventures, investing at the frontier of tech bio, which is, of course, the next way of transforming biotech with AI and automation.
00:02:08
Speaker
Anil was also a co-founder and advisor to Precise.ly, a venture spotlighted by TechCrunch to expand personalized genomics into India.
Comparing TechBio and Traditional Biotech
00:02:17
Speaker
Anil joins us today to unpack what sets true tech bio apart, how the investment landscape is shifting, and what he believes is the AI and biology revolution's next chapter. Anil,
00:02:28
Speaker
Welcome to Sparktime and how are you doing today? Hi, Will. Great. Nice to be here. Yeah, great to have you. Well, let's just start off. You know, TechBio is gaining recognition as its own asset class.
00:02:41
Speaker
It's distinct from the traditional biotech and AI, but I'd love to hear how you define TechBio and what makes it an especially attractive category for investors right now. Well,
00:02:52
Speaker
tech Tech bio, I think people read as, is this just another way of saying biotech? And it's really ah quite different ah because as the name suggests, ah it is technology-driven biology.
00:03:08
Speaker
And you may think, well, hasn't biology always had some technology in it? And that has been true. Technology has always played a role in biology. But I think that there's been a shift by um because of a couple of trends.
00:03:23
Speaker
One is we've had an accumulation of many scientific techniques, and very much it's been the arrival of ai and compute at scale, which is now allowing us to do things that we haven't been able to do before.
00:03:40
Speaker
Biotech, I think, is coming out of ah the academic labs of the 70s and 80s, really out of our understanding of you know the central dogma, understanding molecular biology and the tools and techniques that scientists developed to develop.
00:03:59
Speaker
you know manipulate DNA and primarily to discover the pathways that underlie disease processes and life in general.
00:04:10
Speaker
And biotech really was you know, I'm painting with a very broad brush here, but biotech was really about a discovery process of finding out where to develop ah new drugs.
00:04:26
Speaker
And the typical model ah was that we would find, you know, an interesting pathway, an interesting bit of biology. And those you know those those target that that would identify targets for discovery. And there's a part of the pipeline which is called target discovery,
00:04:48
Speaker
After that point of target discovery and validation, you would really get into developing small molecules in a way that has been done for a very long time by the pharma industry. And I was at Millennium Pharmaceuticals where that's what we did. We produced targets at scale and then threw them over the transom for relatively traditional ah small molecule drug discovery development.
TechBio's Disruptive Potential
00:05:15
Speaker
And that's that's biotech.
00:05:18
Speaker
those Those companies you know tend to have like a single asset, like 60% or something like that of biotech startups have a single asset, a single drug candidate that they're bringing to market.
00:05:31
Speaker
Whereas tech biocompanies are really technology platforms which are working on a new way of, say, developing drugs and built ah bringing many, many different assets to market. So that's generally how I would say they're different.
00:05:46
Speaker
And tech bio is more general, too. um um I think of it as biology becoming an engineering platform, and I compare it to the arrival of the Internet, where you have a basic enabling technology where we can really do engineering at scale. And of course, you know, the boundaries between these two things are a little fuzzy, but something has definitively changed now, really because of AI and compute at scale.
00:06:15
Speaker
Yeah. Well, I want to put you in the hot seat right away. and And by pointing out to our listeners that you said biotech was biotech in the past tense.
00:06:27
Speaker
Well, tech bio is going to dis disrupt biotech. That's the way we look at it. love it. A little bit here. But yes, tech bio is a disruptive approach.
00:06:38
Speaker
It is. ah You know, theyre there are big problems in pharma. And I think that we can see that in some of the basic metrics where we have a 90 to 95% failure rate ah in clinical trials.
00:06:55
Speaker
And that those numbers haven't really changed. And there's some argument that they've actually gotten a little worse over time as you know kind of patent busting strategies have been have have proliferated where people are doing small modifications to existing well-known molecules.
00:07:15
Speaker
And of course, there, you're in a game of showing that you can get a little bit more efficacy and you have novelty and you're really looking just for patent novelty and maybe some more efficacy.
00:07:28
Speaker
And you're not really exploring novelty in terms of the product. And the underlying problem, you know, in my view, is that small molecule drug development is, you know, working on a platform that fundamentally has problems.
Challenges in Drug Development
00:07:46
Speaker
And we do that because medicinal chemists know how to make small molecules. And when they try to make bigger ones, they fail.
00:07:55
Speaker
and that ah you know that poses a limit over the entire industry small molecules because they're small they're they'll bind to many things so they have off-target effects and then you have a very narrow therapeutic window where you know if you go higher ah you want to have efficacy so you want to have binding of your target but you also have off-target binding. So those off-target binders, those events can often lead to toxicity. They may interact with other enzymes and produce side products.
00:08:30
Speaker
So those things can be problematic. And so we're in this narrow window where we're essentially rolling the dice ah each time ah we are you know bring a new drug to market.
00:08:41
Speaker
right So the opportunity is, can we do things that are very hard for ordinary humans to do? Well, Anil, first of all, I'm a chemist by training. So I just want to point out that we do know how to make big molecules. I spent many, many years doing that, but just to start this conversation off on the right foot, but you know I do recognize that there's a lot of um CMC scale-up manufacturing challenges the larger the molecule gets. so I just wanted to
00:09:13
Speaker
save my two cents here as a chemist, but in's in all reality, like I actually, you're reminding me of a story like um what Unnatural Products is developing, in which, you know, you you're supporting them through Humane Ventures. And, you know, i understand that they're kind of building a technology that's not small molecule and not large molecule. So it's, you know, kind of fulfills that niche that you you
Evaluating Tech Bio Startups
00:09:38
Speaker
just described. So when you look at a tech bio,
00:09:42
Speaker
startup, whether it be them or someone else, what are you evaluating in terms of the potential and what what are ultimately immediate red flags for you that makes you lose interest?
00:09:53
Speaker
Well, red flags, let me let me get to that ah later. I'll talk about what excites me at first.
00:10:06
Speaker
One is that, but you know, the that typically really interesting opportunities are where we're able to do something fundamentally new.
00:10:17
Speaker
And that comes not by automating tasks, ah as as is often the case when you think about AI startups, they're taking processes and just you know eliminating labor.
00:10:32
Speaker
And in the tech biospace, there's a very different kind of opportunity where we're exploring spaces that were just simply too hard for us to do at any scale. um And is that's something that UNP is doing really well.
00:10:48
Speaker
They've combined novel chemistry ah with AI ah in parts of their pipeline where they're basically exploring a very, very large chemical space of some 10 to the 50 different combinations with these macrocyclic peptides. So if you imagine a 10-amino acid molecule, but where you have many unnatural amino acids, say 100,000 each, the permutations add up very quickly and you get 10 to the 50, which is a huge number. That's as many atoms as the Earth contains.
00:11:27
Speaker
That's a huge chemical space that arguably even even you know ah a very brilliant chemist like yourself would have trouble um trying to think about all the different binding modalities that that might have to optimize against multiple parameters, like not just the binding, but also the yeah permeability and digestion um and absorption.
00:11:50
Speaker
um as so sorry, solubility and permeability. That's ah that's a problem actually it comes up when you have a larger molecule because it doesn't just have to bind, it also has to solubilize in your gut, it has to get through um the lipid layers, it has to get through your gut, and and then you know the opportunity is that then you could have a very large and very specific molecule that gets inside the cell and binds those targets that we've considered undruggable until now, where we can very precisely potentially interfere with protein-protein interactions or act as molecular glues ah that bring proteins inside the cell and precisely target exactly the right biology, right?
00:12:37
Speaker
rather, say, than going after biology that's on the outside of the cell that's accessible to something like antibodies. you know Antibodies are are great, they're highly specific, but you have to inject them. And so they only float around in your bloodstream, and they can only access the surfaces of cells.
00:12:55
Speaker
So that's that's the opportunity with UNP, very, very exciting company. I think you know people thought ah that this category of macrocyclic peptides was a dog. ah you know My own PhD advisor yeah this was he said this is not exactly mainstream, in Neil. He has a Scottish accent. and um and you know I thought about it quite a bit when we were first investing, but I thought this combination
00:13:30
Speaker
of Enormous parallel screening in a class of molecules that nature had already shown us was very powerful and effective because there are like some 40 macrocyclic peptides that are used therapeutically in almost all of them except maybe a couple are our natural products or just like slight derivatives of natural products. So it's a very interesting space which violate these rules that chemists have, the Lipinski rule of five, which says, don't make anything too big. You're going to have too hydrogens.
00:14:07
Speaker
You're going to have too many hydrogen donors or acceptors. And know what? We've looked at the data and we've always failed. And I think that this was Pfizer's top chemist, Carl Lipinski, in any case,
00:14:19
Speaker
And, um, so there was, there was a clear opportunity for a bat where if you could be much more precise and you could do things that we've never been able to do before, you have a chance at making very, very specific,
00:14:34
Speaker
ah targets that so have a much wider
Investor Expectations and Storytelling for Founders
00:14:38
Speaker
therapeutic window. And you know that opportunity is one that got us very excited because that you know answers to our core thesis of actually being able to change the underlying metrics that are holding the industry back. And that's the huge opportunity.
00:14:55
Speaker
Can you actually, are you just, you know, do you just have a thesis for know solving a disease, which is a great thing, or can you actually address the underlying metrics? And that that's really one of the you know powerful patterns that we look for, that AI is being combined with you know sort of a novel scientific insight that gives access, say, say to a new chemical space that um makes new things possible and solves these really fundamental industry problems. like Imagine that we could go from
00:15:34
Speaker
90% to 80% failure rate, that's a doubling in productivity. that's That's an outstanding outcome. And then there are all kinds of other you know ah opportunities once you do that. um It's not just you're not going to fail in clinical trials. I think these are going to be better products. They're going to be safer products for people.
00:15:57
Speaker
So those those are those are the things I think that I look for on the on the positive side. I think, and this kind of became evident even more so at Bio last week when we were spend a lot of time talking to a lot of different people and that generally speaking, the funding environment has shifted dramatically in the last several years. And no doubt you you know this as well. And I'm now thinking like as investors' expectations have evolved over what founders are they're looking for in founders,
00:16:31
Speaker
I feel like what you just laid out is something that's maybe universal to all investors seeking to invest right now. And that's something that really is a ground shift in how we think about so preparing drugs and putting them into humans.
00:16:47
Speaker
So having this is my long-winded way of asking is, aside from demonstrating really interesting science, what else do you think founders should be thinking about right now when they're when they're preparing to pitch investors?
00:17:01
Speaker
Well, you know one of the most common you know challenges in this space is storytelling.
00:17:12
Speaker
um And this this is kind of a natural consequence of having founders who are very, very profoundly deeply technical um They come out of academia.
00:17:29
Speaker
ah they have They have spent their time you know giving talks in lab meetings, largely. And there's a very different way of storytelling in academia, which is much more discursive.
00:17:44
Speaker
And I think it's really optimized for giving people so many details that you get ah you get helpful responses from the lab.
00:17:55
Speaker
They say, well, did you think of this? Well, did you think of that? And that's that's the whole point of that kind of conversation. It's very winding and You know, but but you know there's there's a lot of detail, unnecessary detail, and it's almost the pleasure in all of that detail that drives academia, you know, just really understanding a domain so thoroughly.
00:18:18
Speaker
And by nature, everybody's coming out. I mean, the the best founders are coming out of those spaces because they have to have very deep technical training in places that you would never get anywhere else. You know, they have to have probably done their PhD.
00:18:35
Speaker
and Tech bio ah companies as well are more like tech than they are biotech. like There's a system that is perfected on the East Coast where venture studios essentially go hunting for insights. I'm thinking of Third Rock.
00:18:53
Speaker
ah third rock um and you know ah other other venture firms like that where they go hunting for biological insights and then they put kind of a well-oiled team together to productionize and spin up a biotech company and they're extremely successful at that but i think that what's missing there is the kind of skunkworks mentality that is very much like the Silicon Valley garage startup, where some people can get together and hack you know a website into existence, and then they've got a business.
00:19:30
Speaker
Well, obviously, you cannot do that in tech bios. So much more de-risking that has to happen. There's so much deep science. You need a lab. You need to be able to ah develop the science before it's anywhere near ah ready for investment.
00:19:47
Speaker
And so that comes out of an academic system with a great deal of complexity. And, you know, having been an entrepreneur and and a technical entrepreneur as well, i know what it's like for founders who are deeply technical because you're basically building and insight from a kernel of an idea and then adding layers and layers and layers to it over time. And there's so so much detail and there's so much complexity, so many different problems that you've solved.
00:20:22
Speaker
And together with the academic storytelling style, you kind of want to explain all of your journey to people. And what's missing is that very, very high-level story, which yeah in Silicon Valley, people are just pitching relentlessly and knowing how to give 37-word, very tight Mad Libs stories about what their startup is doing, the critical problem they solve.
00:20:53
Speaker
So that, I think, is like the biggest piece of advice is, you know and I don't even know what the answer is, to how do founders overcome that challenge, except maybe to work with, you know, work with companies who can work on communication or have the business partner who knows how to put that framework together for articulating the opportunity.
00:21:15
Speaker
What kills me is that I know that there are great startups and there's money from Silicon Valley that should be excited to invest in this because tech bio really comes out of Silicon Valley in many ways that the name comes out of Silicon Valley but there aren't The class of investors who can easily go deep, there aren't too many people who have you know this combination of understanding optics and robotics and drug discovery and you know the biology and the chemistry and how those things are going to intersect technology.
00:21:57
Speaker
and And compute an AI. I mean, that's almost impossible. And then how to think about how all those things fit together and also have tolerance for, you know, a story that's not going to be really sharply put together ah like we see in Silicon Valley, because that's where the real gem opportunities lie.
00:22:19
Speaker
It's in founders who have that kind of depth. Yeah. So i'm not sure if that answered your question. that You're literally speaking my language here because every day I feel like I'm having this battle with someone who's a founder who's so technical and clearly the best person on the planet to be developing a certain technology, yet trying to get them to come up from that and okay now let's talk about how it affects the world and they just can't they can't even think that way it's a real challenge and i think the thing that like motivates me to continue to do it is knowing there's so much value in investing the time in helping that person communicate because that's really how we're going to bring
00:23:11
Speaker
you know, breakthroughs to the front, right, and and help people win is if they really, even if they can't communicate spending the time to go deep with them and and come back up and go, okay, this is how we say it.
00:23:24
Speaker
And I think... And recognizing that you don't have the skills needed to talk about it correctly is step one. And I think if we could help people do that, that would be a huge change.
00:23:37
Speaker
that That way you would be doing God's work. And I i really, I'm like, I sort of say that as a joke, but sort of not. Yeah. like I really think that this is one of the most – I'm not the only person who thinks that this is one of the most important areas to invest in. um you know There are lots of luminaries who've now spoken about this, like Eric Schmidt and Jensen Huang, and and Steve Jobs was saying this is the most important area. you know Shortly before he died, he was sort of looking forward and saying the next –
00:24:09
Speaker
The next century is going to be all about the intersection of biology and technology. And I think, you know, we are largely here. The opportunity is here, but the depth, the scientific depth is are greater than typical Silicon Valley startups. And you've got all of this Silicon Valley money that is looking for theses around and biology.
00:24:38
Speaker
But what does that mean?
Revenue Models and Market Strategies
00:24:39
Speaker
how How do those companies go to market? There are things that sound very technically appealing. How about a SaaS company for...
00:24:50
Speaker
protein AI, LLMs. ah That sounds great to Silicon Valley ears, but to somebody who's been in the industry and who's worked in technology companies, you know certain rules of thumb that If your product is not a drug, you're in the wrong business.
00:25:11
Speaker
And that is such a you know that's such a troubling thing, especially as a technologist and somebody who's always been a technology person. there's this kind of There's a giant sucking sound that happens inside biotech companies when they get to the point of clinical trials, and then the tech teams disappear. Yeah.
00:25:32
Speaker
Because everything goes into the one asset. And that's exactly what happened at Millennium Pharmaceuticals. And so there's this one, there's that challenge that it it it and and part of it, let let me put it another way that when those companies, when companies are trying to serve biotech companies with say a research tool,
00:25:55
Speaker
They're looking at the entire market, maybe naively at the very beginning, like say a Silicon Valley entrepreneur and saying, oh, wow, you know, the biotech market's so huge. I'm going to offer a tool which is, you know, efficient. Then it's a very, very hard thing to build.
00:26:12
Speaker
um It takes tremendous technical insight, but At the end of the day, if you are producing a research tool, you're pulling from research budgets, and you are not going to be able to do royalty deals.
00:26:28
Speaker
That's just the way it is, no matter how brilliant those founders are. I remember when I was like a snot-nosed 20-something-year-old at Millennium Penteculticals, And ah had, you know, people who are just really brilliant, you know, postdocs and PhDs who had started companies and had done, you know from you know, very sophisticated innovations. And they were coming hat in hand to Millennium. And Millennium had, was flush with cash because they had said, they'd put a stake in the ground saying, we are actually going to develop drugs.
00:27:01
Speaker
And because they were a real drug maker, they That was the real reason that they were able to do the historic deals at that time because they were credible to the biotech sorry to the ph the pharma industry that they partnered with.
00:27:18
Speaker
That's where the money is. Ultimately, it's when you have a pill on the market and you are either you know you're wholly owning that asset or you have royalties from it. It's just very, very difficult for the toolmakers to ah develop meaningful royalties from that.
00:27:36
Speaker
Now, the one place where, you know speaking about patterns and anti-patterns, what I look for is if there is this kind of technology, let's say a measurement technology,
00:27:48
Speaker
you ask, will this technology in some way, shape, or form get a turn of the crank with respect to individual consumers or patients?
00:27:59
Speaker
So a Illumina is obviously a very, very successful company because they're able to ah perform these analyses for genomes of cancer patients and all human beings. And you know in principle. So they're effectively reaching into that consumer domain, and they're far outside of just being a research tool, though, obviously, they are, you know, very much used as a research tool.
00:28:28
Speaker
um So that's kind of how I break it down. I think very carefully about, you know, get so excited about these technologies, ah you know, especially Linda has to, you know, yeah pull me back and say, remember.
00:28:42
Speaker
um And and you know we we really think about, well, what's the business model? And I think that some of the time the journey is that these entrepreneurs realize realize this along their journey, and they find a way to start you know producing their own assets or having some stake in the
Consumer Tech in Diagnostics
00:29:05
Speaker
game.
00:29:05
Speaker
And so I would encourage entrepreneurs to think about that perspective sooner rather later. i mean, ultimately, this is a real market problem because those companies should be funded. They're and they're probably important.
00:29:19
Speaker
It's just those those are very long roads and and there are hard struggles because you know Ultimately, there are many research tools that are going to contribute to the outcome of a drug, and you will only be one of many.
00:29:34
Speaker
But once you have the therapeutic, you are selling that, and you know there' is there's unquestioned value in the market if you produce the best therapeutic.
00:29:46
Speaker
Yeah, that that is absolutely true. But I'm also curious, you know, we're talking a lot about therapeutics, and I have so many questions, by the way. Great answer, an Neil. um But i'm I'm curious, just off the cuff, you know, what what do you think about diagnostic companies?
00:30:01
Speaker
and And where does your mind go when you think about ah AI, biology, and and diagnostics, as opposed to just traditional therapeutics? we We tend not to invest in...
00:30:15
Speaker
And companies which directly interact with the healthcare care system. o And, you know, part of this is a sort of ethical viewpoint.
00:30:27
Speaker
Part of this is our incompetence to accept Anything that goes on inside hospitals or you know doctor's offices, the channel getting into that, that is a whole special thing and of itself. and it's it's a like To me, just from my perspective, and i I'm really under-informed on this, but it just seems so horrible because there's no way...
00:30:55
Speaker
ah it's It's very, very difficult from my perspective. I just just really want to I don't want to make any, you know, large claims beyond my area of expertise, but it just seems challenging for how you develop a very widely known brand. It takes so much time to, ah you know, especially if a doctor is administering that diagnostic, you know, typically these are things that are kind of interesting. They're novel.
00:31:21
Speaker
They advance, you know, they advance the nature of practice. And the thing that i always think about is how far behind doctors are from the latest, you know, latest understanding in the clinical literature.
00:31:35
Speaker
And if you take a look at that, you'll find that it's between nine and 17 years with about 14 years behind. Oh, my God. You know, if there's something really cool, i just don't think doctors are going to put it into practice in any sort of timeframe. That's well beyond the timeframe of a fund. Yeah. um So, ah you know, i just I just have that skepticism around that. But I think that there are models, you know, I know people who go for flip style investment models where, you know, there's probably there are companies that obviously have very well built marketing channels, but then like Medtronic.
00:32:13
Speaker
and so You would look at, okay, you develop this, you taught you have partnerships and conversations with people at these diagnostic companies that can distribute, and you pre-wire the deal.
00:32:27
Speaker
That's a completely different style of investing. It's just not our flavor. and Then the other thing, too, that I've been thinking about a lot is... you know Fundamentally, my perspective is that our healthcare system is bankrupting. ah you know It's the cause of the most personal bankruptcies ah of any cause.
00:32:50
Speaker
And there's a fundamental rot at the heart of our healthcare care system, the way it's incentivized, the way it's getting paid. And we don't want to invest into the core of that system. We really think that we need a model where normal market dynamics prevail between the true consumer and ah the the supplier or the the product maker.
00:33:16
Speaker
And that's a normal consumer relationship. So we're we're very excited about consumer tech companies that are providing new diagnostics. Now, we don't normally think about those as diagnostics companies, but But, you know, my my hope is that we are, you know, there's so much there's so much angst from everybody inside the industry, you know, from deans, I've spoken with deans who started new medical schools, health policy analysts, doctors, you know, so many people.
00:33:47
Speaker
are chafing about the misaligned incentives. And I think that there's a great appetite and possibly some new energy to do new things in that space.
00:33:59
Speaker
so But for the time being, you know that's that's a very complicated and fuzzy and long conversation. I don't want to get into that, but we've been thinking about that a little bit. But the opportunity in the consumer space is that we can get to that promised land where we have a much more efficient system that is direct to consumer, direct to the home, um and that's using novel techniques take a look at different kinds of blood measures. There are companies, for example, that are looking at um protein, all proteins in the blood, not just the traditional ones ah that you get at Quest.
00:34:38
Speaker
um There are obviously wearables. So these companies, you know in our view, are going to form sort of the bedrock of a new system that will be focused much more on prevention and longevity.
00:34:53
Speaker
And it will be paid, but has to be paid very differently from, you know, the transactional sick care model that we have right now. So we like
AI Integration in Biotech
00:35:02
Speaker
to invest in that. And so I'm very bullish on quote unquote diagnostics, though I think most people consider them consumer health.
00:35:10
Speaker
applications that are really moving the needle in interesting ways. And then we can surpass, you know, by far in terms of convenience, efficiency, cost, accuracy, what the medical system is providing. So that's kind of how I look at that part of the space.
00:35:26
Speaker
yeah Yes, yes. So, Anil, you know, i I think we've heard you use the phrase wet AI, of course, being a fusion of data-driven tools and and lab-based biology. and And we know that those two have to go together to prove out, you know, the points that that we find in our using data-driven tools.
00:35:46
Speaker
So I'm curious, how do you actually see that integration shaping the next ah wave of breakthroughs?
00:35:56
Speaker
Yeah, i wrote I wrote a paper right when we started the fund, sort of thinking about you know what what what was different about AI and biology, and what what would it look like? what was the most What were these really important and interesting applications?
00:36:18
Speaker
I think... A lot of people are familiar with what DeepMind has produced, and you know those are incredible research tools. They're pure computer scientific tour de force applications, and they're going to be broadly useful across research and um and in and in therapeutic development to understand structure.
00:36:42
Speaker
But, you know, we're turning to that model of saying, you know, if you're not building a drug, you're kind of not doing the right thing, or that's where the more opportunities are.
00:36:53
Speaker
It's really about building new types of products. yeah And building those new types of products requires... you know having AI that is focusing on a specific domain or compute that's focusing on a specific domain. Because ultimately, you're not producing a million different types of products across different categories.
00:37:17
Speaker
One company will specialize on a particular type of drug modality. Now, the thing about wet AI that's very different from commercial AI or, sorry, consumer and enterprise AI like the LLMs is that typically you just don't have the data.
00:37:35
Speaker
You don't have an internet worth of data to draw from, to draw your insights. it's a big problem. but It's a big problem. like Now, PDB, protein structure databases have gotten us ah you know yeah a good long distance in the hands of DeepMind.
00:37:52
Speaker
But for these other problems, which are very tricky binding problems where you've got proteins that are changing shape, they may have cryptic pockets which kind of open up and close.
00:38:04
Speaker
ah we don't really We don't have the data, first of all, and we don't have ah you know the rational drug design techniques to explore those pockets and explore those binding modalities.
00:38:16
Speaker
So how do you get to them? And this is where when when AI, you know in biology and and with AI in general, it's all about data. And so the way I look at it is there's usually a combination of scientific techniques that enables you to sample um properties of, say, your chemical space or your protein space at scale within the domain that you're interested in And there are a broad set of therapeutic applications with this modality that's almost...
00:38:50
Speaker
certainly to be true because you're you're you're playing with molecules, whether it's RNA or protein or small molecules or, or or you know, macrocyclic peptides or, you know, whatever modality it is.
00:39:03
Speaker
um So you've you've got a space of a very general kind of tool. And now the question is, how do you optimize those things? And that you get by running experiments at scale.
00:39:14
Speaker
And the way that differs from... differs from you know, enterprise and consumer AI typically most of the data is going to be proprietary. And then the other thing is that those models are actually going to be relatively small, like but vastly smaller actually than what we're used to in the the consumer and enterprise space.
00:39:35
Speaker
ah Because they're very bespoke small models. They're very focused on the problem and they're really unlocking the superpower of AI, which is dealing with complex data, which we have in spades in biology.
00:39:50
Speaker
Now, the thing is, we we just, we can't, rat we I shouldn't say we can't, but it's still difficult for us to rationally design drugs, especially complex ones, because there's complex quantum chemistry, there are so many different parameters and things that we don't know that are very hard to model. so We just go and look, and we we don't care about the detailed structure, um you know, except to confirm things. What we care about are, you know, is this complex chemical space getting into the right, you know, the right the right part of the space where we've got ideal binding and ideal solubility and low toxicity?
00:40:30
Speaker
And that's a problem you answer with your own data set. So I get the sense that we're not talking about using what's called consumer-facing LLMs. This sounds much more like a natural development um at maybe each company.
00:40:48
Speaker
Is that right? Yeah, that that is that is right. And that is the pattern that we see. And know these companies also have a choice. Do they become kind of like a CRO and yeah offer that capability? Yeah.
00:41:01
Speaker
say we can design these drugs for you i would urge these companies not to do that i would say that they should go and try to develop those assets themselves and actually you know tech bio companies start out as technology companies yeah And then they do partnerships with multiple, you know, the the good ones will book partnerships with multiple different pharma companies, exploring different spaces, and then building up their data set and getting revenue much earlier than biotech companies.
00:41:37
Speaker
um So biotech companies will be much more dependent on just funding their research, whereas tech biocompanies will be using partnerships, will they'll lean more, I should say, on partnerships to fund their continuing development, fund their learning.
00:41:54
Speaker
And then along the way, they are going to start learning which assets they could go after in their own internal programs that then will be wholly owned.
00:42:04
Speaker
So there are multiple different revenue streams, in other words, for a tech bio companies, which is partnership ah ah partnerships where they get milestone payments, you know upfront payments, and then payments as they ah proceed down the pipeline and validate the candidate.
00:42:21
Speaker
They will receive royalty payments and typically, you know, their ability to argue for a larger royalty stake will increase ah as, you know, there's more confidence in the platform. But the most exciting part is when those companies can hold back their assets and, you know,
00:42:40
Speaker
you know develop them as wholly owned assets. And that that just seems you know inconceivable. Yeah. yeah That's right. For biotech companies, but this is possible because of the efficiencies of the platform.
00:43:00
Speaker
And you know as a technologist, that's what I love investing in because I think that this is where the real solutions are going to come from. it'ss It's going to come from those those breakthrough capabilities, and those you know some of these tech bio companies could emerge as the new Biogens, the new Genentex that you know discover entirely new ways of doing the drug discovery pipeline.
00:43:27
Speaker
so Yeah, that's the hope. And Anil, you used a word that I've never heard before in biotech, and that was revenue. So that was interesting to hear. I know, it's so weird, right? Well, what the hell is that? so but but but But let's get into it. I mean, you know we know many founders, and not just in biotech, but also in tech biotech, talk about platform technologies, but but really very few of them...
00:43:53
Speaker
you know, know how to monetize them or are able to monetize them. So so what do you think a good platform strategy looks like from an investor lens? Well, I mean, it's also a matter of when these companies are ready to come to market. You know, some of them, ah these are these are not easy technologies to build.
00:44:14
Speaker
And there is typically a lot of de-risking and the de-risking is pretty variable. Sometimes companies come out of a lab and they have almost everything they need to, and they may even have some assets coming out of the lab, ah which are ready to you know form a basis for their next round or even going into clinical trials. And they're a little bit like a biotech company in in that regard.
00:44:47
Speaker
Sometimes you and the investment has to go in in proving out the underlying platform. There are partnerships with forward-thinking biopharmists like ah Bridge Bio um that are exploring new modalities.
00:45:05
Speaker
and so getting Getting partnerships with them or with you know the the well-known top 10 pharmas, they are almost acting like ah venture firms themselves, yeah taking bets on small biotech companies and increasingly tech biocompanies.
00:45:22
Speaker
um those Those are the important proof points, I think, ah getting out there. And you know it's so funny because you know tech bio also resembles tech in that way because people actually think about revenue, um which is kind of amazing.
00:45:38
Speaker
But it also means the teams are smaller, like the structure, the kind of founders that you have are more like tech founders because they're you know one or two individuals who have a deep understanding of the whole thing.
00:45:52
Speaker
whole you know new concept that they're bringing to bear? And how could it be any other way? Because you're playing with so many different variables in how you're doing drug discovery.
00:46:04
Speaker
It's not the kind of thing that a team that already understands all of the different ah parts of the pharma pipeline ah can get on board with because you'd have to explain to everybody all of these different ways of doing things.
00:46:22
Speaker
And so that's that's why you know Silicon Valley produced Uber. They didn't have a bunch of Anderson consultants go and talk to the taxi industry and say, well, help you make new dispatching software, right? It's just a small group of people. It's a tribe of barbarians who are saying, no, this is this is a completely new way of doing things and they understand it and they can execute on it. So the teams are smaller.
00:46:50
Speaker
They can make revenue like sort of like Silicon, not exactly like Silicon Valley companies because they don't have a SaaS model. Yeah. But they do do partnerships. And so that that funds those companies as they get to that later stage. And then at that point, you know that's where when I expect, you once they have their candidates, that they are starting to resemble more and more a traditional biotech. But my hope is...
00:47:18
Speaker
that the you know traditional biotech investors are also going to recognize the important extra value of the technology platform. And I think that by and large, that's true. The exciting thing will be when we start to understand that these new technology platforms are not just efficiencies, but um But they're actually fundamentally changing the metrics. And then once that insight starts to percolate, once we actually deliver on that, and we will need multiple hits on goal, and it will look something like, you know,
00:47:53
Speaker
three, you know, there's there's a lower than 90% failure rate. And at first, you know, you'll have three, those could be a fluke. But over time, we will hopefully start to see that we have a much higher success rate with these products getting through clinical trials, showing lower talks.
00:48:15
Speaker
And that is going to change the valuation ah structurally in these companies. And we won't be doing what you know, traditional biotech investors do, which is apply the standard ah NPV calculation over somebody's pipeline.
Ethical Considerations in Biotech
00:48:31
Speaker
It's much more than the pipeline. There's an additional value for solving these fundamental problems.
00:48:40
Speaker
So I want to go back to a fundamental problem that sort of touched on, which is the healthcare system. And i think you and I are could spend two or three hours or more just debating the problems there. But um I'm going somewhere with this question that's important. And that is that there are many different considerations for companies when they're thinking about rebuilding an infrastructure for drug development, whether they be social or ethical. And given this is an issue I think that's close to your heart, Anil, I'd love to hear your thoughts on what responsibilities do these tech buyer leaders that we've just been describing have to the whole industry beyond returns?
00:49:29
Speaker
this is such ah This is such a good and important question. You know, i'm i think about this company that to some friends of mine invested in called Devoted Health.
00:49:47
Speaker
Mm-hmm. They made a pledge in that company to treat all of their customers like they were family.
00:50:00
Speaker
wo And they they they decided that they would not do things for the purposes of just making profit. to Now, they are an enormously successful company because people love devoted health.
00:50:22
Speaker
But what what they did was they made a purposeful commitment to do the right thing. i i I feel like we are in a place where more companies and VCs should think about how we structure our businesses and our purposes so that we are really thinking about that longer-term good for humanity.
00:50:48
Speaker
There's so much of our industry which is you know because of the nature of needing to make capital and produce returns
00:51:00
Speaker
We focus only on the money-making. And I'm you reminded of what um the Merck CEO said when he was he was actually questioned about ah in front of the Senate, I think, and it was Bernie Sanders who asked him about the high cost of drugs, ah what unnatural products is building. And he said, you know this is the future of drug innovation and drug innovation is going to lower the cost of drugs.
00:51:28
Speaker
And that's actually probably not true. um Because that that that's what studies show, is that the costs just go up. um They still keep going up.
00:51:39
Speaker
ah you know i don't have a structural solution for the incentive structure. I think that is solved on the insurance side of the world. But I do think that founders should think seriously about that greater purpose of being a great company that does great good.
00:52:00
Speaker
And that you know that resonates tremendously. I think You know, to put it in terms of self-interest, which I don't think that this is what it's about, but I do think that there are all these beneficial impacts of having...
00:52:16
Speaker
A really luminary, far-reaching vision that is also very purposeful is that it inspires everybody who works at that company to do their very best and have the deepest insights and have a deep calling in life, which is going to... you know that's ah that's something that's very fuzzy and gooey and very far away from the typical concerns of somebody who's been trained in science and working in biotech and doing business deals with ah you know people who are in middle management and pharma.
00:52:50
Speaker
That's just not part of the language. But that i I firmly believe that that sort of perspective is very important for our world.
00:53:01
Speaker
And when we start thinking like that, ah we have ideas about how to fix things. I've come across companies that have had their business models focused around how they can drive down the cost of drugs.
00:53:16
Speaker
And I think you know with these increased efficiencies, I think that there's an opportunity for this special class of founders to make not just lots of money, but a very positive impact on the world and maybe even change the public's view of this sector directly.
00:53:37
Speaker
ah from one that is like rapaciously taking as much as it can and hiking up insulin prices to one that is really making ah grand commitment to the world and you know in helping more people. I know how naive all of this sounds, believe me.
00:53:55
Speaker
But I do but think that there's there's a possibility in that because of this increased efficiency, And then we can get there and figure it out. It's almost like you want to build a business model that you can eventually ah invalidate.
00:54:10
Speaker
And that's you know a truly great and purposeful company. You know what, this is, don't know this is appropriate to say, but I'm going to say it anyway. Like this concept feels foreign in the US for some reason.
00:54:26
Speaker
But, you know, when I go to Europe and talk to European companies, that's almost like the first thing they want to tell you is like how they want to make lives better for patients or something to that effect. Like, do you think there's an element here of a cultural difference in the US? Yeah.
00:54:43
Speaker
It's been a long time since I've been in Europe. I went to school there, and I've forgotten the vibes. I only know the vibes here in the U.S., and especially on the coasts.
00:54:57
Speaker
And the coasts are very materially minded. And I would say that in terms of materialist science and rationality, efficient market hypothesis, all plays into a default set of assumptions that in how our world is arranged and should be arranged.
00:55:21
Speaker
And, you know, at the same time, I think, you know, we have been marinating in that for a very long time, for the last 40, 50 years, and it's showing its wear. And I know, i know for a fact, especially here in Austin, but I've been hearing it um on the coast as well, that people are yearning for ah more purposeful world, right?
00:55:43
Speaker
I know that there are many efforts, and I know also that there is increasingly capital that is seeking ways ah to invest in companies that think in that new way.
Conscious Capitalism in Biotech
00:55:56
Speaker
So maybe maybe there's a change on the horizon, though you know mechanistically, ah it it is it is a hard thing to articulate because ultimately, ah capital's job is to deliver returns, and that is all about a single calculation.
00:56:17
Speaker
don't know. I don't know. i and Then again, you know look, there there are other people who have argued otherwise. you know John Mackey, ah the founder of Whole Foods, lives just down the street from us, and he's a friend.
00:56:30
Speaker
drove... ah you know he drove but has he's He's the founder of Conscious capital Capitalism, and which is really surprising when you hear John speaking. He sounds like very fire-and-brimstone libertarian, but he is also very, very purposefully—I know him, and he is very purposeful and committed person ah who— If you read Conscious Capitalism, who really talks about introducing conscious decisions into all of his business processes and not just asking about the bottom line, but asking about how one does good.
00:57:14
Speaker
He charts those effects into all of the additional value that accretes into the brand. And he argues that, in fact, that actually does deliver superior returns.
00:57:28
Speaker
And I think that that is the thesis of the folks who work on conscious capitalism. So you know maybe maybe maybe the answer is as simple as that.
Conclusion and Future Exploration
00:57:38
Speaker
It's rational, do the right thing, do good, and you will grow.
00:57:43
Speaker
i love that so much. Thanks, Anil. Yeah, I love that, Anil. And I love that you shared it. So especially thanks for for sharing that. But also, everything you've shared with us today. It's been such a delightful conversation.
00:57:57
Speaker
And thanks so much for joining us. Thank you so much. Thank you for having me. It was a really fun chatting with you. Well, thanks again to our listeners for joining Sparktime. We welcome you to join next time as we continue to explore the ideas, the thinkers, and the innovations that continue to drive biotech forward.
00:58:15
Speaker
We hope to see you there.