Sean McClain and Amy Webb at Built with Biology 2022

Absci founder & CEO Sean McClain sat down with Amy Webb CEO of the Future Today Institute at the recent Built With Biology Conference 2022. They talked about the future of AI in drug discovery, Absci’s partnership with NVIDIA, and more.

Interview Transcript:

Amy Webb:
Hello. Hello.

Sean McClain:
Amy, it’s great to be here with you.

Amy Webb:
Good to be here too with you. So, I’m the CEO of the Future Today Institute and we develop original research and deliver long term trends analysis and scenarios and decision frameworks for business and government around the world. And while we focus on all areas of technology, one area that’s been particularly interesting to me. And certainly to everybody that we work with is artificial intelligence and lately both artificial intelligence and this weird new intersection where you play, which is biology. And I think one of the most exciting applications of all of this is new drug discovery. So that’s actually why we’re here. And I think one of the cool things you’re doing, Sean is creating… I mean, to me, you’re setting the seeds for a future in which we approach drug discovery in a totally different way. So, explain to us as though you were talking to my 11 year old daughter, what is Absci and what are you actually building and doing?

Sean McClain:
Yeah, definitely and maybe before we dive into AI drug discovery and what Absci is doing, I just want to say congratulations on your latest book, Amy.

Amy Webb:
Thank you.

Sean McClain:
The Genesis Machine and I’ve definitely been listening to that when I’ve been working out. And again, congratulations on that. And I think it’s providing a great perspective at a high level as to what AI and synthetic biology is doing and so it’s great to have that perspective out there for all the folks here, as well as investors analysts, which is super important for industry. So congrats.

Amy Webb:
Thank you. I appreciate all that. Now let’s talk to all of the investors and analyst 11 year olds and to my lover. What is it, so explain it the way that you would explain it to an 11 year old.

Sean McClain:
Yeah, definitely. So what we’re doing here at Absci is really merging AI and biology together to ultimately get better drugs to patients and ultimately increase the overall probability of success throughout the clinic. Currently it’s about a 4% success rate.

Amy Webb:
Why it so low? That seems really bad. What’s going on?

Sean McClain:
Well, let’s just take a look at an antibody sequence, a protein based drug, there’s more sequence variance or possible combinations of drug candidates that you could create in an antibody than there are atoms in the universe, and that is the reason why it’s a 4% success rate because we’re not searching in the right search space. And when you take biological data and you merge that with AI, you’re able to look at the whole search space to ultimately hone in what is right for that particular indication or patient, which is ultimately going to allow us to go from a 4% success rate to a 10, to a 50 and ultimately personalized medicine from there.

Amy Webb:
Okay. So AI is kind of an umbrella term that means a bunch of different things to different people. So let’s break this down a little bit. What exactly are you? What exactly is going on over there in Washington state? What are you guys doing?

Sean McClain:
Yeah, definitely. So we’ve spent the last 10 years developing what we call our syn bio platform and it takes a very simple organism, E-coli, and we pair that up with our screening platform. And what we’re able to do is build these large billion member libraries and screen them withinside the cell. And we’re able to look at the functionality of proteins or protein-protein interactions, along with the tighter the manufacturer ability. And in a given week, we’re able to screen about 10 billion different antibody drug candidates. And that’s really compared to maybe screening tens of thousands that there’s currently best in the industry and that data throughput and the quality gives us the ability to actually leverage deep neural networks and allow us to not only look at 10 billion drug candidates, but look at the whole universe and then ultimately hone in on what is the right universe to be looking at for that particular indication or target.

Amy Webb:
Right? So, that sounds to me like you’re using some type of transformer model. I mean, does NLP… Let’s use fewer buzzwords, how about it. Otherwise, everybody’s going to be like… So, natural language processing plays a role in this is what I’m hearing from you in some way?

Sean McClain:
Exactly, it’s NLP. So just like NLPs are used with language taking, let’s say English and saying hello in English and taking that phrase and turning it into French, but obviously it’s way more complicated than that. And what we’re doing is learning the language of proteins. With the data that we have, it’s all about what is the language of a protein? How does a protein fold and then how does a protein ultimately then bind to the target of interest? And so these NLP models that have been developed at tech companies like Google, we can take that and apply that to biology.

Amy Webb:
Right. So you just said protein fold and Google in the same sentence, so I got to ask are you-

Sean McClain:
AlphaFold.

Amy Webb:
Right, so, what’s up, man, are you building an AlphaFold competitor or what is this? How are you different?

Sean McClain:
Yeah. So what AplhaFold did was they were able to take a sequence of a, of a protein and be able to predict the structure. Structure is really important. I would say from an academic perspective, this is a huge breakthrough, but when it comes to drug discovery, protein-protein interaction is one of the most important aspects. And even in the AlphaFold paper, they talk about not having enough data on protein-protein interactions to be able to predict what affinity a antibody would bind to the target and so we’re essentially just looking at the other side of the coin.

Amy Webb:
Right. So this is the thing that.. So, yes, that’s all very interesting and as much as that blows my mind, the thing that really blows my mind is a couple of weeks ago, there was like a huge, huge thing that happened that got very little coverage. So your interesting, smallish biotech startup is working with Nvidia, right?

Sean McClain:
Yeah. That’s correct.

Amy Webb:
Well, why don’t you… We can talk about this, yes?

Sean McClain:
Yes we can.

Amy Webb:
Otherwise I’m going to feel real bad that I just brought it up.

Sean McClain:
Yeah.

Amy Webb:
Okay. All right, good.

Sean McClain:
Yes.

Amy Webb:
Good, we can all chat about it. But this was a huge big deal and it makes me… The question is, why didn’t I hear analysts talking about this? Why wasn’t this in the media? And I think that there’s just this,… First of all, I don’t think people understand why this was important and I think there’s a fundamental gap, there’s like this Delta between what’s happening in AI and what’s happening in biology. So you should explain what this thing is and then we can talk about… we can both be upset about nobody talking about it, but why don’t you explain what it is first?

Sean McClain:
Yeah, definitely. I would say I was definitely bummed on the coverage that was picked up, but you have to realize though, when you’re merging two industries-

Amy Webb:
Wait, wait, wait, what is Nvidia? So, explain… Yeah. So just like… This is not a pure AI crowd, so what is this company and why would an AI company want to talk to you?

Sean McClain:
Yeah. So what Nvidia sells to customers is GPUs and what that allows you to do is run your models and be able to ultimately scale your models. The more compute you put around certain models, the more accurate your models get. And so Nvidia is absolutely key to success within AI for healthcare and all the other industries that are out there.

Amy Webb:
For context, so this is a best in… from my point of view, this is like a best in class computer vision company that’s mostly… Their CEO is kind of known for having held a press conference in the metaverse. I know metaverse, but people talked about it that video had two million views, I think, within the first hour, which is significant. So why is a company that is in the gaming metaverse web3 cartoon avatar space, what do they care about health? I mean, I know what the answer is, but I think we should tell them.

Sean McClain:
Yeah. So Nvidia is looking to partner with industry leaders that are going to activate an ecosystem, and they see that healthcare is the next industry that’s ultimately going to get activated by AI. We’ve seen AI transform almost every other industry, and it’s made huge advancements on society, it made great returns for investors and I think investors and us from a technological perspective know that AI within healthcare is the future and it hasn’t been tapped at at all. And Nvidia saw what we were able to do and what we were able to… How we were able to merge AI and biology together to have some of these huge breakthroughs. And they’re like, “We want to partner with Absci to help ensure that we’re developing the GPUs and the software that are ultimately going to scale within healthcare and within drug discovery.” And so that partnership together really signifies for us as an industry, that this is the future. This is where it’s headed. Nvidia is making a bet in this space and it’s really exciting to see.

Amy Webb:
Yeah, I think this is the piece that from my vantage point, analysts get wrong and investors get wrong a lot. Everybody’s so obsessed with when are the robots going to come and take all of our jobs? And then when are the robots going to come and murder us in our sleep? That there’s an entire storyline that’s missing. And when the community talks about automation, so automation as it relates to work, automation to speed, to automatically do pattern recognition, to do the types of computations that even the smartest collection of humans just don’t have the mental compute, the biological compute to do on their own. This is obviously an enormous frontier that gets us closer to therapeutics and other things that we need. But I think that there’s this… Everybody’s so specialized that nobody can see that connective tissue, I guess, in between. I mean, we were talking about that earlier, right?

Sean McClain:
Yeah, no, totally. And I would say the investors that got most excited about this partnership and the data that we had presented and really showing this huge technological breakthrough that we had, it was the tech investors. They’re like, “This is absolutely incredible.” They’re thinking this should have occurred five years from now. And I think that we need as an industry look at tech and biotech and start to understand both perspectives here, because they’re really important. And we need tech investors to understand biotech and we need biotech investors to start understanding tech better and be highly collaborative. Again, when you’re merging two industries together, it’s not one is right or one is wrong. It’s how do we be collaborative and help tech investors better understand biotech and then the biotech investors understand the tech and the huge breakthroughs that we’re currently having and I think that’s what’s really needed here.

Amy Webb:
So since we’re on investing for a hot minute, let me press you a little bit more. So Altos Labs got a $3 billion round of funding and I don’t even know what the investment… I don’t know what the valuation is. $3 billion worth of investment without any products, that to me is a little concerning because when I think back to the 1980s and everybody was very excited about artificial intelligence and there were enormous promises that were made that failed to materialize, there was a winter, the AI winter of the ’80s and then it took a very long time for things to bounce back. So from your… You’re kind of stuck, because you’re trying to convince people, “Hey, this is real.” And I, by the way, think it’s real. But at the same time, there seems to be a lot of money just being thrown at everything. And I worry a little bit about a looming syn bio winter. And I don’t know if you feel… you probably don’t want to say if you feel that way in front of everybody but tell us, we want to know. Do you feel that way?

Sean McClain:
What scares me is you look at the AI drug discovery companies that currently have drugs in the clinic and I think the work that they’re doing is really extraordinary and it’s on the small molecule side. You have Accentia, you have Recursion, you’ve got a Schrodinger. And my fear is that if they have a small molecule drug that ends up failing that was discovered from AI, that that’s going to have a huge setback for AI within drug discovery, but what investors and everyone have to realize is that we’re in their early innings and what we’ve been able to demonstrate is we can take an antibody sequence and predict the affinity to a target. And not only that, we can use these models to actually generate an antibody that has the affinity that we want to target. This is what we are wanting investors and analysts to really focus in on is the technological advancements.

Sean McClain:
I mean, this is a huge advancement within the space and if we’re able to do this, think about all the other extraordinary things we’re going to do. And yes, affinity doesn’t correlate to success in the clinic but just being able to demonstrate that you can do this is truly incredible and we’re going to continue to chip away at it. But we are in the very, very early innings with very huge proof of concepts being shown. And to me, that’s really exciting. We’re going to have failures, but that’s okay. But it’s looking the technical success that’s being seen.

Amy Webb:
So I want to unpack a few things because I think they’re worth just noting because, and Kathy Wood also mentioned this this morning, we’re the very beginning of a multi decade long trajectory. This is a deep, deep time horizon and I agree with you. The worst thing that can happen is we get outsized expectations for what should be happening as though you can schedule an R and D breakthrough the way that you schedule your quarterly offsite or something like that. And then investors get spooked or there is a problem or a failure at some company in the space and then that’s it. This is long-haul, groundbreaking, life altering technology that will pay off at some point. But I don’t know when and if anybody could build that kind of a model, it would be me, but I can’t tell anybody when that happens and so I would even not say early innings for all of this. I mean, I would say that the guys are on the field pitching the ball and we haven’t even started the baseball game.

Amy Webb:
Should we keep with the… We’ll stop the baseball metaphor and move on to data? Where are you getting all this data? So you need a lot of it and I’d like to know where it comes from.

Sean McClain:
Yeah. So we were not an AI first company. We were actually a syn bio first company and we spent the last 10 years developing syn bio technology that really became our data engine. It’s our SoluPro E-coli strain paired up with our screening acids that allow us to, in a single experiment, screen over 10 billion, different drug candidates looking at their affinity or functionality or tighter in their manufacture ability. And that gives us the ability to actually leverage deep learning and actually have these huge breakthroughs that we’ve had. But it all goes back to the data and we’re recruiting top talent from Google, Facebook, you name it. They’re coming here and they’re like, “We came here because you have the data.” I can sit there-

Amy Webb:
But where’s it… Sorry. Where are the data coming from? Not the humans using the data, where does it come from?

Sean McClain:
Yeah, so it comes from our wet lab. It’s literally… We have the technology at our campus in Vancouver, Washington, a non-biotech hub, which is an amazing place to be. And just to show you that any syn bio or AI company can be built anywhere, but it’s all built in our labs in Vancouver, Washington, it’s the technology and that’s why we’re getting all of this talent to come work here at Absci because they can go make a boat load of money at large tech companies and build these models, but they can’t test them. They can’t be like, “My model is correct,” and they don’t also have the data to also create the next model that’s going to be used for drug discovery and within biology. And so they had been rapidly iterated on these models and get the data where they could, they couldn’t get it in any other place.

Amy Webb:
So the problem in the field, one of the many problems, in the field of AI is data sourcing, the corpus, knowing what data got put in accountability, traceability and everything else. So, this is not a question about your data, but more sort of a big picture question. Do you think algorithms used in the space need their own clinical trials of some kind? That would be a regulatory nightmare, but still.

Sean McClain:
It would. So, I believe what’s going to end up happening is that the models become more and more predictive of efficacy. And we’re going to start to see that 4% success rate go to 10% and then ultimately 20, 50% and once that occurs, you’re going to be able to start to go to the FDA and start to lobby and be like, “we need to change how we do FDA clinical trial and design,” because for every one in two drugs we discover we’re able to have success over that or we’re going to have success throughout the clinic. And that’s what’s going to enable personalized medicine. And that’s truly the future. And think about this, you have personalized medicine that, you have personalized that for every single drug that goes through a 50% success rate, that’s going to completely change the paradigm of health insurance as well and so the whole industry’s going to be flipped on its head once you start to see these huge advancements start to progress.

Amy Webb:
Yes. So, that is going to scare a lot of people. So is your message, “Hey, now is a time for us to think through and evolve our business models, our operating models,” is that what you’re saying?

Sean McClain:
A hundred percent, and this is where we have to come together as a community, the syn bio talent, AI talent, tech, investors, biotech investors, and really start to understand what does this mean? If we are successful, it’s going to mean that we’re going to create all this value. How do we capture this value and how do we set ourselves up for success? Because at the end of the day, it’s truly amazing. Why I get up every day is because this is going to change patients’ lives. This is going to get better drugs to patients and ultimately allow personalized medicine and what better thing. But we do have to start talking about this now. It’s super important.

Amy Webb:
I think that’s right. All right. We’re out of time. So I’m going to ask you one last question and it’s a question I ask everybody I meet. What’s the last great thing you’ve read so that I can read it?

Sean McClain:
Well, obviously your book.

Amy Webb:
Obviously my book, but aside from the book, which all of you should buy and read, it’s called The Genesis Machine and appreciate that but that’s actually not why. I do ask this question of everybody. So, what’s something that you just read that we should all go read?

Sean McClain:
So, since I’m currently reading your book, I’m going to go back to one of the books that I always tell entrepreneurs to read. And I think, again, there’s a lot of exciting companies that are being generated in this space and a lot of new entrepreneurs is The Hard Things About Hard Things and it’s a journey about how rough it is to be an entrepreneur and how you have to constantly continue to believe in the impossible and work through hard things that occur every single day. And I read that book. It was hugely transformational for me. It was Ben Horowitz that wrote it and that’s, I’d say one of my top five books that I’ve read as an entrepreneur personally.

Amy Webb:
Sounds good. Sean, thank you.

Sean McClain:
Hey, thank you so much, Amy. It’s been great chatting about AI and biology and how we’re going to change the-

Amy Webb:
The future.

Sean McClain:
… world here.

Amy Webb:
Yeah. Thanks everybody.

Sean McClain:
Yeah.

Related