🔬 The Coolest Diffusion Research Isn't in LLMs — Evan Feinberg & Sergey Edunov, Genesis Molecular AI

This episode has a fun personal twist: There’s a counterfactual world where I was employee #1 at Genesis Molecular AI https://www.genesis.ml/, the company behind today’s episode. A certain introduction happened a few weeks too late and I had already happily signed at Atomwise, another ML-for-drug-discovery startup. Same problem, different company. I was certain ML was going to transform small molecule drug discovery. Early results were underwhelming. Useful at times, but nowhere near revolutionary. In the last year I’ve seen signs that ML is finally ready to deliver on my convictions from a decade ago. Genesis is one of the places that might have finally cracked this problem. I was super excited to come full circle and catch up with co-founder Evan Feinberg https://www.linkedin.com/in/evanfeinberg and CTO Sergey Edunov https://www.linkedin.com/in/edunov/. If you are at all interested in small molecule drug discovery, we think you will find this fascinating! In our nearly two hour chat we cover: * What is small molecule drug discovery, and why is it hard * Structure prediction as a hotbed of innovation in AI algorithms * How advances in AI elsewhere have enabled stepwise improvements in predictive power * How the community benchmarks are essentially calling AI slop good enough * The Genesis flagship model (PEARL) can routinely hit a threshold that is necessary for real-world applications * New agentic workflows enabled by these highly accurate models Read on for more, and also some personal thoughts on the future at the end. The coolest diffusion research is happening at Genesis Sergey Edunov came to Genesis from Meta where he led Llama 2 training and Llama 3 pretraining. Sergey was a former physicist who thought he was done with physics after many years of training LLMs. Then, he discovered Genesis, and was blown away with all the novel architecture work they’ve been developing. It probably surprises no one that modern LLM research has not resulted in fundamentally novel or exciting updates in architectures since almost the advent of the transformer — the entire field is using variants on the same idea that came out in the original “Attention is all you need” paper. Sure, some were quite useful (mixture-of-experts in particular allowed for the massive model paradigm we’re at today), but there was very little conceptually exciting. “We sort of had to wait for the right primitive to get created, and that turned out to be diffusion… Actually, some of the most innovative diffusion research that’s happening in our field is happening in 3D structure prediction right now.” — Evan Feinberg The field of 3D structure prediction on the other hand has been a hotbed of research. Genesis’ recent model PEARL https://www.genesis.ml/news/introducing-pearl (Place Every Atom at the Right Location) is able to understand protein flexibility, and model not just where the ligand goes, but also make small adjustments of the protein so that the two fit better than either alone. The field knew this was missing for a long time, but it was really hard to model until now. Agentic Discovery What makes this problem so hard? As Sergey points out, there are 10^60 possible drug-like small molecules. You’ll never be able to search them all, and trying to find the good ones is something like finding a needle in a haystack — except everything except your needle is dangerous. “There are 10 to the 60 drug-like small molecules in the universe… it’s like finding a needle in a haystack, where everything except your needle is very, very dangerous.” — Sergey Edunov “Or finding hay in a needle stack might be a more apt analogy.” — Evan Feinberg Trying to solve the multi-parameter optimization problem is even worse. What makes a strong binder and a molecule with good “ADMET Properties” are oftentimes at tension with each other. For example, a good binder is likely greasy, but a greasy molecule is likely insoluble so it won’t enter the bloodstream and get to where it needs to go! Genesis’ advances in generative AI have now pushed them beyond the threshold where they believe agentic drug discovery loops are finally possible. We all remember the early days of LLMs. They were great chatbots but terrible agents, as small errors compounded rapidly into uselessness. As LLMs got better, the usefulness of agents rapidly improved. Evan and Sergey argue that their models at Genesis recently passed a similar threshold. Their internal agentic drug-discovery system (code named SAPPHIRE) can now iterate like a chemist: look at and reason about poses, form hypotheses, read literature, use internal tools, create candidates for the next iteration. Combining this with automated lab partnerships like the one Genesis has with Incyte https://incyte.com/, we’re rapidly approaching a time of drug discovery agents running 24/7 making/testing new molecules. Exciting times! Benchmark crisis: Everyone’s favorite benchmark is slop One surprising point that isn’t talked enough about: the academic field of “co-folding” has settled on a benchmark value of “2 Angstrom RMSD” as a metric for a “good pose”. Evan does not mince words: this threshold is just bad. Perhaps even deceptively bad. For many strong binders, there’s a very clear pose, one that you can even directly resolve in the PDB electron density! And yet, with a 2Å RMSD threshold, you can get the pose quite wrong in ways that might even mislead a medicinal chemist. For example, flip around an aromatic ring, and everything looks reasonable, but you’re no longer modeling the right interactions. Evan makes the strong claim that 1Å RMSD is really the threshold necessary to ensure the core of the molecule is sitting where it needs to be, and models all interactions. “If your model is sitting at 1.8, 1.9 Angstrom RMSD, that’s slop, most likely.” — Evan Feinberg As a simple example, he points out hydrogen bonds which are responsible for many of the most important interactions in protein-ligand systems. Hydrogen bonds only have a 0.6Å range to be valid! Clearly if you’re accurately resolving all H-bonds, you generally have to be doing much better than the 2Å threshold. This is clearly a hard-fought lesson for Evan and Genesis. In their opinion, the community is stuck on these benchmarks because academics developing methods were not users. Evan does see signs of life, with the use of new metrics such as lDDT for co-folding. Hopefully soon the community can agree that “1.8Å RMSD is slop”, and start hill climbing on this much harder task. For a more thorough exploration of the weaknesses in conventional benchmarks, see the PEARL technical report https://arxiv.org/abs/2510.24670. PEARL tops OpenBind Which makes what happened next all the more striking. Near the end of the podcast, we talked about a recent “proof-is-in-the-pudding” moment for Genesis — evaluating their PEARL model https://www.genesis.ml/news/zero-shot-pearl-system-surpasses-all-cofolding-models-on-openbind on a recently released OpenBind benchmark. This benchmark featured 802 never before seen co-complexes on a target protein EV-A71. This target seems almost custom-chosen to give most classical docking methods a problem. When a ligand binds to the main binding site, the protein moves around to close off the path the ligand used to enter the binding pocket. This process, known as “induced fit” is notoriously hard for traditional methods to model. The tradeoff is easy to understand: treating the protein as a static structure, it becomes difficult to place a ligand in a binding pocket. Treat the protein as dynamic, and now you have to simulate complicated processes that take a long time to resolve. PEARL was able to model the induced fit of the ligand without running long MD simulations. Across the different evaluation metrics, PEARL came out not just ahead, but oftentimes well ahead of any public model. A truly impressive result. “Where PEARL was exceptionally good is figuring out how to move this loop. We are basically correct for every single pose.” — Sergey Edunov Even more exciting, this was done without any fine-tuning, or using any data on the target or homologous targets — the template PDB was released after PEARL’s training cutoff. Where does co-folding go now? As someone who has followed or participated in ML techniques for protein-ligand interactions for almost a decade, I was genuinely impressed with the results that Genesis has released recently. This has been many years in development, and I’m sure Evan and the team had many sleepless nights trying to get to this point. I also think other teams are making similar progress — both Isomorphic and Deep Origin have released results that seem spiritually similar and combine computation, wetlab data, ML, to achieve genuine predictive power that seemed impossible a decade ago. Sadly, all of the above are closed source so there’s no way to honestly compare them. Looking at the results I think there might be a time in the not so distant future where we can consider protein-ligand binding “solved”. I sincerely hope that the academic community can take inspiration from these developments. Once you know something can be done, it’s much easier to execute. Still, I believe that the key enabler in all of the above was the tight integration of ML, large-scale computation, and real-world drug discovery applications. Sadly academia is just not structured in a way that makes such a development easy. With those parting thoughts, we hope you give the podcast a listen! This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.latent.space/subscribe https://www.latent.space/subscribe?utm_medium=podcast&utm_campaign=CTA_2

• 0:00Speaker 00

  I remember very clearly in like 2017, 2018, talking about GANs and how generative adversarial networks and how they're clearly the future of image generation, obviously. They didn't work very well for proteins or protein ligand systems. And we sort of had to wait for the right primitive to get created. And that turned out to be diffusion, which turned out to be a much more useful primitive for the space. What's kind of cool is right now, for people that are interested in really core fundamental AI research, actually, some of the most innovative diffusion research is happening in our field, is happening in 3D structure prediction right now. No one would have predicted that then, but now like

• 0:44Speaker 00

  that's a pillar of diffusion, I'd say.

• 0:48Speaker 01

  Hi there. Welcome to the Late in Space AI for Science Podcast. I'm RJ Hanke, CTO of Mira Omics. I'm joined by my co-host, Brandon Anderson, and we're privileged to have Evan Feinberg, founder and CEO of Genesis Molecular AI, and Sergey Yarunov, who led Llama 2 and Llama 3 pre-training before he joined Genesis as CTO.

• 1:12Speaker 02

  Hi, I'm Sergey. I studied physics at school. This was a long time ago. And after graduation, I happened to work in software engineering, but I will never need physics again. And the ML came up and became a thing. And turns out that a lot of things you're doing in ML were actually very similar to what you would do in physics. So I jumped in on ML, back then, and I did a lot of AI research, being a part of their team in Facebook for quite a while. I later on led Llama team, Lama 2 and Lama 3 models. And then recently, I decided to pivot my career again and recover my roots in physics a little bit, and I joined Genesis as the CTO.

• 1:57Speaker 00

  I'm Evan. I'm the founder and CEO of Genesis Molecular AI. And like Sergey, also a physics major. I was a bit different from everyone in my family, growing up. Almost everyone was a medical professional of some kind. And my sister became an accomplished TV writer, playwright, novelist. And my love was physics and computer science. But my mental model of an adult was, well, you should help people and help patients ideally. So I was always searching for the right, the right way to do that. And after arriving at Stanford, doing my PhD in Vijay Pande's lab, we were excited in the mid 2010s of everything amazing going on in machine learning,

• 2:39Speaker 00

  to use a dated term, for images and for language. And around the same time that Sergey was at Fair, working on a lot of big graphs, I was at Stanford, working on many small graphs. Because molecules are really networks of atoms and bonds and spatial interactions. And we were at the right place, the right time to bring to bear our backgrounds in physics to improve AI algorithms for looking at molecules. We published a few papers in the area of graph machine learning. And sort of like Sergey, I thought, well, I won't need this physics again. Because, you know, there's machine learning. But turns out, the massive amounts of simulations on GPUs of proteins that we ran

• 3:22Speaker 00

  also came in quite handy as Genesis evolved. And we've been really excited in the past few years to figure out how to build foundation models for a totally new domain and make them useful for patients.

• 3:33Speaker 03

  Yeah, that's a really nice lead into my first question, which is, we've both been in this domain of sort of machine learning for molecules and bio for roughly 10 years. It's kind of an entire generation of tech bio has, you know, come and gone since then. While a lot of machine learning for molecules has been, I think, quite effective. One domain where it's been not effective has, or historically really resisted, machine learning modeling has been the world of protein, small molecule interactions. And with some of recent advances that Genesis has put out, it seems like you might have actually started to make real improvement on this in a way that we haven't seen for a long time.

• 4:14Speaker 03

  Can you talk about what you have done, what Genesis has done? The developments which led to improvement and why you think this is actually a real improvement over kind of some of the traditional machine learning strategies, which were ambiguously helpful.

• 4:30Speaker 00

  I totally agree, Brandon. And the amazing thing is, or one of the, one of the really remarkable things is that when we were founding the company initially as a spin out of the research we were doing in AI at Stanford, we were afraid that we could be too late.

• 4:50Speaker 00

  I mean, you remember that, well, you and I first met around that time, right? Just seven plus years ago, as you rightly point out, almost a different generation in the area. And at the time, we were a little seed company. There were already incumbents in the space who had raised, or is the magnitude more, more money than us? There are already some really exciting academic papers that were out. And we had to constantly answer the question, sort of like, is there room for another company doing AI in drug discovery? Which is a crazy thing to say, looking with, with a backward facing light and, you know, hindsight's 2020, I suppose. And the claim that we made then is the same that we made, we're making now, which is that there's 20,000 protein coating genes.

• 5:35Speaker 00

  And each of them can, can cause a disease. And in the same way that in other domains of AI, it's not like they were zero to one. And then suddenly that problem was solved. The reality has been, there's been no single iPhone moment. Even the iPhone required iterative development over time to become what it is today. And people love to talk retrospectively about the chat GPT moment. But the first chat GPT models that became widely used were vastly less useful than, than they are now. And I think we should look at drug discovery from an AI context in the same way that as we go, we should expect to solve more and more problems.

• 6:18Speaker 00

  And there will be large ethyphonic leaps. There will be iterative improvements, but they'll compound over time. And that's why I think in the same way that 10 years ago was very early, but Clearly, value was starting to be created for drug development then from an AI perspective. Fast forward 10 years, our technologies, which we'll talk about in this discussion, are vastly more useful than they were then. In the next 10 years, we'll see another exponential improvement as well. In the same way that your car, in terms of autonomy, can do things today. We're recording this in San Francisco. If you look outside and see Waymo, we'd be blown away a few years ago by those technologies.

• 7:01Speaker 00

  So I think we should both be looking at it as every few months, every year, the technology gets more and more useful for discovering new medicines with artificial intelligence. And that will continue to be true for years to come.

• 7:18Speaker 03

  Maybe we can go back in time a little bit and explain like, what was the state of protein small molecule drug discovery, let's say, about a decade ago? Like, what could we actually expect to do with machine learning models? And what were some of the failure modes that people, I think, didn't really see coming from a practical standpoint back then?

• 7:39Speaker 00

  It's funny, just I've been breaking the fourth wall a little bit. I came in not really knowing any of these questions, but Brandon really is knowing how to speak my love language. Because like the other people in this room, you know, though my background is much more in the physics and CS side, I've been very, very focused in a deeply passionate way about the area of how can we make medicines with AI for over a decade. So I've seen so, so many tectonic shifts as a view. To give you one concrete example, and I'll give a little background for those who don't have as much biological background. Drug discovery is akin to finding a key for a lock.

• 8:24Speaker 00

  Where the lock is usually a protein, in some cases a nucleic acid, right? And where the drug is the key, and it's usually some, some small molecule, or a peptide, or, or an antibody, or some other modality, that wants to bind that lock, that protein, the receptor, and change something about it. If it's an enzyme, you want to usually stop it from functioning. Sometimes you want to activate that endogenous protein, but it's to, is to introduce an external molecule, an exogenous molecule, to change something about the biological pathway. A necessary, but not sufficient, part of that process is finding a molecule that binds well to that receptor, that protein. Of course, that's not sufficient.

• 9:07Speaker 00

  We also need to make sure it does not bind to certain anti-targets, certain proteins you want to avoid, that often mediates toxicity. You need adme and toxicity, which is typically like 30-ish properties, to make sure your molecule is safe, in addition to being effective and gets to the right tissue. These are all critical, and all are necessary and none are sufficient. But one concrete example, to answer your question about protein ligand interactions, is that we had a hypothesis for a long time, that if we can predict the 3D structure, the 3D coordinates, can imagine for the CS people, like a point cloud, the 3D positions of your

• 9:51Speaker 00

  drug and the 3D positions of the protein, and if you could model that with high accuracy, that will lead naturally to measuring binding affinity, or predicting potency more accurately. That was a hypothesis that fundamentally could not be tested, because models for predicting those 3D poses, the 3D structures of complexes, were so bad.

• 10:15Speaker 03

  Or if you can predict this, it requires so much computational tech resources that you might as well just solve the problem.

• 10:21Speaker 00

  Right, like you might as well solve the 3D crystal structure or cryo-EM structure. It's what you're saying, which which can cost tens of thousands of dollars. It can be months or years. Right, entire PhD theses, post-docs can sometimes work literally 24/7 trying to solve one 3D structure. And the hypothesis was that if we make this orders of magnitude faster, sometimes more accurate, which we can talk about with artificial intelligence or before that with molecular dynamics, that you would thereby not only accelerate drug discovery, but enable the discovery of medicines for targets that were previously thought to be undruggable. And that was just a hypothesis and it and it took until the last few years to show that

• 11:04Speaker 00

  if we systematically improve those, the accuracy of those, those predictions, if we can prove the accuracy of a protein ligand 3D complex, we can make potency prediction more accurate as well. And that's something that I am personally extremely gratified to feel like we've been showing because that was just a hypothesis when we started a company a few years ago. And and now it's become a reality thanks to the convergence of a lot of different ideas that we're able to to to enable Pearl and other foundation models like it.

• 11:38Speaker 03

  Yeah, can you talk about Pearl a bit? So the, I think one of the interesting things about Pearl in my mind is it was an attempt to kind of scale what seems unscalable to me. RNA small molecules, you know, normally if you scale them, models just come out as just like pure pattern matchers. These things love to tell you what you already know. And sometimes they're maybe not so great at telling you something you don't know. So maybe it seems like there's hints that Pearl has actually moved beyond this. Like, what were the key insights that went into this?

• 12:09Speaker 00

  Sure. And by the way, Brandon, like what you described, that shouldn't surprise any like pure ML practitioners in the audience, just because machine learning is initially constructed for a pattern recognition. And it's the same with when you fire up claw. Or ChatGPT or what have you. It's going to do best when it's closest to the train data and it's going to pattern match often. So I think that has been the big sense of urgency in AI, meaning the physical world, is how to extrapolate, how to make generalizable models. And that's us that we've been really hard at work on. Maybe, maybe Sergey wants to give some more, more color on that.

• 12:49Speaker 02

  Maybe to step back a little bit and discuss what PRL is. It's a structured prediction model, which basically means it takes as an input a protein sequence and a ligand representation, but we try to attach to this protein. And it predicts how this protein and ligand are going to look together as a structure in the 3D space.

• 13:09Speaker 03

  So like a term for this, people sometimes use as like co-folding.

• 13:12Speaker 02

  It's called folding.

• 13:13Speaker 03

  Alpha Fold 3 and then some of the, and then Bolts and Open Fold 3 and so on, have also implemented some of these ideas in their own way.

• 13:23Speaker 02

  Yes, but several models, some are open source, some are closed source, but are pursuing similar, similar direction. Our fundamental difference is that we're focusing on small molecular space. There are a lot of models that are doing protein, protein interaction, interactions. And it works well for small molecular space. On the first glance, it may sound like it's an easier task because like, well, it's a small molecule event, right? So like, why would it be challenging? But the reality is the search space is so vast for small molecules. We're at 10 to the 60, drug-like small molecules in the universe. So good luck searching for the space. And there are also all of, all variations, like you can rotate it.

• 14:08Speaker 02

  You can have different conformation of those molecules.

• 14:11Speaker 01

  Maybe the intuition there is that, yeah, like when you type in a Google search, if you just do a few words, you're going to get a huge list of matches. But if you have a very specific query, then you're going to be, you know, you're going to have a few matches and they're more likely to match. So similarly with a small molecule, if you have a very small query, which is your molecule, then it's likely that you're going to have to search a huge space of, of possible matches. Whereas if you have something very complex, then there's, it's easy to rule out a, a match at that. Is that a good way to think about it?

• 14:46Speaker 02

  That's one way to think about it, definitely. Uh, for me, it is just the computational complexity on figuring out which small molecule will attach to this protein.

• 14:57Speaker 01

  Hmm.

• 14:58Speaker 02

  It's such a vast, um, problem to solve that is im- impossible to do without good models.

• 15:05Speaker 01

  Hmm.

• 15:05Speaker 00

  It's like finding a needle in a haystack.

• 15:07Speaker 02

  Yeah.

• 15:07Speaker 00

  Where everything except your needle is very, very dangerous or just doesn't bind. Or they are needles. Right, right. Or finding hay in a needle stack. Yeah. Might be a. Oh, yeah, yeah. Probably. Yeah.

• 15:19Speaker 02

  Yeah. So now we can go and think about how we train those models, right? So a lot of training data that people are using is so-called PDB, uh, which is a public database of like all of a historical crystal structures. And it's not that big. It's like 200,000 crystal structures and it's very hard to expand. It takes a lot of, uh, time and energy and money, uh, to create new crystal structures. Although there are some projects that are pursuing this, it's still expanding at a glacial pace. Uh, so expanding this database is pretty hard, but what we figured out is possible to do and it's very relevant to our small molecular space is that in small molecular space, you can actually model your small molecules with physics.

• 16:05Speaker 02

  Uh, you can model their behavior and that allows you to create more data that you can train model on.

• 16:10Speaker 01

  Hmm.

• 16:11Speaker 02

  Something which is not necessarily possible in protein to protein.

• 16:14Speaker 01

  Right, because they're too complex to, for a protein.

• 16:17Speaker 02

  Yeah, those are very big molecules.

• 16:19Speaker 01

  Yeah.

• 16:19Speaker 02

  So it's very hard to model them with physics. Not impossible, it's just computationally very, very hard.

• 16:24Speaker 01

  Right. So, so I guess you, what you're saying is you do a really good job of modeling these small molecules using N D or something like that. And so you can kind of put together those structures in your training set at a much lower cost and so that you can use that as your synthetic training set.

• 16:41Speaker 02

  May- maybe just to step back a little bit about our, and talk about our roadmap. Um, so it's not fundamentally different from the roadmap for L L M scaling. Like remember in L L Ms, we have all of the stages. You have pre-training scaling, then you have post-training scaling, where you do like either fine tuning or R L. Now everybody's doing R L. And then you do inference time scaling. So all of those three concepts, they connect.

• 17:07Speaker 01

  Yeah.

• 17:08Speaker 02

  And that's what led us to state-of-art L L Ms these days. It's not fundamentally different on our side. We also have pre-training scaling, where we create a lot of synthetic data to train better models. We do that. Then the second thing we started doing is actually the third step in LLM scaling. We started doing inference time scaling. It's fundamentally very similar. Like in LLM, when we talk about inference time scaling, we're talking about thinking tokens, where LLM, instead of giving you the answer right away, it kind of goes and thinks for a while and then comes up with a response, right? Well, we are doing very similar thing with our models, where a model is forced to think,

• 17:52Speaker 02

  except it's not thinking in language tokens. It's thinking in terms of crystal structures, not fully materialized crystal structures, but some sort of a crystal structure representation in memory. And model kind of goes back and forth with this. And we use physics-based guidance during this process to steer the model output in the right direction. And what we found is that it improves model performance by a lot.

• 18:18Speaker 01

  It's easy to understand how thinking tokens work, because it's just basically parts of the transcript that you don't see. But I think that at least for other models, you have like sort of some sort of loop. Is that similar to what you're doing? Like a loop transformer or something like that? Or is there? There's, I know you mentioned that there's some physics-based verification as part of that in the loop. Is there more to it than that?

• 18:42Speaker 02

  One fundamental block of our models is diffusion-based head. So it's like the same diffusion models what people are using for imaging video generation. Well, we use them for crystal structure generation, right? And diffusion head is by nature iterative, like multiple steps. You're refining your predicted structure. And as you're doing this process, you can steer the model in the right direction.

• 19:08Speaker 01

  So there's like, you have like a steering simulation or something that in the, in the loop in that, so you, you go through a passive diffusion, you look at what comes out, you run some verifier and use that to say, I like this, or I don't like this, or give a vector to go towards, is that kind of the idea?

• 19:28Speaker 03

  Yeah. So it's like, your diffusion head is basically learning something like a force field? And that you're kind of bouncing off between a diffusion based force field and like a physics based force field? Or is that what you're thinking about it? Or?

• 19:40Speaker 02

  I struggle to understand like what those models are really learning underneath. And I think the whole idea of explainability is a big research challenge. Yeah, you can identify what a neuron and a transformer potentially thinking about, but it's really, really difficult problem. So we, we have some internal representation, but at the output comes out as a crystal structure, it's really hard to tell what exactly is happening inside.

• 20:06Speaker 01

  Is it possible to do something to the effect of the standard mechanistic interpretability, things like sparse auto encoders and the more complicated things in this context? Or is there a problem with doing that kind of thing?

• 20:19Speaker 02

  I think it would be interesting as a research direction. That's not something we are pursuing. I guess potentially we could explore this.

• 20:26Speaker 01

  Okay. Yeah.

• 20:27Speaker 00

  Sergey's a lot more credible than I am talking about the LLM translatability to our space. Sergey is being humble, but Sergey led the Lama two research team at, at Meta when, when he, when he's still there. So he's trained one of the most widely used language models in the planet. But I guess why I can add on the, on more the, the physical interpretability side is I like to look at, at AI models in terms of the inputs, the model themselves and the output. And we find it really improves outcomes as well as interpretability to your question,

• 21:09Speaker 00

  when you could include as many physical priors as possible without, of course, you're biasing the model to what we puny humans believe. But I've always had the view that AI fundamentally is representation learning. And to me, that's actually not that different than what's happened in the language and the vision domains, where we're enforcing a real human prior on images when we start using convolutional neural nets. Is we're saying, well, pictures are grids of pixels. There's nothing inherent about that. That's a human construct. And we constructed ConvNets on top of that. And language as sequences of tokens in one dimension.

• 21:52Speaker 00

  We first built RNNs, we then built transformers, but it's still, that's baking in a pretty, pretty serious prior on how to look at that data. And we don't view it very differently here. In our case, from the input perspective, the disadvantage that we have in our field over the more traditional domains of AI is that we don't have the internet to work with. Like, we can't just download Reddit posts and you buy some subscriptions to the Wall Street Journal and like, train a model and voila, the pre-trained model works fairly well. The closest equivalent that Sergey mentioned is the RCSB Protein Data Bank, which has more like a couple hundred thousand crystal structures, prior instructors and others,

• 22:36Speaker 00

  although there's a lot more, there's a lot of tokens per structure, so there's a lot more latent information that that figure would describe. So we have to be clever on the input side, which is generating more pre-training data as Artis mentioning, using physics as much as possible. In terms of the model architecture itself, the idea is to not have the model have to relearn as little physics as possible, so is less likely to overfit. And then at the output stage, enforcing physicality, I think that also goes to one of our main focuses as a company, which is, we've been focused from the beginning on small and medium-sized molecule discovery. What does that mean?

• 23:16Speaker 00

  So, small molecules tend to be drugs that you can take as a pill, so it's orally available. It's how most people think of medicine. And medium-sized molecules, so think about backrecycles, peptides. Um, modalities that break the traditional rule of five, but are still growing, uh, growing modalities in medicine. That said, of course, small molecules are still 65% of FDA approved drugs. So we're talking about the biggest part of the pie. We're building this with the judicious focus from day one on what do drug hunters? What do medicinal chemists? What do CAD scientists? What do they need to just to discover drugs faster? And better than they could before. And so we wanted to to build that sort of um.

• 24:02Speaker 00

  Human, not interpretability, but usability from the beginning and also physical usability. We want outputs from these models to be used by. Physical methods that require 3D coordinates to actually make sense. Use the term force field. We want the outputs of our model to be useful as inputs to a force field. There is this, a few papers that came out. One was in Cell, I think last year, which showed that for all of the claims about. Alpha fold solving drug discovery, people try to take alpha fold produced protein structures. Use them for traditional docking and found no value in it.

• 24:43Speaker 00

  Those pockets just weren't high resolution enough. They couldn't be used by a physical screening method. And so we wanted to be able to build our systems that be useful by humans, useful by all the many adjacent and powerful tools. From the physics and computational chemistry communities from day one for to have that interoperability.

• 25:06Speaker 01

  Yeah. So this might be a good time to just step back for a minute. When you develop a drug, there are many steps to that from discovery and toxicity and availability. And. So there's names for all these things, which all you you state. And then there's the clinical stuff. And finally, uh, eventually approval, hopefully. Where do your models sit in that? Or which parts of that process do they say? And what what are what are the steps? Just like briefly, I've read that there's 12 steps, maybe.

• 25:39Speaker 00

  I don't know. There's 12 steps in drug discovery, except it says the last one.

• 25:44Speaker 03

  Except for your faith in phase three trial failed.

• 25:47Speaker 00

  Yeah.

• 25:48Speaker 01

  So where do you do your models help people to do their jobs in the discovery or in the development process?

• 25:55Speaker 00

  As you rightly point out, if we go from the very beginnings to the very end, first, we need to identify what target is causing the disease, what signaling cascade, what's going wrong, what's causing the phenotype of the patient. And that's not a trivial exercise, but there are many, certainly not all, but many diseases where the cause is actually very clear. The most clear are the monogenetic diseases, but even in others, it's from various panels, it's clear that this overexpressed or overactive or deleted gene or protein is causing the disease. So we know many, but not all of the universe of causes of diseases.

• 26:39Speaker 00

  Then, once a target is identified, we have the process of finding a drug for that target that's ideally selective, very potent. It gets the right tissue. Some diseases are multi-systems, some are very specific to a certain organ or tissue, but I make sure that drug can get that tissue. And then there is the process of GLP tox and IND enablement, IND investigational new drug.

• 27:08Speaker 03

  So GLP in this case is not.

• 27:11Speaker 00

  Not the.

• 27:12Speaker 03

  Not.

• 27:13Speaker 00

  Yes. We're not.

• 27:15Speaker 03

  Good lab practices.

• 27:16Speaker 00

  Yeah, thank you. We're not talking about the metabolic space right now. There's clinical trials, usually ascending size, your phase one starts typically more with safety than phase two and phase three. And as you point out, ideally approval at whatever regulatory body, the FDA or E N D A or what have you. Our contention is that the highest leverage application of artificial intelligence is the drug discovery and drug design process. And the reasons are that even though they're all valuable and I cheer on all of the peers, many of whom I know who are working at all the different parts of the stack.

• 27:56Speaker 00

  But I think the first thing to make clear is it just like a vision model is going to be very different than a coding model. They might share some similarities under the hood, but requires real focus to do any one area correctly. There's a similar, if not greater difference from this sort of models you need for target identification, i.e. biology, to drug discovery, to preparing for regulatory filings, to helping segment patients for clinical trials. These are all very important, complementary, but ultimately distinct problems. I think about all of the times where a patient goes to a doctor, gets told that they have

• 28:41Speaker 00

  a very clear diagnosis. And in addition, the physician will tell them there's been breakthroughs where we understand why this disease happens. We've sequenced your genome even, we know that or sequenced your tumor, we know it's causing your condition, but we do not have a selective therapy. There is no precision medicine for your condition. And it has the hope that still at the moment, at least they know what's wrong with me, but they don't really know how to treat it. And our aim is to have as many moments as possible where patients are told they have a very specific condition and there's a specific treatment for them. That's going to require bending the cost curve of developing new medicine, but also

• 29:26Speaker 00

  importantly, solving those cases where there is certain quote, "undruggable targets", undruggable proteins that we need to figure out how to drug, where those targets have proven resistant to traditional methods or even intractable in some cases. Just to clarify my understanding.

• 29:43Speaker 01

  There's the discovery part and then there's the clinical part where and and so there's there can be effort there where, okay, I want to find what's the target that is impacting this disease the most, or how do I maybe what's the thing that I want to go after in order to solve a medical problem? And then there's the clinic where you're answering, does this thing actually work? And maybe there's a feedback loop between them even. But if you don't have the thing in the middle that's like, okay, here's how we actually build this drug so that I can't hit my target. It's not selective enough. So it like kills cells or does things to cells that I don't want it to do. So then it doesn't matter, right? Like I can have the answer. This drug we

• 30:28Speaker 01

  should go after this target. But then if I don't know how to do that effectively, quote unquote, et cetera, then then the answer doesn't matter. Is that kind of what you're saying?

• 30:36Speaker 00

  Yeah, RJ, if you because people love to debate about success rates in in of clinical trials. But the reality is if one focuses on those drug candidates that are aimed at proteins that have a close genetic linkage to a disease and or at least where the biology is well understood, where the animal models translate well to the disease, where the molecule is predicted to have good pharmacokinetics, as in the levels in the blood in patients or in serum are going to be high enough. Those molecules have fairly high FDA approval rates. The success

• 31:16Speaker 00

  rate from phase one to end of phase three is substantially higher than average. People love to cite the 10% success rate, but that's really a low ball. Because we often know what genes, what proteins, what targets are causing disease. They're just really hard to drug, or the molecules we put in the in patients are inferior to the the target product profile that that they deserve. In the cases where if you just look at the preclinical data, you focus on good targets with good biology that are predicted to be distributed well and have good safety Profiles, those molecules are very likely to get approved and it creates tremendous

• 31:58Speaker 00

  value for patients. And so our view is, I think about the physicians that run trials that are clamoring for those kinds of molecules. I think about the patients that are looking for more selective therapies. And so our view is, that is the highest leverage application of AI in healthcare medicine or broadly.

• 32:19Speaker 03

  Just a business question here. What is the space of things which are, the target is, the biology is understood, the structure is understood, and that, you know, the only thing you really needed to do is find the right lock. That seems like you probably would have already picked over that class of targets. In some sense, those are easy targets. Right? So how much is opportunity is there for that?

• 32:42Speaker 00

  So there's two orthogonal concepts. Known biology is orthogonal to the ease with which one can drug that target. Sometimes unfortunately, it seems they anti-correlate. And that often it seems that the most appealing targets from a validation perspective seem to be really hard to drug.

• 33:04Speaker 03

  Yeah. Do they anti-correlate? Or is it just that we have picked all the low hanging fruit, which do solve both?

• 33:09Speaker 00

  I think it is likely the case that many of the quote easier targets, for a variety of reasons, have been sort of, many of them have been drugged, not all. But even in those cases, people love to sort of have this false dichotomy of easy versus hard targets. And well, this one's drugged versus it's picked over, therefore it's not. But we at Genesis do most of our work through large pharma. Like, most of what we do is providing AI services basically to major pharma companies like Gilead. And we recently announced our expansion of our collaboration

• 33:50Speaker 00

  with Insights, which we're excited about. And I can't give details of specific targets that they're working on, but what I can say is there's a real range from first in class chemical matter. What that means is, we believe this target causes a disease. There's no known molecules that bind to that target. We need to find the first binders ever, the true zero to one cases.

• 34:13Speaker 00

  But there's a wide variety of targets we work on that are more, I'd say, one to 10 cases, where sometimes there is an approved agent, or sometimes there's molecules that are only preclinical, but they're not optimal, where if you can improve upon those preclinical molecules and get them to development candidates, so ready to get into patients, or if you can improve upon the existing clinical or approved agents, you can create a lot of value for patients. I'll give a public example, we're not working on this target at all, but you just look at the progression of ALK inhibitors, ALK inhibitors, and you look at, to make it very concrete, charts of patient survival. You might have said, well, when the first

• 34:58Speaker 00

  ALK inhibitor came out, why do we need another? We've drugged ALK, we're done. But it is a qualitative, not just a quantitative difference, when you compare patient survival curves from first generation ALK inhibitors to later generation ALK inhibitors, of how much longer those patients live, how many more get benefit from it. And so I think there's not only value in first in class, there's enormous value for patients in best in class too. And I think both are areas that we focus on, if that makes sense.

• 35:33Speaker 03

  Okay, yeah, so we so far have talked about Genesis modeling in terms of, let's say, this one specific problem of protein ligand binding. How does your overall world work in terms of developing this? I mean, I assume that you must, I mean, I do know that you have other things that you've worked on in solving the broader early phase drug discovery problem.

• 35:57Speaker 02

  I keep joking about that when Nobel Prize was given for alpha fold free, a lot of people thought that drug discovery is solved. And it's really far from the true.

• 36:08Speaker 03

  Very, very far.

• 36:09Speaker 02

  Yeah, you can maybe, maybe, so maybe you can predict crystal structure, right?

• 36:14Speaker 00

  At low resolution.

• 36:15Speaker 02

  At low resolution. Well, we're doing better, I think.

• 36:18Speaker 00

  Yes.

• 36:19Speaker 02

  I assume you can predict crystal structure. Mean the drug discovery solved? No, obviously not.

• 36:26Speaker 03

  The broader community, like protein folding community and drug discovery community immediately said, there's other things you care about, like dynamics, like just a single static structure is not enough to understand what's going on, interactions. Yeah, so there's lots of things there in addition to just like having a static structure. I think many people thought those were very useful, like they've really accelerated science, but it was very clearly a starting point and not a in condition.

• 36:51Speaker 02

  Right, but if you look outside of like people working in the field, like general popular community, it was a pretty wild wide belief. The problem is solved. Now, in reality, besides crystal structure.

• 37:03Speaker 03

  I just have to clarify that. Otherwise, like there's an entire ML structure biology community, which will all just jump on us and say like, no, no, not the problem. Gotta, gotta throw this caveat in.

• 37:13Speaker 01

  Absolutely not a solved problem.

• 37:15Speaker 02

  In reality, you need to predict so many other properties, like other properties that I haven't mentioned, like basically your design in the key for a lock, like that small molecule that sticks to your protein in the body. But you don't want the small molecule to stick to everything in your body, because that's probably going to cause issues. You also want to make sure that small molecule can be solved, like is soluble, so that you can ingest it as a pill. Those are important properties, too. You want to make sure it doesn't have any other side effects. And predicting all of those properties is also just as important as, or maybe even more important than predicting the crystal structure itself. And of course, as a company, and we're very proud of it, we're not just building models for predicting crystal

• 38:00Speaker 02

  structure, we're building models for predicting all of those properties and basically enable drug hunters to kind of become hunter texts more effective in their daily job.

• 38:11Speaker 01

  I'm curious about this. The pipeline that you have, you mentioned you have different pharma partners, GSK or whomever you're working with. Do they have targets that were sort of drugs that were designed by you that are in like clinical trials approved? Like where, where do we stand with all that?

• 38:30Speaker 00

  We're limited in what we can say about Of course. Companies are famously secretive, which makes sense. Yes. What I can say, which was recently publicly disclosed as an example. Yeah. Is we just expanded our partnership with Insights. And that started approximately a little over a year ago, the initial work together. And that spanned the two sort of bookends of the drug discovery process in some way. So one of the programs we worked on together was a case where a very challenging target, where chemical matter existed, but there is a key binary event, which is getting to a DC or a development candidate.

• 39:15Speaker 00

  And that is the molecule, or ideally a set of molecules, all which are possible to be the agent for a phase one clinical trial. And we had to do some work together where we take our foundational models, our base models, we fine tune them on Insights data and use them in a close collaboration to sort of work together to get to a DC faster. And we're getting substantially closer in that case, which is one of the reasons we're excited to expand our work together. And on the flip side of that is one of the other areas of work together was on a protein

• 39:58Speaker 00

  with also very nice linkage to a very severe disease where there was no known chemical matter, like no patents, no papers, in this case, no co-crystal structure of a ligand, another synonym for a small molecule that bound to that protein. So we had to find the first ever known chemical matter to bind to it. And then we progressed those, what are called hits. Hits are like the first molecules that bind to your protein. Progress those into inhibitors that are active and called biochemical assays, which are more enzyme based than cellular assays, which means that in a actual cell, living cell model, your molecule is active as well.

• 40:40Speaker 00

  So it was really based on a concrete set of accomplishments together that we want to expand the collaboration. And unfortunately, outside of that, there's really limited any version of what we can say. Yeah. But the objective of Genesis.

• 40:56Speaker 00

  Is to create medicines that patients wish they had. And the way we'll be able to do that is by working with as many pharma and biotech partners as possible for whom their comparative advantage is discovery, clinical development, commercialization. And our comparative advantage, of course, is in AI, where we can put those two expertises together in a synergistic, not just an additive way, and make medicines together that otherwise would not be possible.

• 41:30Speaker 01

  Yeah.

• 41:31Speaker 00

  So that's the name of the game that we're in this for is those clinical outcomes that you're pointing out. And I'm very excited to be able to share those as they rise in the coming years.

• 41:41Speaker 01

  So one thing that is, you guys talk about in your website and whatever, is this one Angstrom threshold at which a protein structure prediction and the binding between that and a small molecule becomes useful. Can you talk a little bit about like, okay, why is that the case? What, how did you do it when others have failed? And, you know, and you've spoken a little bit about that with, you know, sort of biases in the model, the synthetic data, maybe there's more. How does that impact, you know, sort of downstream, the actual things that we're talking about here with, you know, the Admat and all that stuff?

• 42:21Speaker 02

  When we're talking about molecular interaction, the scale of two Angstrom that people typically measured is just too big. So you can think about like in image generation model, you would generate a picture and it's like fuzzy. So yeah, it's like in general, right? But you can't discern the details and the details really matter here. Like with two Angstrom accuracy, your entire aromatic ring can be flipped and it will still be a valid output.

• 42:49Speaker 03

  The worst part is that unlike an image, which is blurry, you, you know, you don't even know it's blurry, right? You flip around an aromatic ring and it looks just fine. And yeah.

• 43:02Speaker 00

  Or even worse, a heterocyclic aromatic ring and then you're really in trouble.

• 43:06Speaker 02

  And it really matters in this case because like individual atoms need to establish connections here. So that level of accuracy what they're pushing for one on stream is really important.

• 43:17Speaker 03

  So maybe it's almost like with a lot of generative image models, they will fill in a detail which just doesn't exist. So if you were to try to do like forensics and you tell, you know, your, your vision transformer to, to, to enhance an image and suddenly it just pops up a face, but that's actually not the person's face, right? But like you don't know that and it feels just fine. And so if you're trying to use this as a structural hypothesis as a med chemist, you're kind of screwed. If you're looking at the wrong faces and investigators, it's not going to take, you know.

• 43:49Speaker 02

  Yeah, I really like your framing. And then as Evan said, that like it was as a downstream, you run a bunch of other models, like physics based models, and all of those little issues may compound. And then obviously, if you made their own prediction, like fundamentally wrong, then the downstream predictions are also going to be wrong.

• 44:08Speaker 00

  I just wanted to say for your question, I really appreciate your mid 2000s TV reference about, you know, enhancing forensic images. So I just want to say I appreciate that before we get back into, get back in the details here.

• 44:20Speaker 03

  We talked about poses. So maybe the, the contrarian take is that a pose isn't even a well-defined concept in that, you know, the best way of thinking about these things is this like, you know, probability distribution over things. And there's a small molecule which probably lives in a binding pocket, something like this. I mean, is, is this pose, like this is an abstraction that humans use and not really in some sense of ground truth. So I'm actually almost kind of, kind of interested to actually hear that like, oh, we have this one in some threshold. This is like really important for med chemists. I would also talk to some chemists who would say, I mean, this pose isn't even real. This is just like maybe a most probable configuration or, you know,

• 45:01Speaker 03

  how do you think about that abstraction in general? And do you, do you explore confirmation space and provide that as tools? And yeah, how do you interact with just like a single structure versus like an ensemble? Sorry, I didn't know if that question even makes sense, Ben.

• 45:17Speaker 02

  Yes, it's an abstraction, but it's a very useful abstraction. It helps us to build up confidence that a particular model output is actually valid. But you did not just straight up hallucinated something. Because yes, ultimately, what matters is binding affinity or potency. And you can straight up predict that with your model and just skip the entire pose generation step. But then you only have a single number and that number might as well be completely hallucinated. And you have no means to validate whether that number even makes any sense. So as much as pauses are not perfect, they're still a very useful tool for the entire process.

• 45:58Speaker 03

  Sorry, going to that correction, you know, just see if I'm afraid I'm jumping into a technical rabbit hole here. But but like the just because you have a pose also, I mean, there's like entropic and enthalpic contributions. And predicting binding affinity is not just about getting the energy right. But actually, is this even likely to make it into the binding pocket? And is it like to live there long term? So that's much more than just, you know, affinity potency prediction is much more than just is this the right pose? But does this have the broader properties it needs to be a long-lived molecule in this state? And how do you kind of deal with those things too?

• 46:36Speaker 00

  A note for sort of the wider AI audience, probably resonate both with practitioners as well as users. Is, you know, obviously something that's blossomed in the past six months is agents. And we love agents.

• 46:53Speaker 03

  Who doesn't?

• 46:53Speaker 00

  It's there's a lot of, I'm going to say a few necessary, but not sufficient conditions. I'd say in in my my sort of response to your really deep question there. We all remember what agents were like, let's say, mid last year. And let's just say there is positive value and there's negative value and both can be amplified by agents. And why is that? Well, agents are only as useful as the underlying models that they're orchestrating. And let's think about coding. If your coding model even makes subtle but but real bugs, your agents are just going to amplify those issues and you're going to end up with not only slop,

• 47:36Speaker 00

  but something that that may be anti-useful that might give the user incorrect information. We all remember what that what that age was like mid last year. That made a lot of people lose patience, I would say, with claims about LLMs for for agentic engineering. Something changed, like clearly a threshold was was was was met. And even though these models are still not perfect, the utility of LLMs for software engineering are so obvious now. I mean, obviously, it's been a huge tailwind and very useful for us for for replacing a lot of the drudgery of coding and getting a focus a little bit more on some of the the more strategic

• 48:16Speaker 00

  issues that matter. You could draw a direct analogy from that to what we're talking about here in that we obviously know it will be no surprise, but we're working on an agentic platform for 24/7 drug discovery. You can imagine just fleets of hundreds of metchemists and cat scientists working nights and weekends all the time for for your drug targets. The the codename for that gem is Sapphire. The prerequisite for that was we needed the underlying models for pose, 3D, 3D complex prediction, potency, admin, to all be good enough

• 48:59Speaker 00

  for an agent using these models 24/7 to create molecules that medicinal chemists would actually want to make and not laugh at. And if your model is sitting at 1.8, 1.9, Onstrom, or MSD, that's slop. Most likely. And let me be really direct about it. What you're hearing, so you're hearing bimodal distribution from people of the utility of a 3D pose and how real it is. The reality is that for a highly potent ligand, almost certainly, there is a large portion of that molecule with a very well-defined 3D position, down to even half angstrom. And if you don't believe me,

• 49:42Speaker 00

  you can open up the electron density diagram in the PDB. It's all available online and you can, you can literally see, in some cases, aromatic rings with missing density in the middle. So it's not just a construct on a blackboard that your organic chemistry professor showed you. It is, you can literally see a donut, a torus of electron density for an aromatic ring. However, there's going to be solvent-exposed areas, oftentimes, that will be less important for binding affinity, but maybe are important for solubility or other properties of your molecule. Those sometimes are less defined because sometimes, in reality, to your point, they're more dynamic, they're flopping around in solvent.

• 50:23Speaker 00

  But the critical piece as a upstream indicator that's valuable for predicting the free energy of binding is to get the core of your molecule that specifically interacts with the protein to be correct to sub-angstrom resolution. And why is this matter, just to use intuitively? So a hydrogen bond, everyone in the audience has probably heard about a hydrogen bond. It's one of the critical forms of non-covalent interactions. It's how nucleic acids are held together. It's how most ligands and proteins interact. It's how proteins form secondary structures. Hydrogen bonds have a very specific angle and distance. And the distance is from the donor to the acceptor heavy atom,

• 51:07Speaker 00

  it's 2.7 angstroms to 3.3 angstroms. And if I do my math right, that's a 0.6 angstrom gap. And outside of that, it's not a hydrogen bond. If it's less than that, it's a clash. If it's more than that, the interaction is much, much weaker pretty quickly. And angstrom, for those who don't know, is 1/10 of an anemometer. So drug discovery really is a science of resolution.

• 51:33Speaker 00

  And if your accuracy is not sufficient, it will therefore not be useful for the downstream things that you care about, which is both potency prediction, but also prospective design. What molecule do I make next? So that's why our view is that the history of innovation in startups is that the ones that do best are ones that focus on one well-defined but very important problem. And we think our ability to get higher resolution predictions stems from error, judicious focus on small and medium-sized molecule design rather than boiling the ocean.

• 52:10Speaker 01

  Yeah. So that's the what and the why. So what about the how? How did you get to one Angstrom and sub one Angstrom?

• 52:18Speaker 02

  I'm gonna give you an extremely boring answer.

• 52:20Speaker 01

  Okay.

• 52:21Speaker 02

  Which I think is actually true for entire AI field, like where three things matter in AI. It's data, infrastructure and evals.

• 52:32Speaker 01

  Okay.

• 52:32Speaker 02

  Right? So you can only improve what you measure. And once you are very careful about measuring what matters and you have really talented people on the team, we're gonna figure out how to hill climb that measure, right? So from the start, we actually focused on less one one Angstrom precision. And that led to a bunch of small decisions in the process that compound, right? If your team never looks into this metric at all, then you will never train a model that is good at it. And if you're constantly looking into that, then you're gonna achieve that.

• 53:07Speaker 01

  So it's the right objective plus good science.

• 53:09Speaker 02

  Correct. And it propagates everywhere, right? Through the whole stack. It propagates to how you look at the data, maybe like, how do we filter out data? Some data is more noisy, so maybe you don't need to see it, or maybe you don't need to see it later in the training. It propagates through your modeling architecture and propagates through your loss.

• 53:27Speaker 03

  How long have you been working towards this specific goal? And I'm curious, this is not something that you broadly hear in the the community talk about one Angstrom. People do say to have decided two Angstrom is sort of the canonical benchmark. And I've never heard someone say one Angstrom before is the cutoff until like reading things coming out of Genesis. How long have you decided this is the number we need to hill climb on? How direct has this focus been? And in like the evolution of the company, I'm just wondering, like, how did this come about?

• 53:58Speaker 02

  I think one of our most important secret is that we are working on real drug programs, either with partners or in-house. And when you actually work on real drug programs, you see the failure modes. You see what works and what doesn't work. And it's pretty obvious when you see at the outputs of real programs, what kind of failure modes are happening. And it becomes very obvious that Tongstrom is just not working out for those setups.

• 54:27Speaker 03

  Right. So I mean, there's a lot of really smart med chemists who are also, you know, who think very carefully about benchmarks and people I respect very much and who also have prosecuted actual drug discovery programs. So I'm, I know, I'm just kind of curious, like, why hasn't this become part of the community? Is this literally just that the community has never been able to succeed at a winning some benchmarks that they kind of settled it to as a, okay, it's something we can aspire to? Or I, I know, just.

• 54:56Speaker 01

  But I mean, to your point, it sounds like there's a, in my experience with pharma, there are plenty of problems for which people, there's kind of these known things amongst the technical experts in a subdomain and that, but that it doesn't get out of pharma or it doesn't get the attention it deserves because, you know, for whatever, some of the information is proprietary and gets kind of passed from company to company, but never really released into the public. Is this your estimation of what's going on?

• 55:29Speaker 00

  Have you ever heard of Sweet Bench?

• 55:30Speaker 01

  Yeah.

• 55:31Speaker 00

  So Gemini does pretty well on Sweet Bench. Sometimes Gemini publishes models that win on some of those software benchmarks. Raise your hand if you're using Gemini to write code right now instead of, you know, the obvious other name competitors.

• 55:47Speaker 01

  Yeah, not me.

• 55:48Speaker 00

  No one, like, why would you do that? It's, it's obviously worse in, in practice. And I, I'd say actually that in this case, if you look at the provenance here of how that happened, ARMISTY less than two came originally from, again, Brandon, I've been in the space forever, from docking studies, long before AI models for post-prediction when physics-based. Studies from academic institutions because usually large, you know, the proprietary software makers didn't want to benchmark their methods against other methods. And so academics had to get licenses and try it. And then they introduced our M C lesson too, which is not surprising. They're academics. They're not using these things to make drugs.

• 56:26Speaker 00

  They're using it to write papers. And so that got sort of the provenances that that kind of got repurposed by the A I community. But the actual trend is much more in the direction that we're saying. So the first big innovation there was Posebusters, we'd say, put out by a lab at University of Oxford, which, which pointed out that R M S D itself is insufficient and we need to look at physical validity as well. So that's Posebusters that I'm talking about, not the benchmark, but the metric. So, so Oxford improved that. And in the latest release from OpenBind, we just published our benchmarks at Genesis on OpenBind that we

• 57:09Speaker 03

  We'll talk to that in a bit.

• 57:10Speaker 00

  Oh, great.

• 57:11Speaker 03

  We planned on that.

• 57:11Speaker 00

  But, but if you look in, in there, you know, that's their original publication, their default metrics are using R M S D, Posebusters validity and L D D T. So there's clearly an acknowledgement that's already been made that R M S D lesson two is insufficient. And the field is now rapidly evolving to acknowledge that.

• 57:32Speaker 03

  You think that the academic literature is going to establish some, some benchmark, maybe it's one ingstrom, maybe it's some L D D D or some other metric that will kind of converge. And then hopefully as a community drive forward these like more accurate model, modeling strategies.

• 57:49Speaker 00

  I would say there was an eval crisis in our field that is now in transition right now as our field, which was previously, let's say a lot more quiet. But if you go to NURIPS and I C M L, every year, year after year, their workshops get a lot more crowded. Those evals are, are in transition now that we realize the flaws of what came before it.

• 58:11Speaker 03

  That was a really interesting point about, you know, click the right evals and then just do good machine learning like you normally would do. But I kind of picked up on something Evan said earlier, which is that you started out with, I guess, what you call potential net. I don't think you defined that earlier, but you know, this graph-based network. And then you did a lot of computational simulations kind of after that. And now you're sort of going back into not once generative modeling took off. I'm curious about how that evolution worked and how did you find computational techniques to be like really crucial to sort of building on the ML for a while? And did that that sort of computation lead back into generative? Or did you start collecting

• 58:56Speaker 03

  data to build a generative computational data as like this is a way of data documentation for a generative model? Like, what's sort of the history of that? And then what led to those decisions to the extent you can talk about them?

• 59:07Speaker 00

  I will say that the last line of PyTorch I've written is much further back in history than the most recent line of PyTorch that Sergey has committed. So I want to make sure that he gives his opinion as well. I want to make sure that I'm going to directly answer your question. I want to make sure that I address something you said like a little bit a time ago that I think kind of got lost, which is you asked about how we do other things, including AdNet prediction. So I just want to make a comment about that. Just I can spend all day and I'm happy to diving into details about Pearl and the evolution and we will. I just kind of want to give a shout out to the fact that it is one important

• 59:49Speaker 00

  pillar, but not the only pillar that matters in the trajectory of a drug discovery campaign. And you mentioned the potential net paper, which we're happy to see have been become influential since you and I were working in the space when it was very much in the future, let's say. But now it is the future. So it's great. Another paper we published around the same time was on neural networks for AdMe prediction. And we've published two, actually.

• 1:00:19Speaker 03

  Sorry, AdMe, AdMac?

• 1:00:21Speaker 00

  Oh, sorry. Thank you. Yeah. Absorption, distribution, metabolism, elimination, and toxicity are the five prongs that comprise AdMe.

• 1:00:29Speaker 03

  These are all the properties you talked about before that you just have to get right in order to make a drug successful.

• 1:00:35Speaker 00

  Correct. So I would say there is Over thirty or so assays, each of which you could imagine if you're a neural net person, like a multitask neural network or multi-head, it's got to predict over thirty, you know, three dozen-ish properties, each of which, if it's in the wrong range, means your molecule is not a drug, it's just a tool. And so these are things like solubility, which Sergey mentioned, which is important for formulation, as in can it be made into a pill that you can take orally? Your oral bioavailability, whether or not you're inhibiting certain enzymes, call it cytochrome P450s, and its different variants. HERG, a critical

• 1:01:18Speaker 00

  channel, which if you inhibit it too much, can cause cardiotoxicity. So these are, it's an alphabet soup of things that most people here haven't heard of.

• 1:01:26Speaker 03

  And a lot of these are extremely hard, specifically because it's not like a single causal effect. There are oftentimes many processes, many pathways which are involved in defining this one endpoint.

• 1:01:39Speaker 00

  Correct.

• 1:01:40Speaker 03

  And data sets are oftentimes comically small.

• 1:01:43Speaker 00

  Right.

• 1:01:44Speaker 03

  Unless you have, you know, I guess Pharma to help you out. But at least open source, it's quite a hard problem.

• 1:01:51Speaker 00

  It's sparse in the public domain. I'd say there's actually a range from really directly predictable properties to ones, let's say, that, as you point out, are actually amalgams of other signaling events. So like whether or not you're inhibiting like CYP3A4, that's really specific. It's a certain protein that you're inhibiting.

• 1:02:09Speaker 03

  So that's something that Pearl, for example, could actually model and expect to see some performance here.

• 1:02:15Speaker 00

  Yeah.

• 1:02:15Speaker 03

  Useful predictive power.

• 1:02:16Speaker 00

  We've been doing 3D work on admin prediction before anyone else was. And there was a bunch of work that ended up launching the company. Back in the day when AI could still be in peer-reviewed journals, not just sort of like random white papers on archive, we published Molecular Net, which has been cited a few thousands of times at some point. We also published this paper showing that multitask graph neural networks were the best at the time for doing admin prediction on large Pharma data sets. And that paper is one of the most cited papers on AI for admin ever now. A lot of papers, as you know, it's like you write them and you kind of move on, but that work, both those works have been quite influential as well. So.

• 1:03:00Speaker 00

  Our history from the beginning has been to work on not just one problem, but to focus on drug discovery and in so doing, being able to have the bandwidth to focus on building all of the ML models that are needed for drug discovery and not get distracted by the biological discovery processes that are needed, the target ID side or the clinical trial side, but really focused on all of those tasks that are required for drug discovery and also molecular generation. These are all important. And I just wanted to make the point at the beginning, but as you rightly point out, Brandon, PERL is a 3D structure prediction model and many of these ADMET properties

• 1:03:40Speaker 00

  can be posed as those, so it is therefore useful and not just on-target potency prediction as well. All of them matter and we've had to tackle all of them. Do you use PERL for all these tasks or they're focused on some of them? We'll be sharing some things publicly in the coming period, but I think we've had enough news of results lately, I think so far, if that makes sense. Every time you publish something, it's actually, it's a lot of work for the team to put together. So we'll do that in the coming period, but the most recent thing we published was obviously the open bind results, which was for a standard 3D prediction task.

• 1:04:22Speaker 03

  So one thing I want to ask about, or I kind of keep hitting at this. So you talked about there's a lot of synthetic data and then there's ML modeling, which is very kind of modern general model flavor. I want to talk about kind of the feedback between how your computation, your ML and how wet lab data went into developing this model. And we talked about priors, I think maybe one, it's been typically hard to scale protein ligand models well, in a way which meaningfully generalizes. And this recent PERL paper and some of your other recent results we'll talk about shortly have shown true generalization.

• 1:05:07Speaker 03

  And what I'm kind of curious about is how do these different aspects feed together to give you this power? And specifically, just saying computation is interesting, but you have to be very careful about computation. MD has all sorts of biases. If you're not right, it can give you poor results. And so I'm curious about how all this worked together and especially in the history of the development of the company, how did you kind of get to this point?

• 1:05:36Speaker 00

  There's this, the classic concept of the narrative fallacy, where in retrospect, it's, everything seems obvious and you can draw really linear, you know, processes of how we got from point A to point C. But the reality is always more interesting and much, much messier and much more nonlinear, I think. And for Sergey and I, who've been training or taming neural nets for, for over a decade, we remember a lot of different eras in the space. And we would have been so excited if we could tell ourselves the capabilities of our technologies 10 years ago. And if we told ourselves how we got there, it might

• 1:06:19Speaker 00

  have seemed somewhat obvious, but, but still, still there were some things that would have been really difficult to foresee. So one example is, you know, the concept of using generative AI for, for the molecular space is not a new concept, but the reduction of practice would have been very difficult a decade ago. And so as one concrete example of that, I used to co, uh, co-run the Stanford AI salon. Uh, we had fun organizing, you know, it's Stanford and the Gates Computer Science Building. We'd get some interesting speakers and like 25 people having wine and cheese and they're all now like some very like just

• 1:07:00Speaker 00

  straight up famous people, right? But back then, hey, I was a much smaller field. And I remember very clearly in like 2017, 2018, talking about GANs and how generative adversarial networks and how they're clearly the future of image generation, obviously. And so there was a lot of work then also in, well, can we just apply GANs to produce confirmations of proteins? Or can we try to use them to produce protein ligand poses? And for all the same reasons that those models were really tricky to train for images, Mode Collapse was the most famous. Sort of problem, they didn't work very well for for proteins or protein ligand systems.

• 1:07:45Speaker 00

  And we sort of had to wait for the right primitive to get created. And that turned out to be diffusion, which turned out to be a much more useful primitive for the space. And interesting thing is actually a lot of image and video models, some are using diffusion, but some of those have actually gone autoregressive. And so what's kind of cool is right now for people that are interested in really core fundamental AI research, actually, some of the most innovative diffusion research is happening in our fields. It's happening in 3D structure prediction right now. No one would have predicted that then, but now like that's that's kind of a pillar of diffusion, I'd say.

• 1:08:28Speaker 00

  That was unpredictable 10 years ago, everything I just said. And in parallel to that, long before the advent of 3D diffusion models for generative tasks in chemistry, we were building a variety of tools for the problems of drug discovery at hand, some of which were using physics-based methods for predicting potency or even for helping predict certain ME properties. And same with molecular generation, that is like using different techniques to generate new molecular ideas. As Sergey pointed out, there's 10 to the 60 drug-like molecules. Searching that space efficiently is hard, so we've been working on that problem.

• 1:09:13Speaker 00

  And those things happened to be available when we wanted to to take what was then the very nascent area of co-folding, which was clearly exciting, but not at all ready for prime time, not all ready for what? A medicinal chemist who want to use in their day-to-day work and take it to the realm of Useful and in some cases irreplaceable, like clearly superior to a non-coffolding method. And it kind of just happened that we had been building those other primitives. And so they were available to us so we could put them together to build out,

• 1:09:56Speaker 00

  for example, the synthetic data pipeline, or to use some of the inference time techniques. Like, we were, we were ready to do those rapidly because we had been working on orthogonal techniques in ways of approaching problems. So there was some thing here that was contemplative, but like, unless any discovery, like the discovery of penicillin, I'm not saying what we're doing is on the same order of magnitude or benefit to humanity as, as penicillin, but there's always this element of, if you're laser focused on a problem for long enough, and spent enough time in the lab, or in our case, on a computer, banging your head against the wall on these problems,

• 1:10:37Speaker 00

  you can create the luck, as it were, that will enable some of the developments.

• 1:10:42Speaker 01

  How does the lab interact with the development process?

• 1:10:46Speaker 02

  We, as I mentioned before, we train other models, not just structured prediction models. And for those specific models, lab outputs are extremely useful right away. Like potency prediction, right? Those outputs can be used directly for, for model training. But what I'm most excited about going forward is reinforced learning. I think that's coming in our field. And we've seen early signs of it working already with our models. So where you basically put initially maybe physics-based feedback to improve your models through RL, typical RL loop.

• 1:11:27Speaker 02

  But eventually you can go all the way down to the lab in the loop setups, where you're not, your model is not only producing like predictions, but you synthesize based on those predictions and then you measure downstream properties and then you feed back those predictions into that model.

• 1:11:44Speaker 01

  So how do you get enough volume? I mean, do you have automation to do that? Or do you like large campaigns that aren't necessarily automated, but generate a lot of data?

• 1:11:54Speaker 02

  One thing I'm super proud of is our partnership with Insight. They're such an amazing company. They are extremely, extremely good in producing data.

• 1:12:05Speaker 02

  Basically taking the compounds, generating them, creating them, and then measuring the downstream properties and then sending those results back. This is such a partnership kind of born in heaven for us

• 1:12:19Speaker 02

  between Genesis and Insight, where we are able to train models, give predictions and have results back from Insight super quickly.

• 1:12:28Speaker 01

  So your like sort of rollouts include a lab, like a lab iteration?

• 1:12:34Speaker 02

  Essentially, yes.

• 1:12:35Speaker 00

  That's amazing. Yeah, and diffusion's slow, so it's about the same speed of that sometimes.

• 1:12:41Speaker 00

  Give it a week. It is true. It's like I'm a big believer that companies are typically really good at one or two things and do best when they can focus on it. And in the same way, we've been just really doggedly focused on developing the best AI models for drug discovery. Like Insight has that level of maniacal focus on optimizing drug discovery and development. And another sort of trend people like to be talking about is sort of, of course, the rise of China in biotech. It's the elephant in the room. There's no point avoiding it. It's so in the zeitgeist right now.

• 1:13:22Speaker 00

  And where so many Western companies have, you know, had become to rely more on CROs to do a lot of their white lie work. Insight became like so state of the art in terms of the experimental capabilities they had in-house. Their productivity is extremely high. And as Sarah Kay said, it's a match board in heaven for us because what we thrive on is continuous learning of the models. So we wanted to have design, make, test, analyze cycles that are as rapid as possible and continuously fine tune, in some cases, depending, retrain the models based on what we see in the lab.

• 1:14:06Speaker 00

  And so that partnership is one of, if not the first ever, that enables that sort of true joint foundation. Well, training on historical and also perspective data. And as ML practitioners, we all know here that data is is just such a critical input in the whole part of critical ingredient. So it's it's just extremely exciting for us. And that's a range, you know, from reflecting our our conversation so far here, you know, from structure potency and a variety of ad net properties as well.

• 1:14:41Speaker 00

  So so I think it immediately will will improve the strength of our models and and be really powerful for for generally accelerating drug discovery.

• 1:14:50Speaker 01

  And and just we were joking around about the time it takes to do diffusion. But like, I don't know, and I don't know if you can disclose this, but how long is it? Does it take to you? You know, you email them with here's some here's, I don't know how many 1000 compounds, and then they get back to you with the measurements of those realized compounds in what a day, a week, a month, an hour?

• 1:15:15Speaker 00

  It depends. Like, some compounds are really easy to knock out, really known, like a very well characterized reaction where the usual conditions just work. And sometimes it's, oh, this coupling actually didn't work. Like the literature said it would, and we have to try different conditions. So it depends. I'm saying this so the audience understands that for all the claims out there about just like robotic labs, automating synthesis, the reality is just a lot more complicated than that.

• 1:15:46Speaker 01

  Sorry. What are some of the complications there? We've talked about

• 1:15:50Speaker 00

  Radically.

• 1:15:51Speaker 01

  Radical. So they all do some form of automation. What like, and again, not trying to get you to get in some fight with them. But like, what what was the contrary? Take here on, okay, what goes wrong with automated lab?

• 1:16:07Speaker 00

  Okay, so this is also one of the reasons, as Sergey mentioned, we're really excited about reinforcement learning because

• 1:16:17Speaker 00

  It circumvents the problems of having to do very fast design, make, test, analyze cycles from the lab to feedback in the model. Ideally, we can throw more and more GPUs at the problem and have the model self-trained in the same way that we once did for board games in the late 2010s or most recently for coding. Like, that's a key north star for us. However, before anyone tells you that there will be a country of geniuses in a data room just solving drug discovery, there are some problems in our space that don't lend themselves to RL as naturally and that will require experimental intervention.

• 1:16:59Speaker 00

  And so to drive in the specifics of why, to give some examples, and so, you know, coding was always my north star, what I wanted to do with my life, but I spent a fair bit of time myself in the wet lab as a very mediocre experimental scientist before I found my true joy back in, you know, when I got to graduate school and I can just focus, focus on coding all day. The reality is quite messy. So in order to make a new molecule, you have to synthesize it. And what that means is different chemical reagents and catalysts and temperatures and solvents coming together in the right way, in the right protocol, to try to make your new molecule. And then, once you do that, you need to purify it.

• 1:17:43Speaker 00

  If you don't do that, your positive or negative assay read could be an artifact. You need to characterize it. So you need to use NMR and other techniques like mass spec to indicate that what you made in your vial is what you actually sought to make in the first place, which is not as trivial as it sounds either. That's not true, only true for us, by the way. For anyone doing material science for AI, it's the same sort of challenge where, did I make what I actually made is actually a not trivial question at all. It's easier in small molecule drug discovery than it is in materials, but it's still not trivial and it takes, it takes time. We've seen so many cases where screenings of large molecules.

• 1:18:24Speaker 00

  Sounds great in paper because you can test millions, billions of compounds in one go. So you have more shots on goal and, and as a bonus, you have data for your model, right? You just got millions of data points. Wow. Okay. The reality is that the translation of high throughput screens, whether it's Dell DNA code libraries or more traditional screens, the R squared of those predictions to like the actual business of resynthesizing a molecule de novo and doing a low throughput, high fidelity experiment is a shockingly low R squared. That false positive rate is enormous for a variety of reasons. And so that's, it's one of the many reasons why people are so excited about AI making

• 1:19:07Speaker 00

  those predictions because in theory they can be even cleaner than a lot of the white lab work, especially the high throughput work and yield to higher enrichment, higher true positive rates. So.

• 1:19:19Speaker 01

  Why can it be more? Is it just because the, the, the experiments go wrong a lot? And even, and even, does it matter that you have some sort of very precise robot doing it? It's still, there's too many variables. There's too much slop in the system.

• 1:19:35Speaker 00

  It's a few things. So one is that drug discovery frequently involves finding the outliers.

• 1:19:43Speaker 01

  Yeah.

• 1:19:43Speaker 00

  One of the many aspects is so wild about our space is that the properties that one is seeking in a molecule very often anti-correlate. Binding tends to improve the more hydrophobic, the more greasy your compound is because protein binding pockets are greasy. Guess what? That makes your solubility worse.

• 1:20:03Speaker 01

  Yeah.

• 1:20:04Speaker 00

  And then you want to improve your solubility often by adding polarity to your compound. And suddenly, for those that remember high school biology, cells are surrounded by lipid bilayers. Your molecule might not get through the cell membrane because you made your molecule too polar. So. The properties you want often anti-correlate. And that's where you get where multi parameter optimization, the molecules ends up feeling like playing whack-a-mole. So you're often searching for real outliers. And that often requires making molecules that are fundamentally novel. And the kinds of chemistries that today can be automated are fairly constrained. And so your ability to search chemical space broadly for those really top, top Pareto optimal

• 1:20:49Speaker 00

  compounds is actually very limited. So the benefits you get in speed have a very harsh trade off with the actual quality and novelty of the molecules that you can make.

• 1:21:00Speaker 01

  So you can do a lot of boring stuff very quickly. Is that?

• 1:21:03Speaker 00

  If you're lucky, yeah, that would be, yeah, if you're lucky.

• 1:21:06Speaker 01

  But to get to the, to actually the answer to your, the sort of cutting edge questions, you really, it's hard to get a robotic system to do those kinds of things.

• 1:21:17Speaker 00

  Yes, we would love for that to happen. It'd be a huge tailwind for what we do, but that's not, that's not available today.

• 1:21:25Speaker 01

  Now where we're at today. Right. And so, and so, I mean, I know, again, not asking you to disclose anything that you can't, but what is insights approach that makes them so fast and effective?

• 1:21:37Speaker 00

  I think if, if one is willing to focus on one problem and say no to distractions, it's amazing what can happen. And I think one is that their talent density is very, very high. And they're willing to, to what is not universally done now in large or medium sized pharma, which is build so many experimental capabilities in-house. And I actually think that it's possible more companies to try to do that, given the global dynamic of pharmaceutical development that's changed so dramatically.

• 1:22:15Speaker 01

  Right. So like CRO is taking over everything, then now suddenly it's become commoditized and then you, but you're also your capability to do the cutting edge experiments goes down.

• 1:22:26Speaker 00

  It can.

• 1:22:29Speaker 00

  It's been a debate for hundreds of years about the benefits of vertical integration. It comes to all the areas of risk and challenges and upfront costs, but has all the benefits of controlling the process end to end. When it works, you can do amazing things when you do things end to end.

• 1:22:47Speaker 01

  Yeah.

• 1:22:48Speaker 00

  Of course, it comes with greater challenges and I think an appetite needed to do it.

• 1:22:53Speaker 01

  And this kind of brings me to another question I have, again, about more industry. But so you guys, so maybe now's a good time to talk about your pivot to like, so there's name change from Genesis Therapeutics to Genesis Molecular AI. And that that I think goes hand in hand with maybe a pivot in the strategy of the company? Or is that in terms of, okay, we're gonna like, what it sounds like to me is we're gonna focus on the AI portion and then sell that to pharma rather than be a, you know, sort of a drug company with that also sells the tool. Is that, is that sort of accurate?

• 1:23:35Speaker 00

  When we started that company, the origin of the company was fundamental deep learning research methods that we developed at Stanford.

• 1:23:46Speaker 01

  Right.

• 1:23:46Speaker 00

  And the objective we were trying to solve for was how can we have the greatest impact for patients? And at the time, you know, the founding of the company was purely in AI research. There was no real precedent at that time for a pure model company really making it in the realm of just creating value for partners in pharma or really most legacy industries. And the other, I think, component was the company thing founded on AI research.

• 1:24:25Speaker 00

  I think we really wanted to show people that we were serious about the biotech domain. So, you know, I think that's part of the origin of, let's call it some Genesis Therapeutics.

• 1:24:37Speaker 01

  Yeah.

• 1:24:37Speaker 00

  We'll seem a lot more serious. To the, to the, you know, all the tried and true folks in pharma that died in the wool, med chemists, drug developers who we, who we really wanted to work with. So I think there was an element of an attempt at nominative determinism there a bit, but, but also an acknowledgement that I don't think anyone had really shown at that time that you could be a pure AI model company and really make it in, in the space. And that was, obviously, this is 2019. So we were, we had the benefit of being real forebears, pioneers in the space. We had the downside of being pretty early.

• 1:25:17Speaker 01

  Yeah.

• 1:25:17Speaker 00

  Thought to be late, but turns out we're still early innings. As the company grew and, and evolved, we really restructured kind of like a double helix in that we had extremely strong focused AI research and we were fortunate to be able to recruit some of the most experienced and accomplished drug hunters, like people on our management team and board who have discovered, co-invented, developed many FDA approved drugs. You talk to most med chemists, most med chemists are, feel very fortunate if they're able to work on one molecule that gets FDA approval in their whole career. And we were really fortunate to convince these sort of accomplished

• 1:26:00Speaker 00

  drug hunters to come join our management team and board at the beginning and steer the show. So the outcome has always been the same. Everything we're talking about, it only matters if we end up helping patients and their families at the end of the day. And we have felt from the thing that the, the way to do that at scale is to primarily partner with large, medium sized pharma companies and biotech companies. They can do what they're best at, which is novel biology, clinical development. We can do what we're best at, which is AI and bring those two together. In addition to that, as Sergey has said, there's numerous advantages to having in-house wet lab, data generation, clearly, dog fooding the models, getting really direct human feedback from real med chemists, what we need to build in an extremely candid way. And of course,

• 1:26:45Speaker 00

  the third thing is that the single most valuable Thing in the universe, what I like to joke is the the mirror of Arisad test. If you ask people, what is, what is the one thing you most want? And you really thought about it, most people would say a medicine for themselves or a family member. And so there is just immense intrinsic value to having one's own pipeline of programs. And when, when we interact with pharmaceutical partners, on average, they love that we're working on molecules ourselves because it means that we're not just a bunch of keyboard jockeys, but we live the dream that we are pitching.

• 1:27:26Speaker 00

  And we need to use these models in practice for ourselves too. So they're battle tested in an additional way. And we're sort of domesticated in that we, we know what's really difficult about the space. We're not just sort of selling some pipe dream.

• 1:27:40Speaker 01

  It seems to me like there's been a shift in the last, I don't know, year to two years, where suddenly pharma has become interested in buying tools off the shelf. Whereas previously, there are a lot of companies that started trying to build AI and then had trouble. And so they became pharma companies because that's the only way they could make money. Right? Is you raise much money and hope that in 10 years you have a drug and that drug is successful, then you've made it. And otherwise, then you fold or like sell off the assets or whatever. And, and it seems like now that's changing. You have a lot of recent AI acquisitions and, you know, in drug discovery and pathology and lots of different related areas.

• 1:28:22Speaker 01

  And so have you seen this shift? And is that part of the, are you shifting towards selling models? Or is this just a rebranding just because, because of that shift? Like, what's going on?

• 1:28:33Speaker 00

  To finish that, that, that narrative arc there, the, the name change was just to reflect what we were really doing in practice. Which is we are an AI company as we have been from day negative one.

• 1:28:44Speaker 01

  Yeah.

• 1:28:44Speaker 00

  When we were, you know, coding in PyTorch point 17, building the other first, you know, when Sergey was scaling transformers on PyTorch, the first one to do that. And we were scaling graphene nets in PyTorch, one of the first ones to do that. So it's reflecting what, what our identity has been from day negative one. But also alluding to the fact that we are very serious molecule makers ourselves. And we are full stack in that way.

• 1:29:14Speaker 00

  We are also, to be clear, our mission is to create as many medicines as possible as would otherwise not have been created. And I think the way to get there is to put our technology directly into the hands of as many drug developers as possible, not just our own, but in big pharma, midcap pharma and biotech. And that's partially why we've been building our agents, so that our models, which have become so much more powerful, can be easily used and orchestrated in the hands of med chemists, CAD scientists who might have never ever written a line of code in their life.

• 1:29:47Speaker 03

  I find it interesting to hear that you have now started bringing in agents into your development process. You know, it sounds like from the very get go, you, your philosophy is much more, these are tools for scientists and you make the best possible tool. And this lets them optimize whatever pipeline they're going to use. But ultimately, this is a tool for med chem. And now you are switching and you are saying, we are actually going to have automated systems which make decisions and then advance. First of all, you did allude to, you've kind of hit maybe the threshold to, you know, the like last November threshold where it just magically became useful. So maybe that's part of it.

• 1:30:31Speaker 03

  But are there other reasons for going in this direction? And like, you know, why, why now?

• 1:30:38Speaker 02

  Well, let me ask you a question in return. How many tools do you think a human med chemist or CAD scientist can realistically learn and like be proficient in?

• 1:30:50Speaker 03

  Oh, I was gonna say, how many can they learn and how many do they use? Because I've seen that they like to use lots of them, but maybe they're not necessarily good with.

• 1:30:57Speaker 02

  But it's really hard, right? A lot of those tools are extremely complex. They have a lot of parameters that you, but you need to set properly and configure them, right? And this is where agents are extremely good at. You, they actually know how to use all of those tools and how to orchestrate them well.

• 1:31:15Speaker 03

  So I wouldn't interrupt you. When you say agent, then some people like using the term agent to just mean something automated, but other people use agents specifically. Really mean you have an LLM, which is responsible for orchestrating orchestrating decisions. I mean, in outside of biotech, it's almost universally the latter, but sometimes in the space of bio, people are still using agents to refer to any sort of decision-based system, which is automated out of humans. And maybe some of them are even more like traditional rules-based systems or something.

• 1:31:45Speaker 02

  Right. So in our case, I mean, AI definition of agents, an LLM, which is able to use all of the tools that we have been painstakingly building over the many years of existence of this company. And that's a very powerful concept. Because now, as a COT scientist or a mad chemist, you don't need to deeply understand all of the details and all of the hyperparameters that you need to set. And you can just let this thing go and figure out and solve the problems.

• 1:32:16Speaker 02

  And this is where all of the previous questions that we discussed came up, come back as very important one. Like, agents need to be able to understand what those models are producing. Why do we care about crystal structure? Well, the agent can actually look at your predictive crystal structure and make decisions based on that. Something that we believe is super important to iterate and improve.

• 1:32:38Speaker 03

  You're saying that you can take a crystal structure, which I guess in this case is framed as a graph or like a three-dimensional image that MoStar is interpreting, I mean, in some way. And then it it actually can make decisions upon what is binding to what and make hypotheses. Like, you've gotten to that point where you have seen it making effective decisions.

• 1:32:58Speaker 02

  Right. So there are definitely multiple layers to this this question. You can start with like basic image understanding right here. Or you can use additional tools to understand what is happening in the crystal structure. Or you can even train a model which is which is able to natively understand crystal structure representation. And like image modality is a modality for LLMs these days. Crystal structure can be a modality for LLM as well.

• 1:33:26Speaker 01

  I think you can fine tune it on sequences that look like that are somehow capturing tokens that are 3D crystal structures.

• 1:33:34Speaker 02

  Right.

• 1:33:34Speaker 01

  Yeah.

• 1:33:35Speaker 03

  So if you historically have treated humans like you're making tools for humans. So now you have agents. How do you reimagine the interaction of humans and agents and the decision making process to go? Is this you hope to essentially automated humans out? You want to scale up and do more hypotheses? Or, yeah, and then also the kind of the skeptic would ask, how do you also tell? Make sure that your agents are not making weird decisions where they tweaked the wrong hyperparameter? And now you give you the human a bunch of crap.

• 1:34:12Speaker 01

  You don't even understand the parameters well and know that that happened.

• 1:34:16Speaker 03

  Yeah.

• 1:34:16Speaker 02

  It's a great question. And I think the answer here, you can see in how coding tools have evolved. Right? You remember Cursor when it just appeared and it was like top top autocomplete. It was still useful, but not to the extent that it is useful right now where I can fire up Cursor and it just goes and does everything for me. So I think we're gonna see a very similar trajectory here. Now those agents were already useful, but now humans need to provide a consistent feedback and basically steer and guide this agent along the way. But over time, we're gonna become more independent. Still, I don't believe in the full automation, like replacing humans.

• 1:34:59Speaker 02

  I believe humans are gonna become way more efficient in the job that they're doing using these tools. So a human will be providing strategic direction to what this agent needs to do. And the agent is gonna be able to go and execute on the health of a human.

• 1:35:14Speaker 00

  I think this is the best time ever to be a creative human being, or really be any human. Because like the long, the long arc of history, starting from when we decided that eh, hunting and gathering, I don't know about that. I think I'm gonna settle down and, and, and do this whole agriculture thing. I think that started thousands of years of drudgery.

• 1:35:36Speaker 03

  Sometimes you wonder, why did we actually do agriculture? Because everything I've seen seems like agriculture was probably a nut loss on the short term.

• 1:35:44Speaker 00

  It was the only benefit I see in the long run is that if you were a hunter gatherer and you got sick or born with a genetic illness or got injured, that was like, you're in trouble. There's not much more we could do for you. Medicine just didn't really exist. But ultimately, I think the crowning achievement in my mind to civilization is where the sick people among us in society are not seen as burdens, but people who we want to help. And that isn't true in the history of the animal kingdom, including humans. It's a sad reality, but it's true. Whereas now, because of medicine, we see people who are sick as people who we want to help, who we can help.

• 1:36:28Speaker 00

  And we can help them, they could become even more productive members of society, if that's the thing that you care about. To fast forward that to now, I think that so much of creative work, day-to-day work, if you look at it as a pie chart, time is our most precious resource, so much of it was being taken up by repetitive tasks that actually didn't require much thought or creativity. And now, it's like we've created this additional headspace for deep thinking. And for us, you're solving some of the most technically challenging, but important problems, we can use these tools to become wildly more productive and creative. And that goes to the med chemists, drug hunters, cat scientists, who are our direct users,

• 1:37:13Speaker 00

  they can be grand strategists for a drug discovery campaign. And when you have hundreds of drug discovery scientists working 24/7 on large data centers, just the output you get in terms of discovery is just going to be greater, especially if it's used in the hands of expert scientists who know what they're doing.

• 1:37:35Speaker 03

  We probably should have talked about this earlier, but you have some really exciting results from your Pearl model based upon essentially a recent open dataset challenge that you alluded to earlier. Sorry, we didn't talk about it before, but I would like to give you a chance to talk about this. So there's this open bind, it was an EVA721A protease, which was a very, very hard target, had a lot of things which made it hard for the community, or for co-folding methods to generally work. And you ran this internally and I think had some fun results.

• 1:38:12Speaker 02

  Yeah, I mean, we can also go back to our conversation on evals and what's wrong with evals in general. Like, if you look at the open source models and their performance on public benchmarks, it's kind of very similar across the board. And what was surprising when OpenBayne came up with this new benchmark is that you see a wide variety of performance of the models on that. And, well, and part of it is because it's just a single target, okay, I get it. But it's also because this is the target that those models haven't seen during the training. So nobody has optimized for that specific target before. Nobody here climbed on it.

• 1:38:54Speaker 02

  So it was very interesting for us to see how our models is going to perform out of a box on this particular target that we also have never seen before during training or when we developed these models. So when it came up, we obviously wanted to look into it. We ran Pearl role system on that. And we were genuinely surprised and excited about results. Our numbers are way higher than other published openly available models. And to me, it reflects how Pearl actually behaves in real drug discovery programs. Because we have seen this kind of number differences shifts in internal programs or partnership programs before, but we were unable to talk about them.

• 1:39:39Speaker 02

  Because a lot of this data is like, it's either partnership data or internal data, we cannot disclose it. Here you have a perfect example where it's an external target, but we didn't know. And we just run on it. And we see this staggering difference. So that's kind of what makes me excited. And there are some specifics of this target. Like, it has this flexible loop, but needs to move when you insert your ligand into the right location. And a lot of methods just are unable to handle it. And where Pearl was exceptionally good is figuring out how to move this loop. And we are basically correct for every single pose. Like, we're predicting really well this movement.

• 1:40:23Speaker 02

  Yeah, so that's the exciting part.

• 1:40:27Speaker 01

  I'm curious about this. What is it about Pearl that allows it to model a dynamic target like that?

• 1:40:34Speaker 02

  Pearl model, the way it's trained, right? It's trained to predict together protein structure and ligand. And like protein structure and ligand together, they're not necessarily in the same shape or form as the protein in isolation. So the whole training process, it kind of nudges Pearl into that direction.

• 1:40:56Speaker 01

  So are there lots of examples in the training set like that where there's this dynamic nature where once when during the binding event, the confirmation changes?

• 1:41:06Speaker 00

  There is induced fit throughout the training set. Yeah.

• 1:41:09Speaker 03

  This is probably inherited from the physics-based simulations that you're using.

• 1:41:14Speaker 00

  It certainly helps. I think the other component that Sergey alluded to was we haven't just been benchmark maxing on the same. We obviously publish in Runs and Poses. We published the state-of-the-art numbers in Runs and Poses last year. When we published the Pearl technical report with Nvidia at GTC at the end of last year. We use Runs and Poses. We use Posebusters. Pearl does the best on those benchmarks. We did that. However, as Sergey is leading to, the gap from Pearl to next best model is even larger on the partner drug targets that we work on. And what that's in part reflection of is the objective of our company is not maxing benchmarks.

• 1:41:58Speaker 00

  The objective of the company is creating value for biotechs and pharma companies that are usually working on hard drug targets that are different from what's in the PDB. And so we've had to build the models in a way that are able to extrapolate. And the open buy-in challenge was just the latest example of that.

• 1:42:18Speaker 01

  Before we wrap up, we always ask two questions. One is, if you could remove a bottleneck from

• 1:42:27Speaker 01

  Your industry by Fiat, right? So this is not like for inputs, outputs, business. What would that be?

• 1:42:37Speaker 02

  I can say where my main bottleneck right now is is in GPUs.

• 1:42:41Speaker 01

  OK.

• 1:42:42Speaker 02

  Yeah. Look, GPU prices are going up and up and up. And LLM companies are kind of sucking up all of GPU capacity out there. And I firmly believe that what we're doing is genuinely important for humanity, like discovering new medicines for for yourself, for your loved ones. It's something that we all need at some point in our life. So I think it's very important to unblock that GPU bottleneck for drug discovery.

• 1:43:11Speaker 03

  Is Anthropic actually throttling science because of GPU purchases?

• 1:43:17Speaker 02

  So are getting spicy.

• 1:43:19Speaker 01

  I don't know if you can ask this, but are you looking at non Nvidia? I mean, I guess you guys are Nvidia partners or whatever. So but like they're I've seen that some companies are pivoting towards a multi architecture model so that you can take advantage of other suppliers because of the shortage.

• 1:43:37Speaker 00

  Nvidia has been a great supporter of us. They've invested twice in Genesis and we collaborate closely like, you know, not just in terms of the capital, but we optimize kernels together. You know, we cope Pearl was co-authored by Genesis and Nvidia. They publicized the OpenMind results with us. We've been very grateful for that. They're a very, very, very big company and, you know, we're not the scale of a customer like a meta or something. So we're grateful that they've given us time and attention nonetheless. But I think part of that is a self-serving aspect here, which is that obviously like

• 1:44:17Speaker 00

  no one can time time the market, but the amount of hype and investment in the traditional LLM space for chatbots, coding sort of thing, at some point.

• 1:44:30Speaker 00

  The amount of alpha will not be what's desired compared to the amount of investment. And I think regardless of the vicissitudes of the economy, medicines will always be high in demand. And I think whether it's a desire to help humanity or a self-serving interest in looking for the next, the next big thing. I do think that chipmakers, including Nvidia, are going to want to get a lot more invested in life sciences because it will always be high in demand. And the amount of alpha left in pure LLM space is just getting a little questionable.

• 1:45:08Speaker 01

  I mean, all the LLM companies are starting looking into life sciences right now anyway, right? The GTP Rosalind, there's the Claude initiative. There's lots of startups and whatever.

• 1:45:20Speaker 00

  Yep, we think those are all great tailwinds for us.

• 1:45:23Speaker 01

  And then what about you? What's the bottom line besides GPUs?

• 1:45:26Speaker 00

  Oh, I had the same answer as Sarah, okay? For sure, if I could wave a magic wand, I'd love to have just a enormous H100 issue 100 cluster. That'd be, that'd be amazing.

• 1:45:38Speaker 01

  And then the second question we ask is just, do you have a call to action for our audience? So meaning AI engineers plus scientists get involved with something, come apply for jobs. Do like, what's the call to action for them?

• 1:45:55Speaker 02

  We're definitely hiring. We're definitely looking for top talent in this space. And I personally, I transitioned from LLM space to drug discovery. I've never done drug discovery before joining Genesis. I find this field fascinating and extremely interesting. You know, in LLM companies, you often find research scientists excited about working on architectures, but like kind of forced to work on boring stuff. And honestly, LLM architectures are relatively boring. I don't know, I probably alienate half of your audience.

• 1:46:29Speaker 00

  I think they probably think the same thing.

• 1:46:30Speaker 02

  But it's like, it's a transformer layer in the end. Like paper was published in 2017 and you go to any LLM lab today, you will see very, very similar pieces in their models. Yes, there is a little bit of flavor of MOEs, but otherwise architectures are fundamentally very similar to what was discovered back in 2017, right? Well, our architectures and our models are actually very different and very, very interesting to work with. So if somebody is excited about working on architectures, well, this space is actually very, very interesting space to work.

• 1:47:05Speaker 03

  Yeah, I like to comment, not only are they very, your architecture is very different from the LLM space, but you're actually, you're also very different from what a lot of the other bio ML space is doing as well. Like it's, it's sort of unique, both in their subdomain and more broadly.

• 1:47:22Speaker 00

  We saw as a recent guest, you had one of the, the, the progenitors of ESM Fold.

• 1:47:28Speaker 03

  Yeah. Alex Reeves.

• 1:47:30Speaker 00

  Yeah, they cite our model architecture paper that we published in Open Source a few months ago. So the work that you do here at Genesis, as an AI researcher, there's lots of opportunities for publication, open source and influencing the field beyond the obvious ways to influence the field by working with top customers that are actually making medicines that change people's lives. So it's both incredibly intellectually engaging and a lot of opportunities to influence the field and, and get your name out there. But if you want, you could also just stay being a cog in the machine of an LLM company and no shade, no shade. Yeah.

• 1:48:08Speaker 01

  All right. Well, thank you so much for, for coming and speaking with us. I've personally really enjoyed this, hopefully the audience as well. Evan, Sergey, thank you, and we look forward to seeing how things develop. And hopefully you get some good job applications out of it.

• 1:48:26Speaker 00

  Yeah, we, we love the show, so we really appreciate taking time to, to chat.

• 1:48:30Speaker 02

  Yeah, thank you for having us.