Next up is Schrödinger. Schrödinger is a tech-enabled drug development company. I think of it, known for computational chemistry. You know, a unique or an aspect of your business that I think is of interest and we'll cover is also have a portfolio of proprietary work, you know, through partnerships or otherwise. Yep. I'm happy to be joined by Ramy Farid, CEO, and Richie Jain, CFO. Thank you both for taking the time. Yeah. Yeah, thanks for having us. Appreciate it. Thank you. It's our first time attending this conference, so thank you for the invite. Yeah. I really appreciate it. Great conference so far. Thank you. Yeah. Great. Well, glad to hear it. Yeah. I wanted to start with just talking about the end market, maybe starting with biotech. I think, you know, there's been a lot of attention focused on a relatively strong, you know, funding environment for biotech companies able to access capital. you know, maybe see successful exits as a way for validating you know, more biotech funding. I guess the question ultimately is you know, how is that translating into spending into your business? You know, are you starting to see that? Yeah. Yeah, absolutely. I mean, the contrast between quarters last year and this year are pretty stark. I know a lot of people are reporting this, but the way that's manifested itself is in our 1st quarter, we actually saw new customers being added, you know, versus the opposite last year of, you know, the depressing situation where there were quite a number of companies that were going under and couldn't get funding, right? Or diverting all of their preclinical and research, you know, funding into clinical programs and just shutting down, you know, their research organizations. That's really a big difference this year. That's quite encouraging. You know, just the mood, right? Just the sentiment that things just feel better and the funding, the IPOs, even though that might not be impacting every company, obviously it's just impacting IPOs, but just the, you know, just the atmosphere, right? The tone just feels stronger. The other thing, and people maybe forget this, is, you know, things were pretty bad in pharma too. You know, pharma companies were freezing their budgets and, you know, that's not a thing anymore. We can get back to fundamentals, right? You know, thinking about the impact that the technology is having and actually investing in it. That's a overly long way of saying things definitely feel a lot better and, you know, it's showing itself in actual growth in the business. I'll just add to that. biotech capital flows and formation, the emphasis on efficiency and being lean is extreme right now, and we play right into that. I think if you think about efficiency and how to do drug discovery the most efficient way possible, our software is a solution to that. Ajax just announced our sale to Lilly a few weeks ago. They have single-digit employees at Ajax. How do you do that and get a multiple billion-dollar exit? You work with Schrödinger. We were a big part of the effort there and added leverage to their team to discover a molecule in the most efficient way, in the most expedient way. That's a great point. In terms of the timeline or the trickle down. Yeah capital raise to spending, you know, what is a good sort of expectation for how that ultimately flows through? I mean, it sounds like you're saying you're seeing some spending today. Yeah, yeah. Um, you know, but, uh, is this a Yeah you know, a lag? I think we've put a lot of work into making sure that getting access to our technology is seamless. This is not a long process. We have a layer that makes it possible to run our software on all different cloud instances. We have a huge amount of training material and courses that people can take. Bringing our technology on is actually very quick. Of course, if you raise money and you're telling your investors you need to, you're gonna get to a development candidate in, you know, whatever they are saying now, two years, one year, three years, whatever they're saying, you know, there's a sense of urgency. I'm not sure that there's that much of a lag, actually. Again, we've tried to make that, you know, as easy as possible. So Great. Yeah, yeah. Not like some multi-year thing, right? I mean Yeah. Yeah. You had also referenced large pharma as an area that had maybe been, you know, challenged in terms of spending. Last year. Yeah. Yeah. They're Yeah. How would you describe the environment today in terms of large pharma budget? Yeah you know, applying that into R&D? Yeah. There are clearly, the two things. One is the sort of AI, you know, buzz. They have a sense that there's something they should be doing. I think there's something else that's really interesting, right? I mean, that's good, right? Just in general, you know, we keep reminding people that AI is a technology. You really mean is computation. It's, you know, it's relevant to our whole entire platform. The I forgot. That's the main thing. They, yeah, I can't remember. Yeah. I'll just add. There was something else I was thinking of, but I can't remember what. For us, Q1 results, we were about $28.4 million of ACV. A lot of the growth we saw in the quarter was from top 20 pharma customers. They are, you know, introducing, they're trying out new products that we're rolling out. They're closing some of the adoption gaps that we've talked about in the past couple quarters where our largest customers are $10 million plus in annual spend. Our smallest top 20 pharma customers are closer to the lower end, they're almost $1 million in spend. We're seeing some of those adoption gaps close, and as they're trying to bring through efficiency in their organizations, they're understanding that we can help bring that with our software tools. Yep. That, you just reminded me of the thing I was trying to say. That was really important. There's now a recognition that there's a big disparity between what pharma companies are spending on technology. This is interesting. It's not a situation that's gonna stay for very long. The fact that you've got, among the top 20 pharma companies, that you have some spending more than an order of magnitude more on computation, is something that now the companies that are not spending that much are recognizing. I think it's because these pharma companies that have now embraced technology at scale, they're using these physics-based methods that are being amplified, you know, with AI physics plus AI. You know, they're using these technologies. They're going out and starting to talk about the impact that that's having in public forums. There's this sense, of course, of the companies that aren't doing that are being left out, and they're frustrated by that. I think that's going to drive growth because, of course, it can't be the case that you have one pharma company spending, let's say, $1 million a year on technology, on software that's driving drug discovery, and another pharma company spending well over $10 million. That isn't gonna stay that way. It's interesting, you know, that you would think of all the talk about computation and everything, that sure, everybody's using it, but it turns out that's not the case. We're still-- I mean, this is kind of an amazing thing to be saying, but we think we're at this early phase of an inflection where computation, AI, physics, you know, whatever you wanna call it, computation is really truly adopted at scale to drive drug discovery projects. AI is helping, right, to attract that attention, but I think the fact that there are some companies that are really leveraging it and seeing the impact, right? That they're getting to development candidates more rapidly and with less cost and higher quality molecules, that's starting to get recognized. When that happens, then you've got FOMO and all this sort of thing, and I think you start to have the real adoption, you know, at of technologies at scale like you see in other industries that have been completely transformed by computer-aided design. We're behind. I mean, we as in the drug discovery and materials space, but that's just a matter of time. We know where this is headed. Well, that's exciting to hear. Yeah, we're excited about that. Yeah. You know, I think my follow-up question would be, you know, question we get a lot is: How is the conversation going with large pharma in terms of, "Hey, we want to invest in more of this computational chemistry." We're either gonna invest on it on an internal basis, or perhaps we're gonna work externally with partners like. Yeah. Like yourself. Like, yeah. Okay. It's definitely both at the moment. That's a great question. They are determined, of course, to bring this technology in-house and drive their projects. One of the things that is new, which we're really encouraged by, is they used to when they wanted to collaborate with somebody, they would say, "All right. We've got this crazy project. There's no way it's gonna work. Let's hand it over to somebody. We don't have to pay anything up front, and if they succeed, there'll be some milestone." That's changing. Now, Schrödinger is working on projects that are their high priority projects, their must-win projects, the ones that are highly competitive. There are substantial up-fronts now, which is a sign of confidence that it's going to work, right? If these were all back-ended deals, you know, that probably means they're not sure it's going to work. That's really encouraging that they're taking these collaborations seriously, working on high priority projects, investing in them up front, you know, and then there's still substantial milestones as well. That's a really good sign. They're doing both, right? I mean, they're gonna continue working on their own projects. Actually, there's one other thing. They recognize something else, that the technology's complicated and hard to use, and they need to understand. They need know-how transfer, right? They need to know how we're doing it. We tried one thing, which was, "We'll just teach you," but that didn't work. Now what they're doing is they're using these collaborations that we're working on with them in some sense as a know-how transfer. They see how we're doing it, and they're saying, "Oh, okay. That's how you deploy the technology at scale. That's the kind of expertise you need." Bunsen, by the way, our agent, our AI agent, maybe we'll talk about that later, I think is gonna help here. They're using the collaboration in some sense as a way to learn how to deploy the technology at scale on the hundreds of other programs that they're working on that we're not collaborating on. That's another interesting development. Okay. A question or maybe a thesis that I've been, you know, that I'm consistently encountering is, you know, AI is going to generate new discovery at a pace where we should see a fairly rapid and, you know, onslaught of new drug candidates and, you know, is the system really prepared to handle all that volume? I guess the question is, you know do you agree with that? two, where would you say we are in that kind of evolution? Yeah. Absolutely we agree, for sure. I mean, the whole point of this is that we are going to, I mean, we're already seeing this. The projects that we're working on where we deploy the technology at scale are resulting in a much higher probability of success. The number of programs that actually get to the clinic and that make it through the clinic with high quality molecules is way higher than the industry standard. That's happening now. I think you're asking, can the system handle it? I'm sure they'll figure it out. This is what we've been trying to do for a long time. They will, these companies have massive resources to run clinical trials and they know how to do that. I don't think we're gonna have a problem where there's gonna be too many development candidates, too many high quality development candidates. Let's welcome that problem. I'm pretty sure the industry can solve that one really easily. And in terms of Yeah innings, where we're at. Where would you put us? Okay. Innings. That's a tough one. I'm probably not gonna say what inning, but I'll try and describe it sort of qualitative because that gets you into trouble because it's so quantitative, you know. Let me just say where we are. I think we're in the early days. That's the key. The number of programs that the pharma industry is working on where they're deploying this technology at scale, and I mean really at scale. That means exploring hundreds of billions of molecules and not relying on, in any way on traditional methods is still relatively small. There's enough successes now. I mean, the Ajax Therapeutics acquisition is such a beautiful example. I think that's an example, the acquisition by Takeda of the Nimbus TYK2 program is such a good example of this, where it's very apparent what the result is relative to not use, you know, the result of using the technology at scale relative not doing that. I think that's just starting now. We see the opportunity really ahead of us, and the way we've said this is, you know, our software revenue is around $200 million, ACV, around $200 million annual contract value. We think that is really far away from the TAM, and that's a way of saying we're in early innings, obviously. I don't know exactly which 1. No. It's early innings. That is helpful. Yeah, yeah. I wanted to talk about Bunsen agentic AI. Yeah, yeah. You, you mentioned or you had announced, I think an early release this summer or early access this summer. Yeah, early access, yeah, this summer. Any color on that rollout? Sure, yeah. you know, which customers. Yep Any detail would be helpful. Let's explain what this is. One of the biggest barriers to using technology like this, as you can imagine, is that it's pretty sophisticated. Even expert computational chemists really struggle with technology like this 'cause it's sophisticated, there's a very complex workflows, it's hard to run these technologies. You're running, in some cases, technology on hundreds of thousands of processors. It's hard to provision those sorts of compute resources without a lot of help. This is one of the big barriers, we think, to large scale adoption of the technology. Now the advances in LLMs are just extraordinary. They're so powerful that it has a good understanding of obviously the interface between us, our language, their understanding of it. What it's missing is, of course, knowledge of all the science and the physics and AI and so we've built on top of a large language model, you know, all the skills required to actually run this technology and do it in the right way, and do it in the way that, you know, this know-how transfer I was talking about before, well, now obviously Bunsen has that know-how. We have previewed this to our largest customers, and the response is overwhelmingly positive. I mean, it's sort of, you know, this is what they've been waiting for, and it's very easy to develop agents. That's not the challenge. The challenge is an agent that actually understands our software and our workflows and what our scientists do that makes them so successful relative to the industry, you know, average. That's what it is. Lots of positive feedback. We will release this in the summer. I'm sure we're gonna have a lot of it'll be oversubscribed, the number of people that we can support, and once we start getting feedback, we will figure out the best way to monetize it, which will be, because we have throughput-based licensing, our whole licensing model is throughput based. The more agents you have, you don't have less, you know, if it was seat-based, that would be a problem, right? Because it could potentially replace people. This just powers individuals. One person can now do the work of 10 people because they can just run way more software. We see that as a very effective way of scaling up the software business through just essentially more consumption of the licenses. Okay. Does that make sense? It does. Yeah. Wanted to switch gears, talk a little bit about predictive toxicology. You talked about market receptivity being pretty positive. Yeah for this offering. Yeah. Any color on the size of this pipeline? Yeah how that's gonna impact. Yeah ACV. Yeah. One of the most exciting things about this is traditionally toxicity, screening, assaying, you know, toxicity is done later stages. Typical project, you work for a long time, you try and design a great molecule, and you can't do those very expensive tox studies in animals until way later. Way later. It's actually, it's sometimes, like, three years into a program, right? You think you've got a great molecule. It's potent, it's selective to the targets you know about, it's soluble, it's permeable, and then, and then you, you have this scary moment where you say, "Okay, now we gotta start doing some animal studies," and you put the molecule in the animal, and you have some surprise. That's so many programs are like that. What's really exciting about the solution that we came up with is this computational method for predicting toxicity associated with binding to off targets. The binding to targets that are known to cause toxicity. The key is we've moved that now because you have a computational method, it can be moved way up early in discovery. In other words, you don't have those surprises three years in. It's introduced into the whole multi-parameter optimization problem of discovering drug, so that by the time you get three years down the road and you have a development candidate, you have a much better idea of the toxicity profile. That's what companies have latched onto. They're realizing they can bring toxicity screening earlier in the project, which of course means you're running a lot more, you know, a lot more, right? That's, like, consuming a huge number of licenses to do that. That's what we're most excited about. That, of course, I should have mentioned this in the first place, the actual results. It's accurate. It's doing a really good job of replicating what the experimental screens do. It can identify whether a molecule binds to hERG or to cytochrome P450, or some nuclear hormone receptor, or some kinase that you don't wanna be hitting that's associated with toxicity. Really a lot of excitement and we think it's going to contribute to growth this year, even though it's its first year of release. Okay. Yeah. I think we were speaking earlier about NAMs. Yeah you know, predictive toxicology being a NAM and the implications for reducing animal testing. Animal testing, right. Have you quantified what the implications are for your offering to animal models? Yeah, yeah. Quantify, you know, probably not rigorously. Yeah, it's early days, but there is no question that because of work, of course, we've been using this technology internally in our programs for a long time, that we are putting cleaner molecules into animals, and what that means is we're not having to keep using animals to test our toxicity, right? It's validating a toxicity, right? You're putting clean molecules in. It's already definitely, the problem is I can't give you a quantitative number, you know, we're doing less of these, and this is what caused. There, but qualitatively, no question about it's resulting in cleaner molecules, and I think that's why pharma companies that have been evaluating this, I think they're seeing, you know, that potential. Give us a little bit of time. I think it's a good question because we should be able to quantitate it. Directionally, it's going in the right direction, but that'll be a really fun statistic to have where we can actually say somehow this project tested fewer molecules in animal. Yeah. Quantitatively, another way to think about it is the avoidance of a late stage failure. Yeah, exactly. It's, you know. Right years of opportunity, time, and dollars spent on programs, testing while you're discovering the molecule versus when you're at the end of the journey. That is the value of this program. It's what kills most programs. Right. Right? That surprise at the end. Just one other clarification on predictive toxicology. We've released a product, the discovery work is not over. We're still going to continue to add more targets into the panel, this work will be continuing for years to continue to make the panel more expansive and broad. Okay. One last one on this topic. Sure. You're pulling forward some of the toxicology testing essentially, right? Right, exactly. How about the reduction of time or speed to, you know, getting into the clinic? Oh. Is that the other c ompletely. Yeah. That's the right question. Absolutely. What it definitely does is it increases the probability of success, right? Cause you don't, you don't have that surprise at the end, right? Where you made a molecule and all of a sudden you test it. Then here's the other thing it does. I should have mentioned this early. In traditional toxicity screening, what happens is if you have a molecule that's toxic, that's it. Program's done. I mean, you'd have to go all the way back to the drawing board, right? Start all over again from a new scaffold, a new molecule. Usually that's the end of the program. Programs just simply end. What this is doing is not only telling you whether you have a potentially toxic molecule, but it's telling you how to fix it. The program's not dead. If it lights up and you see that you're hitting an off target, you dial it out while you're maintaining all the other properties. That's really the impact. There we go. I wanted to touch on, I mentioned at the beginning, the proprietary assets, the portfolio w ith the collaboration portfolio. You know, maybe if you could just share, you know, what are the next milestones investors should be paying attention to? I think you had mentioned 15 programs, about $5 billion potential milestone payments. You know, what's sort of that next milestone? Yeah. We touched on this briefly, but one of the equity stakes we have is in HX Therapeutics. When that upfront portion of the deal closes, we'll receive a cash distribution from that transaction. We ended the quarter at $406 million of cash, but you should expect some additional cash coming our way from that deal closing. If you think about the remainder of the portfolio, 15 programs on which we have royalties, those range from low single digit to low double digit royalties, and $5 billion in total milestone opportunity, that's not POS adjusted. The way we think about that is growing that portfolio, and having, working with our partners to have those assets advanced through the clinic. Those assets now have reached phase II and phase III. I think it's a further demonstration of our long track record of success of getting assets into the clinic and having those advanced assets advanced through the clinic and survive. Got it. Um, I think Wow. Yeah. That's amazing. In fact, good timing. That does it for time. Ramy, Richie, thank you both for joining. Yeah, thanks a lot. Thanks. Great discussion. Thank you. Thank you. Appreciate it. This was great. Appreciate it. Thank you. Thanks a lot. Appreciate it. Very helpful.