Welcome. Dan Brennan, here at the 46th Annual TD Cowen Health Care Conference. I cover tools and diagnostics. Pleased to be joined here with me on stage, senior management team of Seer. To my immediate left, we have Omid Farokhzad, who's the CEO, and founder, and then to his left, we have David Horn, Chief Financial Officer. Gentlemen, welcome. Thanks, Dan. Thank you, Dan. Awesome. Yeah, listen, it's, you know, it's an interesting time to be in the space, right? There's a ton of innovation and excitement. At the same time, you're dealing with some of the headwinds from, you know, in the U.S. and some of the global macro. Maybe just from a high level, Omid, if you want to start it off, just kind of walk us through how you think, you know, last year played out and kind of, you know, expectations and, you know, outlook for 2026, and then we'll dig into some of the details. Awesome. Thanks, Dan. Look, 2025, I would categorize as probably the year of validation for us. Our install base grew substantially. We grew 67%, finished the year with an install base of 82 instruments. It was a year of significant growth in terms of the consumable kits that we sold. Year-over-year consumable volume was up 70%. Then it was a year that our publications went up from about 35-ish total pubs from customers to 70 year-end in 2025. By the way, today we're at 76 as of this morning. Many of these papers were published in journals like Nature Genetics, Nature Aging, PNAS, Cell Metabolism. These are like super high impact papers coming from really, really important people who know the space well. It was a year that what we had predicted would happen happened, which was that for the first time, population scale studies using deep unbiased proteomics would actually happen, and we did three of them. This, you know, Discovery Life Sciences, +10,000, University of Akron, +20,000, and then a biobank that hasn't publicly announced it, so I don't want to name them, but you can probably predict which one it is in Asia that had picked a targeted approach but wanted to also do a pilot with us, and that was a 10,000 sample pilot. They'll announce that, by the way, soon, and when they do, I'll tell you more about it. It was also a year that we launched our Proteograph ONE Assay in June, and that was actually a super important launch. A lot of the traction that happened because of the Proteograph 1. It made it possible to really scale. The runtime decreased from eight and a half hours to four and a half hours, easily fitting in a one day shift. Throughput increased to 80 samples per run. Mass spec time decreased. It really became possible with relatively modest CapEx to do large scale studies. I would say with all of that done, it was also a year that we were, you know, as capital efficient we could be. We decreased OpEx by about 23% despite of everything that I just said. I think it was a good year in terms of what I'd expected. My hope is that the flywheel would turn, where revenue would start to jump, not by, you know, 10%, 20% a year. We're a growth company. I'd like to see my revenue jump by 50%, 100% a year. For that to happen, we've built a great product. We have to now execute on building the market for that product. That's what 2025 was. Okay. Maybe still zooming out, I'd be interested just to get, and then we, you know, dig in a lot of the details here, but maybe just to kind of you set the table at kind of 3% growth at the midpoint. I think total rev, $16 million-$18 million. You as many others, we had 10x Genomics up here a day or two ago, are trying to find innovative ways to get instruments in the field in a tough environment, right? Still for CapEx. Let's discuss a little bit about some of the key drivers for your outlook there, you know, Omid, and maybe we'll touch upon, like, what would be some of the key factors if we thought about taking that 3% up to 5%, 10%, 20%, maybe not this year, but as we look out beyond 2026. Yeah. Maybe I'll open it up, and I'll hand it to David in terms of a little bit more detail, a little bit more color in our placement. First of all, we, when we came to the market, keep in mind that we had exactly 0 publications about Seer from customers. In fact, the first customer paper actually came in 2023. Early on, we realized that in the absence of validation in the hands of the customers, we need to lower the barrier for adoption. We introduced the Seer Technology Access Center or STAC. That's our in-house service business. The goal isn't for us to become a service company. That has never been the goal. The goal was to give the customer an opportunity to experience what the platform could do for them because a lot of these customers with samples, these are folks that are very familiar with sequencing. They're very familiar with antibodies, but the notion of getting unbiased information off the mass spec is something that they're not very familiar with. You gotta show them what the power of that is so for them to become a adopter. We did the STAC. The STAC has been great. 50% of our install base in 2025 were folks that had previously used our STAC. That turned out to be a really good mechanism to get the customer comfortable with the product that we're launching.The SIP program or the program where we offer the customer a loaner to test the instrument out, and this is the Strategic Instrument Replacement Program, and then the customer then chooses to buy or return 60% of our instruments that were placed with the SIP program. Importantly, the SIP requires that a customer make a reasonably considerable upfront consumable purchase. We know that that's going to get used, and a significant number of our purchases have also been in SIPs that have converted into purchases. Both of those decrease the barrier for a customer to adopt. By the way, we've been trying to tackle breaking bottlenecks one at a time. You may remember that the biggest bottleneck in terms of access to unbiased proteomics was initially the upstream workflow, which the Proteograph solved. With the flood of information and content that was being now generated at scale, the data analysis became the biggest bottleneck. We solved that with the Proteograph Analysis Suite and various upgrades over the years that we did. Then the final bottleneck was a detector, and we've been working on that since 2022, and I actually made an announcement then about it, which we can talk about later, in terms of lowering the barrier, and really making an end-to-end solution, letting the customer become super comfortable, with access and content. With that, let me see if David wants to add anything else in terms of what it would take for us to go from the single-digit% up to much more higher traction. Yeah. Thanks, Omid. Dan, as we mentioned on our call, I think, you know, the guide for this year, it assumes the kind of the same end market challenges that we saw on the back half of last year, in terms of the academic markets and NIH and those types of funding challenges that we've seen at some of our customers, and they've relayed to us. I think as I also mentioned, you know, I think we are having quite a number of discussions with the population scale cohorts, and I think that's an exciting development. What we're seeing there is that a lot of them have already run a targeted Olink or Soma, and what they're realizing is they need and want more information, right? They want that unbiased, you know, peptide-level resolution with their datasets. we've always said it's complementary, right? you just learn a lot more. you know, I didn't bake any of that in, or we didn't bake any of that into our guidance just because, look, they're pretty binary and they take a while. we do feel like if we start to get, you know, win some of those cohorts, and we've already won a couple of them, as Omid mentioned, that will help not only spread the word, if you will, but help us accelerate from a revenue standpoint as well. How do you think about just on population, you know, I know on, like, the genomic side, like some of the players there who've tried to push in there, they've really had to figure out a way to get their consumable price down in order to enable it. How do you guys feel about, like, your pricing for the value you're delivering, but to enable the scale? Are there ways to optimize more and, like, do you feel like your pricing is a factor in these population studies and kind of. Yeah. Anything on that front? I'll start, and I'll have David Horn finish it, which is, there's two parts to why you price something the way you price it. One is your cost structure, one is what you believe is the value exchange. You know, our margins are known, Dan Brennan, as you know, over the last few quarters, we've been running kind of in the you know, low 50 is our% gross margin. Our gross margin from the instrument is very little. The margin is really being driven by the consumable, which is a very high-margin business. We have plenty of room to go down in terms of price, because our structure in terms of COGS is actually extremely favorable in this business. The reason we price the consumable the way we price the consumable, which today is to essentially match the existing targeted approaches, that's kind of the way we're priced right now, is because as a relatively new technology, new approach to the market, where you don't get fired for choosing Olink, in order to choose Seer, we don't wannt to create a barrier in terms of price. We price it to be comparable, so the customer uses it. I think in terms of the value exchange, what the customer gets from Seer is vastly more valuable. I mean, I think the ultimate proof point is that, the company that we spun at PrognomIQ, they launched their LDP last November. This is for early detection of lung cancer. Fantastic data. Those biomarkers could not have been discovered with any other platform except Seer. Why? If you chose targeted platform, you would have missed those biomarkers. Some of them are not even available on those panels. If you went to an untargeted approach, you would never have been able to do depth at this scale to discover those biomarkers. There's exactly an N of one commercially available option to do that. Now, that's value. We will capture that value when the market is appropriately educated to what we're offering. Until then, we have plenty of room to compete on price. By the way, that includes the pop studies pricing. I have told every population scale folks that we talk to. Is that while I believe you will get something out of this that's enormously valuable, I will never lose on the matter of price, right? the thing is, in order for a biobank that is sitting on super valuable samples to be able to risk those samples to run on Seer or somebody else, there needs to be a large enough body of data that supports the use of that biological sample, which is precious and it's very difficult to replace. Seer today has 75, 76 publications as of this morning. That is actually a very small number of papers to be sitting at a table with a biobank having these kind of discussions. The other folks got to where we are today with hundreds of publications in front of them. I actually feel where we are to be ahead of the game, to be having the opportunity to do these pop studies, we will not lose them over price. Okay. Maybe jumping over to instruments, I think last year, you know, you placed maybe north of 30, the year before that, roughly 10. Just, you know, what drove the, you know, some of that was through your different programs, right? What, what drove the strength of the placements? A, you've kind of talked about the publication cadence obviously, but just any more color you can provide around that. B, kind of what are expectations for this year? David. Yeah. In terms of the placements, we did have a great year last year in terms of the instrument placements up, you know, dramatically, relative to the year before. As you saw, the install base went up by 33 instruments, which is great. Now we sit at 82 instruments. A number of factors. One, I think the environment did improve a little. I mean, I think if you remember, 2024 was especially rough, especially the first half of 2024. That was certainly good. I think the programs we've put in place, not only STAC, because people, you know, try before you buy, that's been great. About half of the instruments we installed last year, they were previous STAC customers. The STAC is working its intended impact, which is to get data, realize they really like the data, and then to acquire an instrument. The SIP program, again, trying to help customers work within their operating budget rather than a capital expense budget, which for some customers is harder to come by in any given year. That program with the upfront kit purchases, they were able to, you know, acquire the instrument on a loan with the idea that we place it, they start to run it, and then they apply for the budget to then buy the instrument down the road. It's been a great program for us. This year I think we're feeling that those programs are continuing to have an impact. Both the STAC and the SIP, I think will continue to lead to a healthy year in terms of continued placements. Great. How about on customer utilization? How should we think about the trajectory of the pull-through consumables per instrument? You know, could you discuss maybe the average pull-through per user is for a customer who's fully up and running? I think we announced on the call the pull-through was around $113 per instrument. What we've seen is there was such a dramatic increase in instrument placements last year, with that upfront kit purchase, it takes the customer a while before they come back to reorder, right? Because they get the instrument, it gets installed, they run their projects, that upfront kit purchase is, you know, a healthy purchase. They run their projects, it's probably, you know, another, I'd call it six-ish months, six months to nine months before they come back for an order. That tends to be a bit of a lag effect. I think we're going to probably see pull-through in kind of in the similar range this year. Also, you know, we fully expect it to grow over time as people get larger and larger studies that they're going to continue to undertake. We're pretty optimistic that pull-through will continue to be a positive for us. How has the Thermo partnership gone, you know, with the Astral? Just how material about, you know, kind of what is it doing for you? Obviously a huge stamp of validation. Is it driving revenues? Kind of how do we think about the impact going forward? Look, I think the Thermo team has been fantastic partners to us since the very, very beginning. They've been very supportive. Our teams have regular calls. David Horn has regular calls with their marketing team. We did a two-step training for their sales team where they got educated about the Proteograph, the consumables, the value proposition, their sales team, allowing them to be able to also pitch a Proteograph as they're selling an Astral. As you know, the cell cycle of that instrument, the capital equipment is not short. They have actually successfully now placed. Have we disclosed how many or not? No. Well, we've said it's a couple to a few, different accounts that. That they have now placed, including at a large pharma, which is a great account for us. It's been a great relationship then. I'm not expecting the Thermo placements to be a significant part of the way we place our instruments. My own team has to do the hard work. They have to, you know, have boots on the ground out there selling. To leverage the Thermo relationship to place that instrument is a good validation. By the way, as you know. Both times where they launched the Astral and when they launched the Astral Zoom, at the ASMS, they exemplified the power of their incredible instrument with the Proteograph. That speaks volume to the quality and to the depth of relationship that we have with those guys. What is it you think makes it unique for the for the couple of customers at pharma that have adopted the Proteograph with Astral and the significant placement opportunity that, you know, that would reside with other pharma customers who could do it? Like, why are they adopting? What could be learned from those early adoptions that could drive more adoptions? Yeah. Look, I think the number one is use case. Not every Astral is used for proteomics, keep that in mind. I mean, Astral is a Swiss Army knife used for a ton of different applications. That's the first point. Second is the utilization of Seer Proteograph is when you're looking at large amount of contents in a deep and unbiased way. That is a particular customer segment that does that. The final point I would make, Dan, is that for any instrument that's out there that a Thermo instrument doesn't sell, we still go after those customers. In fact, many of the instruments that were placed, the majority of them, lion's share of them, that our own team placed last year, went upstream to an Astral instrument that we placed ourselves. I think, you know, as long as the customer's in the proteomic space, and looking more at a discovery content versus a targeted proteomic, then I think a Seer Proteograph is a useful platform for them to have. The relative incremental CapEx cost of a Proteograph on top of an Astral honestly is in the rounding. Can you speak to the detector and the full workflow, which you mentioned a few questions ago, and kind of what that means? Yeah. Look, let me make no mistake to say this, which is I think the Proteograph fundamentally changed the arc of proteomics. I don't say as a scientist, I don't use those terms lightly. Prior to Seer Proteograph, if you chose to do untargeted proteomic at scale using a mass spec, you were limited. You had to choose between depth of coverage or size of the studies. If you look at the deep proteomic studies that were published, the deepest one came from Broad Institute Nature paper, from Steve Carr's group, where they went to a depth of about 4,500 proteins and a 60 fractionation step. That study was only 16 samples. If you look at the largest proteomics in plasma that was published prior to the Proteograph, that did deep proteomics, that was 48 samples, and that went to a depth of about 1,800 proteins. This is just literally five years ago. Fast-forward 5 years, now we have customers doing tens of thousands of samples at depth of 7,000, 8,000, 9,000, 10,000 different proteins across their studies. The arc of proteomics did change, and we're building the market for that entire new content that is coming. I'll talk about that in terms of AI and why it's important in a second. Thing is, as important as that was, the detector that we have developed, that we've been working on for four years, I've been sitting in this chair, Dan, with you for the last three, I've never mentioned it. The reason I didn't is because as a scientist, we've been only talking about things that we've done because of our credibility in this industry. This was a super high-risk project that may have worked or may not have worked. It worked, and it exceeded any expectation I ever had in terms of what we would develop. The platform will allow the scientific community broadly, not just the experts, broadly, to bring an instrument in-house that on its own will change the landscape of proteomics. It is a purpose-built machine for proteomic analysis. When it works with the Proteograph, then the end-to-end solution will exceed anything that's possible. On its own, it will be the best detector for the proteomics that we have in the market today, and by a large margin. Why do I say large margin? That was a needed target product profile because we are small as a company, almost at the nano scale. What I built needs to be so much better that it's years ahead of any other competitor, and I think we've built exactly that. I'll be sharing data on that instrument the second half of this year. I'm super excited about what it means and what it does. By the way, and I also want to congratulate my team for swinging for defense and really building a fantastic machine. And I apologize, is this detector a mass spec? Like, what is the detector actually? Dan, I have not deliberately disclosed what it is. I'm not saying it is a mass spec, and I'm not saying it's not a mass spec. What I will say to you today is it is a detector purpose-built for deep unbiased proteomics. We will talk about it more, and I'll be able to give you more color in terms of timelines and our development plan. By the way, commercialization strategies with or without partners. All of that discussion will come second half of the year. Have you disclosed how much you spent to develop it? We have not, but if you look at my OpEx, year after year, it has come down. If you look at my OpEx in 2026, it'll go down again. Why is that? How can OpEx go down when innovation like this is actually possible? The reason is extreme discipline when it comes to execution, extreme discipline when it comes to being a steward of our invested capital. We make concentrated bets, not broad bets. We make concentrated bets that have the potential to be revolutionary. The Proteograph did exactly that. I had raised $165 million privately for Seer. When we took the company public, I think, and don't quote me because my numbers may be off, I had spent about $35 million on OpEx, about $8 million on CapEx when I took Seer public. The amount of money we spent to bring that instrument to the market is remarkably low. I think the amount of money we spent to develop this detector, relatively speaking to the eyes of the big boys, is remarkably low. I'm very proud of that. Just zooming out a bit, you've talked in the past about and you said that when you came public, there were 0 customer publications. You had one, I think, in 2023. Now you've got, what is it, 76, right? I think you had said in the past, like, that was the gating factor, and obviously we're dealing with a tough NIH environment. Pharma's not great. Just maybe as a review and preview. You the publication momentum is happening. We spoke a little bit about, you know, where you sit today and what would unlock it. Like, was there would you have expected with this many publications and the awareness and understanding to have seen that inflection already happen? If so, like, why hasn't it? I'll answer, and I'll have David maybe make a broader macro picture comment. I think, Dan, it helped in giving us a seat to have biobank discussions, and the value proposition to the customers became very clear in the context of these very important papers. That is validation that one cannot buy in terms of marketing dollars, right? I think the macro picture is a headwind to our customers and therefore by extension to us. The tailwind is getting stronger and stronger. By the way, the reason or one of the reasons, not the only reason we started working on a detector four years ago was also to decrease the barrier for adoption. This is a challenging space, and we've been picking at it. I'm so optimistic about what the future is because I think once you deliver proteomic content at scale for discovery, literally the world will change. The world will change analogous to the way it changed when you gave it content and the genomics at scale. Let me have David comment on the macro picture. Yeah, Dan, I think you nailed it. I mean, I think the challenge with the NIH funding and some of the pharma budgets as well has provided the headwind that Omid talked about. I can tell you that the customers who do use the technology are thrilled with it, you know, making discoveries. We've seen, as Omid mentioned, PrognomIQ, with their early lung cancer detection test developed on our platform. It's coming. I mean, the science works, the product works great. I think we just need some improvement in the overall macro, and I think you'll start to see that flywheel turn. Maybe we'll have one question on the competitive environment, then I want you to sum up where we sit in the future. Maybe on the competitive environment, on sample prep, are there, obviously you got dia-PASEF from Bruker. I don't know if there are other players out there. What's the state of the competitive environment? Stable, getting better, getting kind of more modest? B, you do have a lot of new platforms, right? I know you guys can look at, kind of post-translational modifications, the SomaScan and the Olink. Now you're even proliferating beyond that. You've got Quanterix, who are going to host next NULISA, and then you've also got Nautilus. There seems to be more product coming. Like, how do you think you fit in the context of the competitive landscape? Dan, first of all, in this field of protein enrichment, which sits upstream to a detector, whether it's mass spec or any other detector, this is a field that Seer invented, and it's based on 20 years of my own academic career. When I say Seer invented it, I don't say that lightly because +200 patents speak to that. In that space, if you are sitting on an island and no one visits you, maybe that's not a very pretty island. The fact that a bunch of people began to copy it, and I won't name them, tells you that's a pretty nice-looking island. That's nice because every one of those players infringes on our IP. There is a time and a moment and place where we'll exert that. When? When I feel that they actually are getting any traction. To date, none of them are getting traction. Why? Because every one of them, performance-wise, is deeply inferior to what Seer has developed. That is being shown by our customers' publications. That, to me, is the best way for me to let the market itself declare who are the winners. I rather spend my money on innovation. When I do that, I constantly move the goalpost forward and forward. As people try to catch up with us, the new goalpost is downstream again. That's the first one I would make. In terms of the competitive market, I have always said targeted approaches are best fitted when you know what you're looking for and you go after it. The company that hit that out of the ballpark in terms of product market fit is our friends Alamar. Why? Because they went after a very specific protein, phosphorylated 217 threonine residue amino acid of tau or phosphorylated 181 amino acid of tau. Not tau, because tau exists in about 1,000 different proteoform, which is relevant clinically. Anyway, that is how you use a targeted approach. I have always loved Quanterix. I thought David Walt company is amazing. I think they're going to do great. I love Alamar. That is how a targeted approach should use. When it comes to untargeted approach, there is exactly an ML one company today commercially that's relevant in that space, and that is Seer. Terrific. All right. Well, I think we're out of time. Omid and David, I don't know, Omid, I think we've covered a lot here. Maybe I'll just give you a, kind of a final floor just to kind of wrap it up if you'd like. Well, first of all, we appreciate your partnership a ton. I'm super excited. The one comment I would make, which I think I've said probably about 20 earnings in a row, is I have never been more bullish about Seer and our opportunity ahead as I am sitting here right this very minute. Terrific. Well, with that, we'll conclude. Thank you again, guys, for being here. Have a great rest of the conference. Thanks, Dan. You got it.