Hey everyone. Good morning. Thanks for joining us on day two of our healthcare conference. I'm Tejas Savant, I cover the life science tools and diagnostics sector here at Morgan Stanley. Delighted to have Olink join us today, representing the company, we have Jon Heimer, CEO, and Oskar Hjelm, CFO. Welcome, gents. Before we get started with the Q and A, I just have to read a quick safe harbor. Please see the Morgan Stanley research disclosure website at morganstanley.com/researchdisclosures. If you have any questions, please do reach out to your sales rep. With that, Jon, given that you're in your first year as a public company, I thought it would be great for you to just set the stage and give an overview and a background of the company and the technology for people who are not as familiar with Olink. Great. Thanks very much, Tejas. Let's do it. We actually see the very same strong rationale now as we did when we kicked the company off five years ago in 2016. The community sees some really large challenges in drug development and healthcare. For example, the top 10 selling drugs, only one out of five patients respond well to those. Basically, 80% of patients have no effect or only side effects of those drugs. Cost in drug development is skyrocketing, and failure rates are very high. I think we all as a community had really high hopes that genomics would help us to solve many of those challenges. 10 years, 15 years into genomics, it perhaps hasn't really delivered as we hoped. Looking at biology and how it works, it makes a lot of sense to add proteins to that mix. Obviously, proteins are closest to our phenotype. They drive all the biological processes in the human body. They are dynamic between health and disease in real time, and obviously the target of most drugs. It's just that proteins are so vastly more complex than genomics, so technologies hasn't really delivered. That's basically what we saw, and with our unique, disruptive, proprietary PEA technology, we thought that we had a great opportunity to actually overcome those challenges and help the community to overcome those. We are super excited to be five years into that. We've had a fantastic growth over these last few years and building our customer base around the world rapidly and so forth. That's sort of the backdrop and the background to what we're doing. If we take a more narrow perspective of just in the past six to 12 months or something, what's going on. There we are super excited. We have just recently transferred our platform also to Illumina's NGS platform. Looking at the installed base that we can target, we've gone from 500-5,000 just in the past months, which also presents itself with a very exciting business opportunity for us moving forward. Also how we're extremely aggressively and rapidly building out our offering. We know that our customers want to look at many proteins simultaneously with very high quality. Just in the course of the past, say, 15 months, we've grown the library from just over 1,000-3,000 proteins. Lastly, also, what is happening as we speak, we are right now, as I'm sure you remember, that we were selected by 10 major biopharma companies to be their proteomics partner for a very significant population proteomics project with the U.K. Biobank to run through 56,000 samples. We are, as we speak, running through those samples and be starting delivering data here in the next few weeks and over the course of the next few months. That is, I think, super exciting for our field and for the whole research community as well. As we all know that all of those data will be public available after the biopharma companies have exclusivity for nine months. I truly think that all of these things will really spur this market and the scientific insights and hopefully really contribute to sorting the challenges out that I mentioned initially. We are very excited to be here and where this field is at and going. Got it. That was great, and lots to unpack. Maybe just to start with, in terms of just the market itself, right? One of the questions I've been asking all the proteomics companies at our conference is the why now question. What gives you the conviction that customers have the appetite today for the slew of new technologies that have emerged over the last 12 months? Yeah. Very relevant question, Tejas. Perhaps I don't really look at it that way. I really think that if we go back, the first true gold standard technology that was developed in proteomics was the ELISA, which by the way, also was a Swedish invention, but from 1971. I think we, as a scientific community, have known throughout that period of time that proteins are vital to biological function. I honestly think that we've been trying to solve those with different technologies and methods throughout that period of time. We have Meso Scale Discovery, fantastic platform, but perhaps in the smaller plex. Luminex is another type of technology. Mass specs have been used widely for a long period of time and trying to prove, but still haven't really delivered to where I think the scientific community want this technology to be. Here, actually, I think that our Proximity Extension Assay really has overcome all of the challenges here to be able to look in a very high multiplexing with basically a single-plex quality in a very high throughput setting in a cost-effective manner. I think actually those are the things that are, when we see that happening, I think scientists are now ready to really jump on the proteomics bandwagon and to contribute to the scientific questions we just talked about. Got it. Can you help us think about, from the technology standpoint, the focus on throughput? For how many NovaSeq, how many samples per NovaSeq run, and how many data points per run are we really looking at? Yeah. If you look at the theoretical maximum, we know that a researcher can actually create up to 40 million data points per week per system on one NovaSeq. That is pushing it to the limits. It's a very important point to raise, because, to be able to take proteomics to where genomics is and has been, throughput is really important. Going back to what we just talked about, the U.K. Biobank project with 56,000 samples, if we would've tried to do that with a mass spec strategy, it would've taken actually decades to run through all of those samples. Here we are in a much, much more hurry than that and need to have this data fast to be able to act on them and drive the science forward. Got it. How do you go about selecting targets for a panel, Jon, and how quickly can you validate a target? Yeah. Very important question as well. Basically, we are listening very closely to the scientific community and the experts here. What we hear them say is that the primary focus, and the lowest hanging fruit in proteomics will be from proteins in circulation. Hence, we are very much focused on the proteins that we hope we can find in blood. If you look at from a biological standpoint, which proteins will show up in blood, it's going to be basically from three major buckets of proteins. It's going to be proteins that are actively secreted into circulation, so secreted proteins. It's going to be tissue leakage proteins from basically all organs that will leak into circulation. Then it's going to be the family of inflammatory proteins. We are very carefully working through that together with the experts, and our aim is to very rapidly come to a comprehensive, complete perspective of coverage of those functioning native proteins in circulation. That is our focus. Got it. Talk to us about post-translational modifications. That's a question that comes up fairly often. It comes up in two contexts, right? People come at proteomics from a genomics lens, at least investors do. They look at sort of all the genetic variants out there, and then they look at sort of post-translational modifications, and then they wonder whether targeted approaches will be able to ever handle that. If there's a limitation in terms of your theoretical maximum size of the panel or whatever, given the number of PTMs out there, how will you stay relevant? Yeah. Actually, when we talk to the top scientists in this community and our most strategic customers, they tell us, "Well, don't prioritize on PTMs right now. That's going to be the corner case study. We need to look at the functioning, circulating proteins first. Don't put too much attention on the PTMs at this point in time," is their recommendation. Will our technology be able to cover PTMs? Absolutely. We're really listening here to what our key customers and scientists, experts in this space tell us, and focusing more on those circulating native proteins. Got it. When you think of your key customers today, can you just give us a sense of the mix between core labs, academic labs, and biopharma labs? How do you see that mix in, let's say, 2025? First of all, if you think from the broadest perspective, and Oskar, if you want to fill in here, but if we regard the core labs, as we say, as academia, we have a roughly 50/50 mix today between academia and biopharma. When we do our market analysis, we think that this space will be much driven by the multi-omics perspective on the next-generation sequencing instrument. It's going to be users that look at DNA, RNA, and proteins on a sequencer, and that will be driven from biopharma. Honestly speaking, the academic community is very excited and interested about this as well. If directionally that perhaps biopharma will grow a little bit stronger, and that the [percent] mix will shift a little bit in their favor. Got it. Switching gears a little bit to the competitive landscape here. One of the questions we get fairly often is around your sort of most like for like competitor, i.e. SomaLogic, and they've made certain sort of changes to their go-to-market strategy and the cross talk among their aptamer approach has also gotten better over time. Are you starting to see them show up more often in sort of a competitive bid situation? There is sort of one of their next steps or pipeline projects is to perhaps launch a kit strategy with an NGS readout. Just wanted to get your take on that. Yeah, maybe it is a little too early then to see a strong, big difference. First of all, I think we should take one step back and say that proteomics will be or is a very, very big market. If you compare it to genomics, where you do your genomic profiling on an individual once, whereas in proteomics, we need to do it on a longitudinal perspective, if it is once a quarter or once every six months, whatever it is going to be. I think the proteomics market will represent a larger opportunity than genomics. Hence also that proteins are so vastly more complex than the four letter DNA, that I am convinced that it will be not one winner takes it all, but rather certain technologies might be better for certain type of questions, and so forth. With that in mind, when we listen to our customers and what they want to do, as I mentioned, they want to look at many, many proteins simultaneously in circulation with single-plex quality. They're very much interested in mechanistic biology to see how diseases are developed or how drugs interact with those diseases. We don't think that it's going to be 5,000 proteins that you need to run on every patient all the time, right? Is it a subset of proteins? Is it going to be five, 10, 15, 20, whatnot? I don't know. It is a technology that measures functional proteins that is scalable, where you can trust the data and move from that very high plex down to a lower plex and move it into clinical decision-making. That is how we have developed from our technology, our offering, and our product portfolio, which we feel resonates very nicely with what our customers are asking for. Got it. Walk us through your plans for the clinical sort of market, Jon, because, obviously it's nascent today, if anything, but over time, generally the way this plays out is that the clinic becomes a more dominant part of the overall TAM. Where does the market stand today, and what's your strategy, specifically around even, SomaLogic, for example, has a larger in-house database because of that customer data-sharing requirement that they had historically in their business model. What's your sort of go-to-market strategy for the clinic? Yeah. Right. We tried to be very clear from the start that our vision here is really to enable the scientific community with the understanding of real-time human biology. With our scalable platform, we can take it from that very low plex into or very high plex into a very low plex with the exactly the same robust quality. Obviously, we are super excited that our most advanced customer, being Octave Bioscience, will move our technology into clinical decision-making in the U.S. in multiple sclerosis already this year. That is our clear strategy. We hear very clearly from our customers that they're not doing this for fun. They just don't want to cast a broad net. They want actionable outcomes to drive the science in a meaningful way forward. That's exactly what we're doing now. I also have a huge respect that this science and these processes take time. I'm super excited that Octave is here already. Also, as we look ahead, I try to be somewhat conservative in how fast this will happen. Over time, do I think it will happen? Absolutely. Yeah, that's been and is our approach. Got it. The other question that we tend to get as, not just you and SomaLogic and Quanterix, but even some of the unbiased companies out there that are still in varying stages of development, such as Seer and Encodia, Quantum-Si, Nautilus, et cetera, come to market. Will customers sort of struggle with the concordance issue across platforms because everyone's using their own approach? You might all sort of detect an overlapping but not sort of completely overlapping set of proteins in a sample. That might create sort of like confusion or because we've seen that sort of play out to some extent on the genomic side of things. For example, between tissue and liquid biopsies, and it took a couple of years for customers to wrap their heads around that. What's your take on that? I actually have a huge respect for our customers and their expertise. When one of our customers, before deciding which technology or platform to use, they do very, very sophisticated comparisons. They have a set of samples that they might spike in certain proteins into certain samples. They might deplete proteins from other samples. They run them in duplicates and triplicates. They run several technologies with highly validated single-plex assays and compare to these high-plex technologies. Basically expecting single-plex quality, even if you're looking at three, four, five, whatever number, 1,000 proteins you're looking at. Hence, I think their decision-making, their process to evaluate and decide how to move forward is based on pure data and solid science. For them, I don't think that is too much of a difficult question. They're going to run through that type of analysis and decision-making before choosing which platform that fits their science and research in the best possible way. Got it. I want to talk about the service versus the kit mix. It's sort of skewed a little bit towards the service side. I'm sure part of that is the PopSeq stuff, which we'll get to in a minute. In your mind, is that still tracking versus your expectations, at least at a high level? Can you, on a price per sample basis, talk about the differential between the kit offering versus the service offering? Absolutely, yeah. For us, it's very important to respond, yet again, to what we hear and see from our valued customers, and that they really would like to run the science in their own hands, so to democratize this technology. Hence, we're definitely driving our business towards the kits business. On the other hand, it's also fantastic, I think, from that customer perspective, to also that we offer people service, whether it's virtual companies or if you want to try technology out and so forth, it's super helpful and easy to send some samples and get some data back. As we rolled out the kit offering and now most recently, at the end of the first quarter on Explore on NGS. Today, we had some early access customers from last year. We now in total have 16 labs that are running Explore in-house, which is great. If we look at the pipeline we have, it's many multiples of those 16. Mostly impressed or sort of excited, I guess, when I actually see the names on those lists, because it's really the who is who in our industry, in the biopharma and academic community. Directionally, where we're heading, we are in a fantastic place. We are very excited about these most prestigious researchers really want to bring our technology in-house. Do I think we'll be where we had planned to be by the end of the year? Yes, I do. How fast is this going to go? I'm not sure exactly. Are we trending in the right direction? Absolutely. We feel very, very good of where we're at today and where this is evolving. Got it. On the price per sample differential? Oh, you. Oskar, you want to? Yeah, sure. I think looking at, was that comparing Explore and Target or kit versus service, Tejas? Yes. kit versus service. Yeah. I think looking at the price per target, a little bit depending on the size of the study, but I would say roughly factor of three between, service and target service and kits. I think looking at our cost of sales for the kit side and our quite dynamic cost of sales with a big part of royalty, we can really sort of drive kit pricing down to a low level without hurting our percentage gross margin. We have that ability to democratize proteomics and to get as much volume as possible onto the platform and making the technology available to researchers. Got it. While I have you, Oskar, math question on the $600,000 sort of pull-through that you communicated. Can you just walk us through how you arrived at that number? I believe you'd used a sort of to-date installed base, so to speak, to get to that calculation. If you can just clarify that'll be helpful. Yeah, sure. As Jon mentioned, we have 16 installations at the end of Q2. We started with some early access customers end of Q3 of last year. We're looking at the total revenues that we've generated on Explore kits in the last 12 months and dividing that by the number of installations. It's a rolling 12 month average on the installations. Got it. Clear enough. Perfect. Jon, going back to you. Invitae recently bought Genosity. That was one aspect of your U.S. expansion plans. How are you thinking about Explore capacity and U.S. expansion in that context? Yeah, no, maybe fortunately or unfortunately, but it's a small blip on the radar, really. We're talking to so many different core labs and CROs and stuff, yeah, it's not a major impact, really. Got it. Shifting gears to PopSeq, Jon, can you walk us through the selection process for the U.K. Biobank? That's something which has come up fairly often. What are the metrics that they evaluated you as well as other proteomics companies on, and what was it that made them decide to go with Olink? Yeah. No, it's a great question. I really think it's better asked to the consortia of biopharma companies. We know that they obviously did a very thoughtful and careful evaluation of the various proteomics platforms out there. It was a matter of quantity and quality and service level, I guess. Here, I imagine it couldn't have been a super easy process either because 12 major biopharma companies with a lot of strong minds, I assume. A very thoughtful comparison on the science to choose a platform which they thought would provide the most meaningful and important data for their scientific aims, which in this study really is twofold, identifying new and novel drug targets and predictive biomarkers of response for such targets. I guess in the whole mix of that, Olink was selected as the most valuable platform for those type of research questions. Got it. On deCODE, you have a pilot project with them underway. How are you thinking about follow-on work from that? Yeah. Also the best person to answer that question is deCODE. Seriously, they run, in our minds, a pretty significant pilot study here across 10,000 samples to evaluate our platform. It's obviously a significant undertaking. We see this as a very, very valuable partner to us as well, because deCODE are truly experts across both genomics and proteomics. I'm sure we, as an organization, will learn from this project as well, and we're working very closely with them to see any feedback and so forth that they can provide on us as a company or our technology or set of products. First things first, get through this pilot, and then hopefully they will think that the data looks beautiful, and from what we've heard to date is that they really appreciate the quality of the data. You will take next steps from there. Got it. Talk to us about your PopSeq pipeline now that you have the biobank and the deCODE work underway. What does that funnel of conversations look like today? As it translates to the financial model, do you expect an element of lumpiness to come in, at least in the near term, as more and more PopSeq projects onboard with Olink? Yeah. No, it's exactly to your point, right? I think, looking back over time as well, that we've done a lot of these, when you say PopSeq, really we're talking about genomics, right? We've done a lot of that research over the past, say, five, 10 years. Hence now, where the science is at and what we need to do to improve the data we have to make better decisions moving forward, I think that community, exactly to your point, wants to now add the next [omic] here and proteomics to that. We are in active discussions on those type of initiatives, on really all the three continents, so both in North America, in Europe, and in Asia. I don't know about the lumpiness, whether we should expect that. In my experience, these discussions, one, they take a bit, it takes a while. Two, the projects are big, so they're also going to take a while, and so forth. Maybe I don't really expect a lot of lumpiness from them, to be honest. Got it. This question just came over email, I think it's an interesting one. Is that an opportunity for you for a tighter integration with Illumina, specifically as it relates to some of these population omics projects? Clearly, you sit upstream to NovaSeq, it sort of plugs in well into that sort of high throughput requirement for a population sequencing project. People generally think of a short-read technology being paired up with a long-read technology to do population sequencing work. In this case, if you can pair up with them for a proteomics project, that would be pretty interesting as well. Just wanted to get your take on that. No, that's absolutely something that we do. I think both organizations really appreciate the opportunity, not only for science and for the scientific community and the societies in a broader perspective, that these experiments are super important to healthcare. We definitely see that, and we are already pairing up in those kind of discussions. Absolutely. Got it. Switching gears to new products and your product pipeline, Jon. Can you just elaborate on why you felt a need to introduce the Focus versus the Flex panels, and what the customer reception has been like so far? Yes. No, but as I've commented on a couple of times here in our discussion, I think the Focus panel is really that you can select any number of proteins, basically, from the larger, the complete library into a custom-developed product. To just to meet that need that customers have to cast a broad net, find the novel proteins to drive the science forward, and then being able to solely look at them with the highest data quality. For the Flex, it's rather, let's say we call it the more mid-plex market, so when you're not looking at thousands of proteins, but say, 50, 100, a couple of hundred or something like that, where we know that basically the 50% spend in that market is on inflammation. I do believe that we sit on the largest library of high-quality inflammatory markers. To be able to meet the market needs there, that they can pick and choose from inflammation markets mostly, that they are really, really interested in on the pathways or the drugs that they're looking at. That is the rationale and the thinking behind the FlexPlex offering. Got it. In terms of just Signature for the PCR readout, who would use it and how do you view the advantages of Target versus Explore? Is it just to meet a current customer need? Are there any specific use cases where you think, even five years from now, there's always going to be an appetite for Target? It's absolutely going to be, Tejas. I think, the simplest example perhaps to take is a biopharma company moving into clinical development. If you want to look at pharmacodynamics, for example, how the human body reacts to a new drug. Keep in mind that we're also now in live communication with the FDA. Perhaps we don't want to share data from 5,000 proteins where we might have some challenges explaining all of them. Here we are a little bit, we know this drug and which pathways it interacts with. We're very interested in looking very carefully across a few selected pathways, for example. Is this covered by the 100, 200 proteins? Yes. Hence, this is where targeted definitely comes in. Will this be a demand in next year, in the year after, and so forth? Absolutely. I definitely think that that's a very important part of our portfolio. Got it. Outside of adding targets to your library, Jon, how are you thinking about the pipeline here for the platform in general? Yes, our primary focus now is to continue down the path we're on to very aggressively cover, hopefully all, but the vast majority of proteins in circulation with that single plex quality. That is our most important driver right now and which we're very focused on. That's building out the Explore. We're adding the FlexPlex product, as we said, to the Target portfolio, and we're building out capacity for Focus since customers find these results and want to move their science forward. We always constantly try to, can we improve the protocol? Can it be easier to work with? Can it increase throughput? What are new flow cells coming out that we can adapt and so forth. There's always things to polish here to make it an even more delightful experience for the customers to work with our products. Got it. I want to pull in Oskar again for a couple of quick ones on the financial model, Oskar. One of the questions we get often is just the degree of seasonality in the model. Most life science investors are pretty used to sort of a heavier 4 Q for companies. In your case, it's almost sort of 40%-45% of your annual revenue, at least at this point. Why is that, and how do you see that sort of normalizing over time? No, as you said, it's driven by external factors. When we do see a lot of sort of year-end money coming in and a lot of spend in the fourth quarter, and we've seen so since the foundation of the company. Since 2016, Q4 and December has been sort of a very busy month, and it's all hands on deck. Something that we as a company, we are very focused on Q4 and the planning the entire year is to sort of deliver out Q4. It's always that sort of eye to the latter part of the year. As you said, it's a big number, but something that we as a company is sort of comfortable working with and also, looking at our pipeline and sort of the opportunities, they are skewed to Q4, so it's nothing that sort of we as a company has invented. Of course, over time, as sort of proteomics becomes bigger and bigger and I think sort of rises on the sort of strategic agenda at our customers and the spending will be sort of smoother throughout the year. I think that's sort of the way to think about it. Again, that's a long it's not going to shift in a year's time or so. I'm sure it will be a sort of a long process to get there. Is that seasonality sort of skewed towards your academic customers or is it more on the kit versus the service side or is it pretty across the board? No, but it is pretty much, it is across the board. There are nuances, but across the board really. Got it. Very quickly, on the margin algorithm, one of the most impressive parts of the Olink story is that you're a consumables and service company by and large. How does that translate into operating profitability for you over time? Yeah. I think, taking a step back and from the inception of the company, we were organically funded, there was no access to outside capital. We made a dollar and invested a dollar. I think that built a very sort of, we have great unit economics on the gross margin, both in kits and in service. We built a very sort of lean and nimble business and generated close to 40% of EBITDA margin in 2019. Now we saw a big opportunity to invest in the business and into the platform. Taking down margin, in 2020, 2021, and 2022, we look at them as investment years. Over time, I think it's Jon's and my ambition to run a profitable company and we have the business model to support that. Got it. Perfect. Jon, in the last sort of 60 seconds here, what part of the Olink story do you wish would be sort of better understood by investors? Or what's most underappreciated in your mind? I think actually we've commented a little bit about it during our conversation. I think proteomics is extremely complex, going on sensitivity, specificity, dynamic range, throughput, sample consumption, cost effectiveness. To have a technology that's overcome all of those challenges in a very high multiplexing with single plex quality, I think we have a very picky customer, scientific customer base that do very thoughtful considerations before choosing platform. I guess the strong evidence of us having more than 700 customers across the most prestigious biopharma and academic institutions worldwide with over more than 600 publications is an extremely strong support, I guess, for that we are really marching down a solid path here and has really overcome many of these challenges and very much appreciated by customers. I hope that that message resonates and everyone brings that home. Great note to finish on. Jon, Oskar, thank you so much for your time this morning. We appreciate it and I hope you have a productive rest of the conference. Thank you very much. Thank you. Thanks.