Absci Corporation

Thank you. Earlier today,

Abcide released financial and operating results for the quarter and year ended December 31, 2024. If you haven't received this news release, or if you'd like to be added to the company's distribution list, please send an email to investors at Abcide.com. An archived webcast of this call will be available for replay at Abcide's investor relations website at investors.abcide.com for at least 90 days after this call. Joining me today are Sean McLean, AFSCI's founder and CEO, and Zach Jonason, Chief Financial Officer and Chief Business Officer. Christian Stegman, AFSCI's FCP of Drug Creation, will also join for Q&A following prepared remarks. Before we begin, I'd like to remind you that management will make statements during this call that are forward-looking within the meaning of the federal securities laws. These statements involve material risks and uncertainties that could cause actual results or events to materially differ from those anticipated, and you should not place undue reliance on forward-looking statements. Additional information regarding these risks, uncertainties, and factors that could cause results differ appears in the section titled, Forward-looking Statements in the Press Release APSEI has read today, and the documents or reports filed by APSEI from time to time with the Securities and Exchange Commission. Except as required by law, APSEI disclaims any intention or obligation to update or revise any financial or product pipeline projections or other forward-looking statements, either because of new information, future events, or otherwise. This conference call contains time-sensitive information and is accurate only as of the live broadcast, March 18, 2025. With that, I'll turn the call over to Sean.

Thanks, Alex. Good morning, everyone. Thank you for joining us for our fourth quarter and full year 2024 business update call. 2024 was a successful year for AppSci. I'm proud to say we executed across all aspects of our business, including advancing our proprietary internal programs, delivering on partner programs, and adding four new partners to our ecosystem of collaborators. We cap the year off with our 2024 R&D Day in December, where we unveiled our potentially category-defining ABS-201 program, targeting the prolactin receptor for the treatment of androgenic alopecia. At this event, we shared new data supporting a potential best-in-class profile for ABS-101, our anti-TL1A program, and preclinical data for ABS-301 and ABS-501 programs. Additionally, we showcased new breakthroughs in de novo antibody design from our leading AI platform and hosted distinguished guest presenters, including our KOLs and partners, such as Dr. Dennis Salaman from UCLA and Dr. Luis Diaz from Memorial Sloan Kettering. For those who did not view the webcast, I encourage you to access the replay on our website. ABS-201 in particular is an asset we are extremely excited about. We see the opportunity for ABS-201 to become a potential flagship asset for Abcide. But first, I'd like to reflect on the capabilities that enabled us to generate these differentiated antibody assets. Looking back on 2024, we're encouraged by the significant advancements in our AI-integrated drug creation platform. Similar to how the broader tech industry is experiencing successive breakthroughs in generative AI, we view the progress of our model and platform in a similar light. Just two years ago, in January 2023, we released our first AI de novo preprint. Since then, we have demonstrated notable platform improvements, progressing from single CDR design to designing antibodies to targets with no known binders. These capabilities were clearly showcased in December, particularly through our collaboration with Caltech, which focused on designing a universally neutralizing HIV antibody. Before discussing that collaboration further, I'd like to highlight the key factors behind our success. At AbSci, we see four key ingredients for success. Our data advantage, our leading AI models, access to scalable compute, and fostering a team with multilingual expertise. While we've spoken extensively about each of these in the past, I'd specifically like to focus on compute. In January, we announced a collaboration with AMD, a leader in high performance computing. As part of this partnership, AMD made a $20 million strategic investment in . This collaboration supports our mission of creating better biologics faster. by offering optimized compute solutions for complex biological modeling. These solutions provided exceptional performance, reduced infrastructure costs, and accelerated cycle times. At the JPMorgan Healthcare Conference in January, I outlined several key reasons we chose to partner with AMD. AMD's chips gave us unmatched training resolution, allowing us to model large protein complexes without the need to crop, preserving biological context and enhancing model accuracy. Furthermore, this collaboration significantly accelerates throughput, scaling in silicone antibody design and evaluation, and ultimately reducing R&D timelines and costs. Stepping back, what's the broader purpose of harnessing generative AI for antibody design? It's not just about designing faster or cheaper. It's about creating truly differentiated candidates and unlocking novel biology for the benefit of patients. In December, we shared a case study illustrating breakthroughs in AI de novo design for our collaboration with Caltech. Using our proprietary de novo design model, we created antibodies targeting a difficult-to-drug epitope in the HIV-Coldera region, potentially facilitating the development of a universal neutralizing HIV antibody. The Coldera region, uniquely accessible only in the GP120 open confirmation has remained untargeted by previously neutralizing antibodies. Successfully creating these antibodies marks a potential pivotal milestone in HIV vaccine research and underscores the capability of our de novo design model to target previously undruggable epitopes. We're also actively applying our generative AI drug creation capabilities to our proprietary internal pipeline. In December, we unveiled ABS-201, a potential best-in-class antiprolactin receptor antibody for androgenic alopecia. This is an indication with significant unmet need in a large patient population, approximately 80 million people in the U.S. alone. Androgenic alopecia, also known as male and female pattern hair loss, affects 50% of men and 40% of women by the age of 50. There's been no real innovation in nearly 30 years, creating a large market potential. Our approach aims to not just slow hair loss, but to unlock an entirely new category focused on hair regrowth. We've nominated a development candidate for ABS 201, supported by preclinical data suggesting high affinity and potency, favorable safety and immunogenicity, extended half-life, enabling convenient infrequent dosing, and excellent developability and manufacturability. Preclinical models demonstrated improved hair growth compared to minoxidil. As we advance ABS-201, we envision a straightforward path to clinical development and have assembled a robust network of renowned hair and dermatology KOLs advising our progress. ABS-201 is currently an IND-enabling site, with Phase I trials anticipated to begin in early 26. Since unveiling ABS-201 three months ago, we've received very positive responses from industry experts and the financial community. Given the compelling data, clear development path, and significant market opportunity, our strategy is to develop ABS-201 internally through later stage clinical development and proof of concept, retaining maximum value for AppSci. Turning now to ABS-101, our potential best in class anti-TO1A antibody. At December's R&D Day, we shared new data indicating ABS-101 does reduce internalization of TO1A complexes in in vitro THP1 immunogenicity tests compared to competitor molecules that had high clinical ADA rates. These data suggest ABS-101 may have lower ADA development risk in clinical settings. Additionally, in January, we shared new ABS 101 NHP PKPD data confirming prolonged target engagement, demonstrating dose-dependent engagement, including a ceiling effect, and significantly improved target engagement compared to competitor molecules at comparable dosing regimes. We plan to initiate phase one clinical studies for ABS 101 in the first half of 2025, with an interim readout expected in the second half of this year. We continue to see active partner interest and have begun developing a potential first-in-class bispecific antibody incorporating our TL1A antibody as one arm. Additional data will be provided later. We recently shared data on ABS 301 and ABS 501 programs as well. ABS 301 targeting an undisclosed immune oncology target discovered through Abcide's reverse immunology platform showed expression across squamous cell carcinomas. Our first in vivo target validation study demonstrated potent anti-tumor response, strongly supporting further development. For ABF501, our potential best-in-class AI-designed anti-HER2 antibody, preclinical data confirms novel epitope interactions Affinity comparable or superior to trastuzumab. Efficacy against trastuzumab-resistant xenograft tumors expressing wild-type HER2. And good developability. Earlier I mentioned that our team was a key ingredient to AbSci's success. This is evident in our achievements in 2024. As always, I'd like to thank our dedicated team at AbSci for their unwavering commitment and effort towards our mission. With that, I'll turn the call over to Zach to walk through our new partnerships, our outlook, and financial updates. Zach?

Thanks, Sean. As Sean mentioned, we added two new partners toward the end of 2024 and achieved our partnership guidance for the year. In addition to our partnerships with Memorial Sloan Kettering Cancer Center and Twist Bioscience, which were announced earlier in 2024, we entered into new partnerships with Alken and in VedEx in December. Our collaboration with Alken combines Alken's novel AI target discovery and agentic AI expertise with our leading AI de novo design models with the aim of designing and co-developing potential first-in-class therapeutics. Together, AbSci and Alken plan to co-develop therapeutic candidates addressing novel targets in immuno-oncology and other indications such as immunology and inflammation. The partnership will leverage Alken's predictive AI models and its biomedical data sets in patient-derived organoids for target selection and validation. ABSA's generative AI drug creation platform, including our de novo antibody design models, will be used to design novel therapeutic candidates against these targets. Together, we aim to streamline and accelerate the development of novel therapeutics targeting novel disease targets. Our partnership within BED-X will leverage our leading generative AI drug creation models to create Novel Antibody Half-Life Extension, or HLE, capabilities for animal health applications. Through this partnership, AppSci will utilize its AI models to design novel, modular antibody sequences that confer half-life extension in specific animal species. Invitex will have rights to use the HLE sequences across its product portfolio to enhance duration of therapeutic effect and customer convenience. both significant potential differentiators in the animal health market. This partnership includes R&D funding, as well as election fees, milestone payments, and royalties on a product by product basis. We believe this collaboration has the potential to address significant unmet needs in animal health, starting with initial applications for large market indications in canine medicine. Since our IPO nearly four years ago, We have continued to evolve our business model as our capabilities have grown. Today, we are focused on building and advancing a balanced portfolio of high-quality, high-value therapeutic programs, including a growing number of proprietary internal, wholly-owned programs. Moreover, while we continue to collaborate with partners to create therapeutic programs for their targets under traditional deal structures, we have also more recently initiated creative co-development partnerships to leverage risk sharing, and cost-reduction synergies for the creation and development of first-in-class therapeutic programs. Based on our business model today, including expansion and advancement of our wholly-owned programs, we believe that the legacy metric of number of external partner programs simply on a gross volume basis has become outdated and is now a less meaningful approach to understanding and valuing our business. Our strategy is based on leveraging our platform and resources to create a portfolio of novel and differentiated therapeutic programs that offer the most promising return. Accordingly, we invest our resources in internal programs as well as a variety of partnership programs that we believe offer the best risk-return profile, rather than simply seeking a potential gross number of partnership programs. We, of course, continue to see value in adding partner programs with high-quality collaborators and will continue to do so on a selective basis. Hence, going forward, we do not plan to provide an exact number of expected new partners or partner programs as a business outlook, as we have in prior years, but rather plan to provide material updates and guidance on our internal and or partnered therapeutic programs when possible. This year, we plan to advance of proprietary internal asset programs as described by Sean earlier. We also anticipate signing one or more partnerships including with a large pharma company for a drug creation collaboration and continue to plan to provide material updates on ongoing partner programs as they advance through development. We will also very soon be a clinical stage biotech company with ABS 101 expected to enter the clinic shortly. and ABS 201 accelerating toward first in human clinical trials potentially early next year. And as a reminder, our business model is focused on out-licensing or selling our internal programs and co-developed programs following value inflection proof points as early as preclinical proof of concept or much later stages. Turning now to our financials, revenue in the fourth quarter was $0.7 million as we continue to progress our partner programs. Research and development expenses were $18.4 million for the three months ended December 31st, 2024, as compared to 12.3 million for the prior year period. This increase was primarily driven by advancement of our internal programs, including direct costs associated with IND enabling studies for ABS 101 and the increase in stock compensation expense. Selling, general and administrative expenses were $8.8 million for the three months ending December 31, 2024, as compared to $9.3 million for the prior year period. This decrease was due to lower personnel and other costs, offset by an increase in stock compensation expense. For the full year 2024, our gross use of cash, cash equivalents, and short-term investments, exclusive of partnered program payments, was approximately $72 million, the lower outlook of $75 million on a gross basis. Turning to our balance sheet, we ended the year with $112.4 million in cash, cash equivalents, and short-term investments, as compared to $127.1 million as of September 30, 2024. In January of this year, we announced a new strategic collaboration with AMD. As part of that collaboration, AMD made a $20 million equity spend in AvSyte. purchasing Abcide common shares at a price representing an approximate 14% premium to Abcide share price based on the prior day's market closing price. Further, this year we have also raised an additional approximately $20 million by utilizing our at-the-market facility, the vast majority of which was raised from two premier mutual funds. This additional approximately $40 million raised since the end of 2024 enables us to continue investing in our internal programs, including accelerating the development of ABS-201, our anti-PRLR program for the treatment of androgenic alopecia, as well as our co-development programs. In sum, we continue to deepen our focus on high-value proprietary internal programs, as well as high-quality co-development and drug creation partnerships with industry leaders who bring synergistic capabilities. We believe that this strategic, balanced approach will provide us with the best return for shareholders. Based on our current plan, we believe our existing cash, cash equivalents, and short-term investments will be sufficient to fund our operations into the first half of 2027. In all, we are very pleased with the progress we have made over the past year and are confident in our ability to execute across our portfolio of programs this year and beyond. With that, I'll turn it back to Sean.

Thanks, Zach. In closing, we've made tremendous recent progress across our platform and internal pipeline. New data for ABS 101 reinforces its best-in-class potential, and ABS 201 presents an exciting opportunity to create an entirely new category addressing a significant market and patient need. In coming months, we'll reach the milestone of becoming a clinical stage biotech company as ABS 101 enters the clinic. As we also advance ABS 201 towards the clinic, we're energized by the potential to bring this innovative product to roughly 80 million people affected in the U.S. Alongside progress on our internal programs, as Zach mentioned, we anticipate adding one or more new partners, including a large pharma partnership. Again, these achievements are possible because of our dedicated team of delimiters at AbSci, who advance our mission each and every day. Thank you all. I'll now turn the call back to the operator for Q&A. Operator?

Certainly. And our first question for today comes from the line of Kripa Devrakonda from True Securities. Your question, please.

Hi, this is Alex on for Kripa. Congrats on all the progress. Looking forward to another exciting year with AppSign. We had a couple questions on our end. One, for Abcide 101, we saw compelling preclinical data last year at your R&D day. We know as we wait for the clinical trial to get underway now and the Phase I data later this year, does Abcide plan to conduct additional preclinical studies with ABS 101, and would investors get to see that, and maybe also with ABS 201 as well, if there are preclinical studies underway and plans to present that in the 2025 timeframe?

Yeah, great question, Alex. Christian, do you want to take that?

Sure, happy to. So, regarding ABS 101, we have disclosed at R&D Day a novel preclinical data. At JPMorgan, we have also disclosed target engagement data in non-human primates for ABS 101. and we are essentially wrapping up our IND-enabling work currently. So we will disclose at a scientific conference the full toxicology data, but we are proceeding towards clinical development, as mentioned. And then for ADS-201, we are in the IND-enabling phase, and we will disclose data in the course of this work, also preferably at a scientific conference at an upcoming conference.

Yeah, I was just going to say, on ABS 201, we have accelerated our timeline on that. We do plan to be in the clinic in early 26 with a potential interim efficacy readout next year on ABS 201. So we'll have 101 in the clinic later this year, followed by 201.

Oh, that's great. And one more, if I may. I think you said that you're not going to provide the number of new partners going forward as some standard practice for the firm, but more material guidance on internal assets and also the partner programs and any additional color you can provide on the partner program updates, what that might look like, given that other people might be taking the lead as part of your contracts over there, but obviously very interested in all that development for investors and for us.

Yeah, absolutely. We're really focused on the large pharma partnership. We plan to execute one new large pharma platform deal this year. And then additionally, we are in discussions on ABS 101 on potentially out-licensing that asset. And so there definitely are going to continue to be transactions, but we really want to be focused in on transactions that can bring in significant upfront payments that can, in a non-dilutive way, extend cash runway. And it also provides really nice validation for the platform and additionally allows us to go into other indications that we would not pursue on our own and really having us to have a diversified portfolio. Zach, is there anything else you would like to add there?

Yeah, I mean, I'll just add that, you know, going forward, there'll be a much greater focus on providing guidance around our internal programs and the CODEP programs, particularly once they're at a DC stage. As you would imagine, in our drug creation partnerships, disclosure of information on those programs is sometimes not as easy because that partner would have to agree to that. But certainly for the programs that we're advancing, and as you've seen, we've evolved our business model to have a greater focus on internal programs and now adding code ads in there as well, I think gives us quite a bit of guidance to provide around those. And so that'll be our focus going forward.

Fantastic. Thank you all.

Thank you. And our next question comes from the line of Arsini Shabashvili. Your question, please, from Guggenheim.

Hi, it's Arsene Jan for VAML. Congrats on all the progress in 2024. We had a question on 201. Given the data that you've seen preclinically, how are you thinking about the design and endpoint for the planned phase 1 study?

Yeah, that's a great question. I can speak a little bit to the design and hand it over to Christian for the endpoints. The way we're thinking about this right now is that we will start with a SAD study that would begin sometime early next year, followed by a MAD study where we do plan to power that study appropriately to be able to achieve a proof of concept. And all that will be beginning next year. Christian, do you want to speak to the endpoints that we're going to be looking at?

Yeah, absolutely. Great question. So when it comes to androgenic alopecia, there is a number of well-accepted endpoints described. And importantly, they can all be measured with a very well-established device, the trickle scan device. These endpoints have also been accepted by regulators. So this is a non-invasive computer-assisted optical measurement of hair density, terminal hair count, and it allows us essentially to progress this program very effectively through clinical development.

Thank you. Very helpful. And maybe one more on the large pharma partnership that we're hoping to secure in 2025. Can you share maybe just on the high level which therapeutic areas and modalities these discussions are focusing on? Maybe what companies are interested in and maybe where they recognize your capabilities and your expertise?

Yeah. So, I would say that the focus is pretty broad in terms of overall indication. You know, I think some of the focuses are definitely on INI and oncology. But I would say more generally where the focus lies and what's really driving these platform partnerships is the de novo models capability that we've been able to show where we can go after epitopes that have no known binders that have been difficult to drug in the past and being able to ultimately drug those particular epitopes or targets with our platform. And I think what we're seeing is that there's a you know a lot of interest in you know ion channels like the ones what we're working on with omarol uh or various uh you know gpcrs uh where they're you know there's been difficulty drugging these and uh and the indications have been pretty broad in terms of uh overall uh interest uh zach did did i miss anything there do you want to add

No, the only one thing I would add, Arseny, is you can think of the partnerships we're working on or working towards with large pharma as being multi-target, and to Sean's point, they can span multiple indications.

Thank you. Thank you. And our next question comes from the line of Vikram Purohit from Morgan Stanley.

Your question, please. Hi, good afternoon.

Thanks for taking our questions. We had two. First, on 101, we just wanted to revisit expectations for the readout in the second half of the year. We're just curious to see any updated thoughts you might have around the size of the data set, the scope of the data set we'll be receiving, and just your current view on what you would consider a really strong outcome here. And then secondly, I think you mentioned that you're working on a bispecific that addresses TL1A in part. We're just curious on what the next steps there might be and what you think the future for that program could be, whether that could be something that's eventually partnered alongside 101, or would that be a separate effort you want to hold on to for a longer time point? Thanks.

Yeah, thanks, Vikram. So, I can answer your second one and hand the first question over to Christian. So regarding the bispecific, we are planning on developing a potential first-in-class TL1A bispecific with TL1A on one arm, and then a target that has been known but has been difficult to drug in the past. And we see this as an exciting potential first-in-class TL1A bispecific, but additionally, developing it out as a monotherapy as well. And we plan to have a lead on that this year and then a DC to follow shortly thereafter. But we do see this as, I think, an exciting evolution of the INI pipeline that we're building out. And we do see combo-based therapy or being able to hit multiple pathways with the bispecific as a way to really meet some of the unmet needs within the IBD and UC space. Christian, I'll hand it over to you for anything I missed there and then answering the first question.

Thanks. Thank you.

Our next question comes from the line.

Sorry, Christian. Christian, are you there? Alex, did we lose Christian? His line is still connected. Yes, we can hear you, Christian. Christian, are you still there? Can you hear me?

Yeah, we can hear you now.

All right. I'm not sure what's going on. I'll try again. So with regard to the interim readout in the second half of this year for AVS-101, what you can expect to see is something quite similar, what we have disclosed for non-human primates at J.P. Morgan earlier this year. So essentially, what we'd like to see is a demonstration of sustained elevation of soluble T1A in serum in humans after a single dose. And we'd like to see that in one cohort after a single dose administration. So that's an important de-risking point. And I will just remind you that the competitor molecules that have claimed prolonged half-lives have, to our knowledge, not disclosed targeted engagement in non-human primates. Hence, we think we are very well positioned to demonstrate that not only in human primates, but also in humans.

Thank you.

And our next question comes from the line of Gil Bloom from Niedermann Company. Your question, please.

Good afternoon, and thanks for taking our questions. One relating to the planned pharma partnership. So did some of the achievements you guys demonstrated, you know, binding to the Caldera region of HIV, did that play into your pharma discussions and have a follow-up?

Yes, that case study was a, has been a big driver in the discussions we're having with large pharma companies. Right now, and I think it's a really strong illustration of how we can go after these difficult targets that still exist. And, you know, we do believe that that case study is an important piece to the value prop and, you know, ultimately driving these large pharma partnerships across the finish line.

Thank you. This is Zach. I might add to that too, just to contextualize it. I think the other thing that we're excited about and we've seen resonate with the pharma companies we're dialoguing with is not only the results there, but if you look over the history of the versioning of our models, you can see the kind of dramatic performance and capability gains as we go from early versions back in 2022 to where we are today with the models that can address the Caldera epitope, for example.

Okay. It's very helpful. And as it relates to this bispecific lead you mentioned for IVD, what kind of dynamic should we imagine here with ABS 101? I mean, you know, there are other companies in the space that are really taking lead on this kind of, you know, combo approach? Is this what we're thinking of here, or is this something really completely different? Thank you.

Yeah, thanks, Gil. We do see this as different, but Christian, do you want to maybe talk a little bit about the biology and what we're trying to achieve with this particular biostatistic?

Yeah, absolutely. So, yeah, you're of course correct that there is competition that... has announced to run combination studies. We think obviously this is quite interesting. We do think a bispecific can potentially provide additional value that will be more difficult to realize in the clinic in combination studies. So we haven't disclosed the target we're looking at specifically, but we do think there is potentially a strong synergy between TL1A and the other mechanism.

Yeah, and I will note this is not, you know, an IL-23 or an alpha-4 beta-7. This is a novel mechanism. Again, that's been difficult to drug, and as a monotherapy, it definitely would be a first-hand class, both as a monotherapy and as a bispecific.

Great. Very helpful. Thanks for taking our questions. Thank you.

And our next question comes from the line of Brendan Smith from TD Cowan. Your question, please.

Hi, this is Jackie on for Brendan. Thanks for taking the questions. Maybe just touching back on your HIV program, could you remind us on how you plan on progressing that program and kind of how that works under the current partnership structure and give me an additional color on when we might see updates on the progress regarding that program?

Yeah, so this is a collaboration that we have with Caltech and it's being funded by the Gates Foundation. And so we're going to continue to develop this molecule and at the appropriate time work with the Gates Foundation on taking this into the clinic, assuming the preclinical data looks good. And at that point in time, once we start to see some Some in vivo results, we should be able to come out with a plan on how we plan to develop this in collaboration with Caltech and the Gates Foundation. So stay tuned.

That's great. And maybe just swinging over to more of the business side, can you give us any update on how to kind of look at 2025, particularly around the expected ramp and spend from the study initiations and maybe anticipated cash burn rates?

Yeah, Zach, I'll let you take that.

Yeah, sure. So I think we, you know, to start off, just to note that we've reiterated our guidance that we have a balance sheet that'll fund our strategy and operations into the first half of 27. So I think we're very well positioned to advance our phase one studies for ABS 101. And as we mentioned earlier, we'll have an interim readout on that trial in the second half of this year. I think the other point, which Sean mentioned earlier, is we'll be accelerating the development plan for ABS 201, and that is within our budget. The plan there is to be able to initiate first-in-human studies early next year with the potential for an interim efficacy readout in 2026. So that's all part of our core strategic plan, and I think we're capitalized to achieve that as well as advance some earlier stage programs that we're working on today.

Great, thank you.

Thank you. As a reminder, ladies and gentlemen, if you do have a question at this time, please press star 1-1 on your telephone. Our next question comes from the line of Devanchana Chatterjee from Jones. Your question, please.

Thanks for taking my question. So given the recent positive data from Sanofi and Teva's duvatic colon ulcerative colitis, But, like, particularly the strong efficacy and low ADA rates in patients, how is APSI positioning its anti-TL1A assay to compete? To what extent does this raise the bar in terms of, like, makeup performance or commercial potential?

Great question. Christian, do you want to take that?

Yes, great question. So we have obviously seen the Sanofi-Teva data. It is indeed impressive data. However, it was interesting to observe that the inclusion-exclusion criteria in that study was perhaps an aspect that makes it a little bit more difficult to compare this study with other data that's out there. In particular, there was a significant share of patients. So that raises the question, to what extent can we expect similar data from other antibodies? The other interesting thing we believe with aegis-101, there may be a potential to potentially dose higher than what we've seen with competitors. So this is an avenue we are exploring as well. So it remains to be seen whether in terms of the efficacy there is a ceiling effect or whether dosing higher can potentially lead to additional efficacy.

Okay. Thanks for that.

Thank you.

And our next question comes from the line of Swayampakar Rayamkanath. from HCW. Your question, please.

Thank you. This is RK from HC Wainwright. Good afternoon, Sean and Zach and Alex. So, you know, a lot of the questions on the pipeline have been already asked. But, you know, in the time that you have spent generating this pipeline, how much of the data that that has allowed you to expand the pipeline, are you able to take it back into your model, into your AI models, and are able to strengthen it? And what sort of benefits are you seeing by developing this kind of a RAS pipeline? And the second part of the question is, from your initial conversations with Merck way back in 2021, what sort of conversations happen now when you start working with potential partners? What's the evolution there?

Yeah, great question. So to answer the first one, RK, we have a lab-in-the-loop process. We're in a six-week time period. We can go from data in our wet lab to training our models to validating that. And that occurs both on, as you mentioned, our own internal pipeline, but also in our partner programs. And so that lab-in-the-loop process allows us to rapidly iterate on the design and the architectures of our models to continually improve them, as well as being able to feed in new data for training, which is obviously helping us increase the overall accuracy and generalizability of these models. And we've made, even since two, two and a half years ago, we've made tremendous progress from being able to design the HCDR3 of an antibody to now being able to design an antibody where there was no known binder to difficult and challenging targets like the cold air region or ion channels. And we see, again, that lab-in-the-loop process as really key to unlocking our ability to increase the generalizability and accuracy of our models, and that's really the reason why we've been able to achieve what we have today. And, Zach, do you want to answer that second question?

Yeah, sure. I mean, it bridges off exactly what you were describing, Sean. RK, our discussions with Pharma right now are really highly centered on the de novo design models that we have and the capabilities that we've advanced over the last year to two years. And so I think the big value proposition, to Sean's point, is sort of unlocking these targets that haven't been addressable with... traditional antibody discovery technologies. And so that's a real central value proposition that we're actively exploring with phRMA right now. There are some other interest points too around the ability to create novel pharmacology profiles, for example, pH-dependent binding. There's definitely some interest in that aspect of what we're doing and some of the data we've put out at R&D Day you can refer to. back to Sean's point, I mean, the biggest emphasis here and I think the biggest value proposition is being able to use those de novo design models to address these targets where there's really, you know, known biology, but they just have not been addressable by traditional techniques.

Thank you. Thank you both for taking the questions.

Thank you. And as a reminder, ladies and gentlemen, if you do have a question at this time, please press star 11 on your telephone. And this does conclude the question and answer session as well as today's program. Thank you, ladies and gentlemen, for your participation. You may now disconnect. Good day.