Schrödinger, Inc. (NASDAQ:SDGR) Q2 2024 Earnings Call Transcript

Schrödinger, Inc. (NASDAQ:SDGR) Q2 2024 Earnings Call Transcript August 1, 2024

Operator: Thank you for standing by. Welcome to Schrödinger’s Conference Call to review our Second Quarter 2024 Financial Results. My name is Madison [ph] and I will be your operator for today’s call. [Operator Instructions] Please be advised that this call is being recorded at the company’s request. Now I would like to introduce your host for today’s conference, Miss. Jaren Madden, Senior Vice President of Investor Relations and Corporate Affairs. Please go ahead.

Jaren Madden: Thank you and good afternoon everyone. Welcome to today’s call, during which we will provide an update on the company and review our second quarter 2024 financial results. Earlier today, we issued a press release summarizing our financial results and progress across the company which is available on our website at schrodinger.com. Here with me on our call today are Ramy Farid, Chief Executive Officer; Geoff Porges, Chief Financial Officer; and Karen Akinsanya, President of R&D, Therapeutics. Following our prepared remarks, we’ll open the call for Q&A. During today’s call, management will make statements that are forward-looking and made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995, including without limitation statements related to our outlook for the third quarter and full year 2024, our plans to accelerate the growth of our software business and advance our collaborative and proprietary drug discovery programs, the timing of initiation of, and readouts from our clinical trials, the clinical potential and properties of our compounds, the use of our cash resources and our future expenses.

These forward-looking statements reflect our current views about our plans, intentions, expectations, strategies, and prospects, which are based on the information currently available to us and on assumptions we have made. Actual results may differ materially due to a number of important factors, including the considerations described in the risk factors section and elsewhere in the filings we make with the SEC, including our Form 10-Q for the quarter ended June 30, 2024. These forward-looking statements represent our views only as of today and we caution you that, except as required by law, we may not update them in the future, whether as a result of new information, future events or otherwise. Also included in today’s call are certain non-GAAP financial measures.

These non-GAAP financial measures are not prepared in accordance with generally accepted accounting principles and should be considered only in addition to, and not a substitute for or superior to GAAP measures. Please refer to the tables at the end of our press release which is available on our website for reconciliations of these non-GAAP measures to the most directly comparable GAAP measures. With that, I’d like to turn the call over to Ramy.

Ramy Farid: Thanks, Jaren and thank you everyone for joining us today. We are very pleased with our progress during the quarter delivering excellent revenue growth and continuing to advance our proprietary pipeline. As you will hear from Karen shortly, our first two clinical programs are progressing well in Phase 1 clinical studies. And I’m happy to report that we have initiated dosing and a Phase 1 solid tumor study for SGR-3515, our Wee1/Myt1 co-inhibitor. Total revenue for the second quarter was $47.3 million and software revenue was $35.4 million. We are confident about our revenue outlook for the full year and our reiterating our full year software and drug discovery revenue guidance. We continue to engage in very productive discussions with customers at large global companies about significant scale ups and expanded use of our software.

These discussions are maturing in line with our expectations and have been facilitated by discussions at the highest levels of these companies about the impact of large scale deployment of computation and drug discovery. It’s clear that interest and deploying advanced computational methods remains high. Collaborations are continuing to be an important element of our business, and our progress across several partner programs contributed meaningfully to drug discovery revenue in the second quarter. Our co-founded companies are also an important part of our overall strategy, and their success provides further validation for our platform and is an additional source of capital. We would like to extend our congratulations to our partner, Morphic Therapeutic, on their planned acquisition by Lilly for $3.2 billion.

We are proud to have worked side-by-side with Morphic scientists on several programs including MORF-057. We would also like to congratulate our partners structure on the presentation of their Phase 2 obesity data and Ajax Therapeutics on their recent Series C financing and IND clearance for their JAK-2 inhibitor. These achievements that our co-founded partner companies demonstrate the power of our platform to serve as a catalyst for innovation in drug discovery. We have developed and deployed powerful physics-based methods amplified with advanced machine learning and AI, and we will continue to advance our platform to unlock the full potential of computation to drive efficiency across the R&D continuum. We recently launched an exciting new initiative to develop a computational solution designed to reduce the risk of development failure associated with binding to off-target proteins which can be associated with serious side-effects.

We have already built accurate models for a handful of key off-target proteins, and have begun using this technology in our partner and proprietary programs. In the coming years, we aim to build predictive models for many of the most well-known off-target proteins. We expect these predictive toxicity assays to contribute to future software growth, given the demand for solutions that can help screen and optimize drug molecules to enhance the likelihood of success in clinical trials. We are delighted that this effort is being funded initially by a $10 million grant from the Bill & Melinda Gates Foundation. We have made considerable progress in the first half of the year and we are executing on our strategic priorities for the remainder of the year, including driving software scale up and adoption, advancing the science that underlies our platform, and progressing our partner in proprietary programs.

I greatly appreciate the dedication of our employees whose hard work and collaboration are critical to achieving our mission. I will now turn the call over to Geoff.

Geoff Porges: Thank you, Ramy and good afternoon everyone. Schrödinger had an excellent quarter in Q2 with software revenue growing strongly, drug discovery revenue elevated by the recognition of milestones associated with the progress of several collaboration projects, significant progress achieved in our proprietary therapeutics portfolio, and validation for our technology and business strategy from the planned Morphic acquisition. Our business showed its resilience as revenue growth in our global accounts offset challenges in certain geographies and market segments. We are maintaining our previously provided revenue guidance for the full year, and we’re increasingly confident that opportunities such as the recently announced predictive toxicology project position us for sustained growth in the coming years.

Software revenue in Q2 was $35.4 million compared to $29.4 million in Q2 2023. The 21% decrease was due to increased revenue from global accounts, as well as revenue recognized from our combined software drug discovery collaboration customers. As we indicated previously, our customers are progressively converting to readable software contracts. And this conversion is reflected in the increase in our hosted software revenue this quarter, with customers making this conversion at both global pharmaceutical companies and established biotech companies. It is noteworthy that our total software revenue grew strongly even with a significant increase in the actual percentage of hosted revenue compared to the same period in 2023. On-premise software grew by 11.5% to $18.8 million in Q2, and hosted software grew by 82% to $8 million.

Maintenance revenue was flat year-over-year at $5.8 million and professional services increased by 23%. As a percentage of our total software revenue, hosted revenue was 22.6% in Q2 compared to 15.4% in the same period a year ago, and compares to 21.5% in Q1 this year. I’d like to make a few comments on the dynamics we are seeing in our software business at the midpoint of the year. First, our strategy has been to focus on increasing the scale of adoption of our technology at our established customers, and this strategy is working. We are seeing more and more customers adopting our suite of software solutions as foundational to their drug discovery efforts. And recognizing that increased scale of utilization results in better outcomes. The benefits of that strategy are reflected in this quarter’s results and in our reiterated guidance for the year.

Second, we continue to see volatility around renewals and scalar purchasing among emerging and small biotech customers. In some cases, these customers are dealing with balance sheet and strategic concerns by reducing or halting drug discovery efforts and concentrating on one or two of the most advanced programs. However, we are also seeing green shoots in terms of new companies that are powering their drug discovery efforts with our software, and on balance we are finding more new customers and increasing accounts than we are encountering retrenchments. In the first half of the year, the bookings increase from growth accounts continues to outpace the booking decrease from reducing accounts, and the contribution from new software customers is exceeding discontinuing customers.

We are confident that the opportunities in large growth and new accounts are considerably greater than the issues in the emerging company segment. Third, geographically, we’re also balancing risks and opportunities. A relatively small business in China has had challenges associated with reduced availability of capital for biopharma research there, and those challenges are exacerbated by geopolitical issues. And other businesses in Asia are advancing and we are very encouraged by the growth opportunities we’re seeing in Europe and North America. Our drug discovery revenue was $11.9 million this quarter compared to $5.8 million in the same period a year ago. The increased revenue this quarter was due to recognition of revenue for the achievement of planned milestones in ongoing drug discovery collaboration projects.

Total revenue for the quarter was $47.3 million compared to $35.2 million in the same period a year ago, and compared to $36.6 million in Q1 this year. The year-over-year increase was driven by contributions from software and drug discovery revenue. Turning now to gross margins. Our software gross margin was 80% in Q2 compared to 77% in the same period a year ago, and compared to 76% in Q1 this year. The increase in gross margin was associated with increased scale of renewals at large customers during the period, and lower royalties and FDA efficiencies contributed to the sequential improvement. Our cost of drug discovery services decreased significantly to $8.8 million compared to $14.7 million in the same period in 2023, and also decreased compared to Q1 this year.

The decrease was driven by the shifting allocation of our drug discovery employees from collaboration projects to proprietary research, and also lower CRO and other project expenses associated with discontinued collaboration projects. In the absence of other collaborations or collaboration projects these expenses are likely to be in the current range in future periods. Our overall gross margin was 66% during the quarter compared to 39% in the same period a year ago, compared to 52% in Q1, the increase in overall gross margin was driven by higher revenue and lower costs. R&D expense was $51 million in Q2 this year, compared to $43 million the same period a year ago and compared to $51 million in Q1 this year. The year-over-year increase in R&D expense was approximately 2/3 drug discovery and 1/3 non-drug discovery.

The drug discovery increase was driven by the shift in allocation of staff from collaboration projects to proprietary programs, and by high headcount and increased CRO expenses for internal programs. The increase in non-drug discovery R&D expense was due to increased technology and cloud expenses and higher FDA expenses. Compared to Q1, the total R&D expense was flat as changes in allocation were offset by lower underlying FDA expenses for R&D. Sales and marketing expense was $9.7 million for the quarter and increased by 7.5% compared to the same period a year ago but decreased by 4.7% compared to Q1 this year. The year-over-year increase was mainly due to higher FTE expenses, including commissions and increased marketing investment. G&A expense of $23.5 million increased by 1.4% compared to the same period in 2023, and declined by 8% compared to Q1 this year.

The decrease compared to Q1 which were lower FTE expenses, reduced severance and lower royalties compared to prior accruals. Total GAAP operating expenses were at $84.1 million for Q2 compared to $75 million in Q2 last year, and $86.3 million for Q1. The year-over-year increase was mainly due to R&D, while the decline compared to Q1 was due to G&A. Our operating loss for the quarter was $52.7 million compared to a loss of $61.1 million in the same period a year ago and total loss from operations was $67.4 million in Q1. During the quarter, the change in fair value of equity investments was an expensive $5.8 million as the value of our investments in structure and Morphic declined during the period. Lilly’s offer to acquire Morphic occurred after the close of Q2 and is not reflected in these results.

Our other income was $4.6 million compared to $4.3 million in the same period last year, mainly associated with interest on our cash reserves, and total other expenses was loss of $1.2 million. Our loss before taxes and net loss were both $54 million. As our tax expense was now negligible, our loss per share was $0.74. This compares to net income of $4.3 million in Q2 last year driven by the Nimbus distribution and earnings per share of $0.06. Our average fully diluted share count in Q2 was $72.7 million compared to $75.1 in Q2 last year. During the quarter our operating cash uses $54 million and a and marketable securities balance declined by $54 million compared to the end of Q1. In Q3 we expect to receive $48 million from the sale of our holding in Morphic, which is expected to largely offset the cash used in operations in the period.

We remain eligible for low single digit royalties or Morphic’s leading clinical program and to variable royalties and other programs in their portfolio. We continue to own approximately 2.4% of the common shares outstanding at Structure Therapeutics. And during the quarter, we invested an additional 3 million into Ajax Therapeutics to maintain our ownership at 5.8%. Looking ahead, we are maintaining our previously provided software guidance for the year. We expect software revenue in Q3 to be in the range of $32 million to $34 million. We expect the recently announced funding from the Gates Foundation to contribute some software revenue in the second half of the year, aligning with our expected commitment of existing and new resources to that project.

We do expect this project to have a modest negative effect on our gross margins, primarily because the profitability of the grant, which is software contribution revenue, is lower than our software revenue from contracts with customers. As a result, we now expect our full year reported software gross margin to be slightly lower than last year, and in the range of 2022. The lower gross margin is likely to persist for the duration of the grant but the profitability of our software revenue from contracts with customers is trending to the same range as last year. The expected dollar increase in our cost of goods associated with the grant is likely to be matched by a similar decrease in our expected R&D expense which should moderate our expected overall expense growth for the year by approximately one percentage point.

We now expect our full year operating expense growth to be in the range of 8% to 10%, or the low end of our previously provided range of 8% to 12%. At the conclusion of the grant, we expect the negative effect on our operating software gross margin to reverse. Our full year drug discovery revenue guidance is unchanged at $30 million to $35 million. And we expect the balance of our realized revenue to be weighted towards Q4. Finally, we continue to forecast that our cash used this year will be greater than our cash used last year. With that, I’ll turn the call over to Karen to discuss our therapeutics updates.

Karen Akinsanya: Thank you, Geoff and good afternoon, everyone. We are continuing to advance our pipeline of collaborative and proprietary programs. We are very pleased to see the progress of compounds and clinical programs that companies we have partnered with and co-founded, including Ajax and Structure which continues to demonstrate the power of our platform when deployed at scale. As Ramy mentioned, Morphic recently announced its planned acquisition by Lilly. We worked closely with the Morphic team on the discovery of several oral integrin inhibitors, including MORF-057 which is a highly selective α4β7 inhibitor. Our proprietary pipeline is progressing well, and we now have 3 programs in the clinic. Beginning with SGR-1505, our MOD-1 inhibitor.

Our Phase 1 study in patients with relapsed refractory B-cell lymphomas is advancing. We are continuing to enroll patients in cohorts evaluating SGR-1505 administered once daily, and we have initiated a twice daily dosing cohort. We now expect to present initial clinical data from this study in the first half of 2025. We plan to present safety, pharmacokinetic and pharmacodynamic data, as well as preliminary efficacy data in a peer reviewed setting. Turning to our CDC-7 inhibitor, SGR-2921; enrollment is ongoing in our Phase 1 study and patients with acute myeloid leukaemia or myelodysplastic syndrome. As a reminder, the primary objectives of the study are to evaluate the safety, tolerability and to determine the recommended Phase 2 dose. The study is progressing well with multiple dose escalation steps completed.

Based on the pace of dose escalation and the timing of abstract deadlines for medical meetings, we now expect to present initial clinical data from this study in the second half of 2025. SGR-2921 was recently granted FDA Fast-Track designation in relapsed refractory AML which underscores the unmet need and strength of our preclinical data. Today, we also announced the initiation of dosing in our Phase 1 clinical study of SGR-3515, our Wee1/Myt1 co-inhibitor. The Phase 1 study is designed to evaluate the safety, PKPD, preliminary anti-tumor activity and recommended Phase 2 dose. The study population will include patients with advanced solid tumors predicted to be sensitive to Wee1/Myt1 inhibition, including breast, ovarian and uterine cancer in addition to other solid tumors with elevated replication stress.

SGR-3515 inhibits both, Wee1 and Myt1, and concurrent loss of function of these two proteins confers amplified vulnerability in cancer cells termed synthetic lethality. This profile allows us to implement an intermittent dosing schedule, with the goal of maintaining efficacy while allowing recovery from any potential hematological mechanism based side-effects. Turning to our SOS1 program. We are very pleased with the preclinical package for this R&D ready program. Given SOS1 inhibition is expected to be used in combination with KRAS and other MAPK pathway inhibitors, we will seek to advance this program through a partnership. Within our BMS collaboration, we successfully advanced a neurology program into the latest stages of discovery. We also mutually aligned with BMS on discontinuing the immunology program on the basis of the project viability and competitive positioning of the initial target product profile.

Collaborations remain an important part of our strategy and we continue to engage with existing and new partners about future collaboration opportunities. We are excited about the expansion of our predict-first approach with the launch of structure-based models for binding to off-target proteins. The ongoing structural biology revolution is expanding access to accurate 3 dimensional structures of increasing numbers of proteins. We are leveraging this expansion to pursue broad computational off-target binding models, and have already began to see the impact on our internal discovery projects. Fully realizing the power of computational toxicology solutions is expected to further transform drug discovery. In summary, we are very pleased with the progress we are making across our collaborative and proprietary pipeline.

We are looking forward to Phase 1 data readouts from all 3 clinical programs next year, beginning in the first half of 2025. Behind these programs, we have discovery programs that represent both, first-in-class and best-in-class opportunities that can generate value through partnerships or by advancing them independently. I will now turn the call back to Ramy.

Ramy Farid: Thank you, Karen. We have made a great deal of progress in the first half of the year. We witnessed continued scale up and adoption of our software, launched an exciting new computational initiative to develop a predictive toxicology solution, and advanced our pipeline. We look forward to updating you on the opportunities that lie ahead in the second half of the year. At this time, we’d be happy to take your questions.

Q&A Session

Operator: Thank you. [Operator Instructions] And we will take our first question from Michael Yee with Jefferies.

Michael Yee: Thanks for the question and congrats on a good quarter. We had two questions. One was thinking about the outlook for the rest of the year. Tell us about your thoughts around any dynamic changes in your end user customers. Obviously, you’re essentially maintaining the guidance and we always know it comes down to the fourth quarter, so maybe just talk a little bit about the dialogue and conversations you’re having as it relates to confidence towards the end of the year. And our second question is more an important financial question. I’m sure Geoff would have been good around, and this relates to your year-end cash. We have our own estimate of about $366 million or so. But the burn is up $100 million for next year, leaving you perhaps less than a year of cash, if you didn’t have additional capital.

Can you tell us about your thoughts around where there would be options for capital over the course of the next year or so? A partnership? Other options? Tell us about that. Thank you.

Ramy Farid: Thanks, Mike. I’ll take the first question and hand it over to Geoff to either add to the response to the first question, and obviously answer your second question. So with regard to the nature of our discussions with our customers, few things; one is we’re really pleased with the level of engagement, the clear level of interest in scaling up their usage of the software, and they are excited about the kind of advancements that we’re talking about including this predictive tox project. Obviously, that won’t have an impact this year but still excitement around the science and advancement of the platform; it’s very helpful in in these engagements. We also remind you that we have — we continue to expect as we have in previous years to continue to have the very high, near 100% customer retention rate.

So we have a lot of visibility into these discussions, we know customers will be renewing, and it’s really just a matter of what the scale up might be. And as you pointed out, mostly in Q4. Geoff, do you want to add anything to that? Or — and then answer the cash question.

Geoff Porges: Yes. Thanks, Ramy. Just to add to that, Q4, of course, Mike is our largest quarter always, and that’s when our big customers renew. And I think the guidance range does reflect that. We don’t know exactly how much those customers are going to scale up their use and the exact nature of the contract; how much service there is or maintenance we are providing, how much of the mix of the revenue is going to be ratable versus on-prem. And that’s what contributes to the uncertainty in the guidance but we have a very high degree of confidence that our customers are going to renew, and all the experience that we’ve had, including the experience we’ve had in the first half of this year, is that they are likely to renew at increased scale.

With respect to the cash position, you saw us — we clearly have the opportunity and will receive the additional cash from Morphic presuming that that closes in the immediate future. And then I highlighted the continued position that we have in structure and, of course, the opportunity to realize that overtime as well. And then lastly, as our programs move into the clinic, we have been pretty clear that our intention is not to continue to take all of our programs ahead, independently. We may, under the right circumstances, advance one or maybe more programs. But obviously, given the number of programs that Karen’s team is developing, we are expecting to transact on some of them. So, now we believe we have a number of opportunities for partnering.

And we also, as Karen alluded to, continue to be actively engaged in discussions about collaboration. So, both of those are additional sources of capital to sort of more or less match what we expect to be the cash burn going forward.

Operator: Thank you. And we will take our next question from Manny [ph] with Leerink.

Unidentified Analyst: Thanks for taking the call. Congrats on the quarter, I want to drill down a little bit on growth in the software-based business. You’ve given some clarity around continuing investments and engagement with your large customers around deepening and expanding use and as one driver of growth. Can you give us a little bit of clarity on what you’re seeing in end-market use amongst smaller biotech companies, new founders, newly founded companies, venture funding companies, etcetera? And to what extent we’re seeing the state of biotech capital markets reflected in your new customer ads?

Ramy Farid: Yes. Geoff, you want to answer it?

Geoff Porges: Yes. Sure. Thanks for the question, Manny [ph]. Look, I tried to sort of allude to that in my prepared remarks. There’s no doubt it’s a turbulent environment for emerging biotech companies but there are still new companies being formed. And the new companies formed, tend to be looking at the most cost-effective way to come up with novel molecules against their favorite targets. And so very frequently, they are reaching out to us about becoming a software customer. And so, when I look at the sort of customers that are scaling back or stopping use of our software because they’ve said we don’t do drug discovery anymore, we’re just advancing Phase 1 or Phase 2 program. When I look at that number of customers, and that revenue effect compared to the opportunities that we’re seeing, we’re still seeing more new accounts than accounts that are not software customers, and we’re seeing significantly more growth accounts than we are seeing reducing accounts.

So I would say, it’s definitely an effect, it’s a fairly modest effect. Now our business is really being driven by the large customers but it’s an effect, I don’t think that it’s increasing. I think that we’re weathering it and you’re sort of seeing us absorb it in the results and the guidance that we’re providing. And I think that there is some signs of optimism in the green shoots that I’m alluding to, and the company is reaching out to us and saying we’d like to become customers. But it’s a little hard to say that that’s going to sort of result in a real recovery in the growth contribution of that customer segment which really has not been there in a large way since sort of 2021 or early 2022.

Operator: Thank you. And we will take our next question from Scott Schoenhaus with KeyBanc.

Scott Schoenhaus: Thanks, guys. Geoff, I guess this is a question for you. You following up on new initiative to expand the application of computational tools for predictive toxicology. When exactly will this launch? When can you start taking orders? When should we start seeing the incremental revenues flow through? I believe you said in your opening comments next year. And then I have a second follow-up on this one as well.

Ramy Farid: Let me take that first, Scott. Thank you. Yes, we’re obviously extremely excited about this project. We’ve already gotten quite a bit of feedback from pharma companies that we’ve discussed it with and it’s been universally positive. And we’ve also as we’ve indicated, started to use this technology internally in our collaborative programs, in our proprietary programs, and getting some very nice results with it. Now with regard to when it will be available to customers, that’s something that we’re building out now. We obviously are very excited to have it funded by the Gates Foundation with that $10 million grant. We expected sign-up additional partners to help work with us on the project; that’s what we’ve done in the past with sort of new and novel technology.

And we can’t necessarily get into details about exactly when it’s going to be released; we’ve never done that, that’s not a smart thing for any software company to do. But I think you can tell from how we’re talking about it that this is something that will be released in the near future, let’s put it that way. And we certainly expect it to contribute — continue to contribute to our software revenue growth over the coming years as we add more and more off-targets to the handful that we’ve already done.

Scott Schoenhaus: That’s really helpful color, Ramy. Thank you very much. I guess just to put a pin on the back half software revenue guidance since it’s been picked and poked at earlier — with earlier questions. I guess, is there any change to the cadence of the back half from your — from last quarter, even say, based on all that color that Geoff provided on the customer, the green shoots and you’re seeing more increasing scale of established customers. Has anything changed over 90 days in terms of the cadence of the back half or the contribution of the back half in terms of software revenue guidance?

Geoff Porges: Ramy, do you want me to jump in [ph]? I don’t think so. All I would say is, I think the middle part of the year is proceeding a little better than we anticipated. We are seeing some encouraging signs in terms of, as I mentioned, both global accounts and established biotech customers scaling up their renewals and that’s kind of helping us in the middle part of the year, but the full year result still is heavily dependent upon the fourth quarter. I’ve sort of mentioned that we have a very high degree confidence about the renewal opportunities that we have and the fact that they will renew. But of course, the range of the guidance, which you can see is really influenced by the scale of the contract and the scale-up in the contract.

The length of the contract, some customers will come to us and say, we want to have a multiyear contract or we don’t. And then, of course, the nature of the contract, as I mentioned. So there are all those different elements that contribute to that variance in the guidance, but we’re very confident that those renewals will happen.

Operator: And our next question comes from Evan Seigerman with BMO Capital Markets.

Conor MacKay: This is Conor MacKay on for Evan. And congrats again on the quarter. Just one for me on the updated clinical time lines that we got today. Can you maybe just comment on how a trial enrollment has been progressing for your key assets that are currently in clinic? And maybe just discuss a little bit what drove the change from the previous late ’24, ’25 guidance to the new sort of more specific first half and second half ’25 guide? Is this just a tightening of your prior range?

Karen Akinsanya: Yes. Thanks very much for the question. Yes, indeed, it is tightening of the range. Enrollment is going well in both our most advanced programs. And as you heard, we just started dosing in our third program. We are very happy with the way things are going. And really, this is around our intention and our plan to share information at medical congress and understanding the timing of abstract submission dates. We’ve been able to now add that focus into first half of MALT1 and second half of CDC7. But overall, things going really well and pleased with the studies and the enrollment that we’re seeing.

Operator: And our next question will come from Michael Ryskin with Bank of America. Please go ahead, Michael, your line is open. And once more, Michael Ryskin, your line is open.

Geoff Porges: Mike, you might be on mute.

Operator: And we’ll take our next question.

Ramy Farid: We’ll come back to him.

Operator: We will take the next question from David Lebowitz with Citi.

David Lebowitz: At the beginning of the call, you referenced your talks with the large-scale global companies about significant scale-ups and expanded use of the software. Is that something that’s actually incorporated into the data that something we would see in the fourth — typical fourth quarter uptick? Or is that a more broad statement that talks about potential acceleration in the coming years?

Ramy Farid: Yes, Geoff?

Geoff Porges: Yes. That’s actually a really good question. Dave, I think it’s both. I think that we recognize, as you know, that we had a large contribution from a large customer renewing a multi-year contract in the fourth quarter of last year. And so we’re very focused on growing through that in the fourth quarter of this year and see a number of opportunities for doing that, which is what’s in the guidance. But also, I think from our comments around predictive tox and from our general comments, I think we think this is relatively early days in the cycle of the industry’s adoption of this technology. And we have had an opportunity in the quarter to talk to many of the larger customers about where they are in the journey of adopting our technology at scale.

And I think they’re starting to realize that it’s still relatively early days and they have a long way to go in terms of capturing the full benefit. Ramy, do you want to talk a little bit more about the long term and the discussions we’ve had?

Ramy Farid: The long term, that’s right. The question was specifically about — right — dated [ph] about this year in the fourth quarter, correct? Or did I mishear?

David Lebowitz: I basically meant to the extent that are we talking about substantial upticks in the use of the software. That’s something that’s already incorporated into the numbers? Or are we talking on a broader scale of them actually increasing their overall adoption going forward and how that might affect future years?

Ramy Farid: Yes. Okay. So yes, I think Geoff covered it very well, but you’re raising something very interesting, which is that as customers scale up the usage of the software, they see an acceleration in the impact that it has and that fuels more growth. We’ve often talked about this sort of chicken and egg problem that you have to have — you have to be using the software at some sort of critical level of usage to really see the impact. So that’s what we’re finding is that as customers scale up, they finally get to the point where they see the sort of profound impact on their projects, the kind of thing that we’re seeing internally and that our partners are seeing and you can see the success of our partners and the impact that’s having.

And like I said, that continues to fuel further growth. And also, as Geoff said, we should be very clear and we’ve said this many times, we’re excited about where customers are going. And we, at the moment, have customers that are really using the software at a nice scale, but they still have a long way to go to really use the technology to its full — to realize the full benefit of it. And as we’ve said many, many times, to use it at the scale that we’re using it in our collaborative and proprietary programs. I think that’s what you’re getting at, and it’s a great question.

Operator: And our next question comes from Vikram Purohit with Morgan Stanley.

Unidentified Analyst: This is Morgan Greg [ph] on for Vikram. I have 2 questions related to the MALT1 program. First, for the initial MALT1 data expected in the first half of 2025, could you provide some more color on what this data set could look like in terms of number of patients level, follow-up and parameters of data you anticipate reporting? And then following this data readout in 1H ’25, are there any potential plans to establish a partnership? Or do you envision maintaining proprietary ownership of the molecule until additional data readouts?

Karen Akinsanya: Yes. So as we’ve described, well, this Phase 1 dose escalation in patients, we are very focused on understanding the safety, the pharmacokinetics and pharmacodynamics. I think we said in our prepared remarks, we’re currently comparing once daily and twice daily and collecting pharmacodynamic data. And so this first half of 2025 that will really be giving insight into what we’re seeing in that Phase 1 dose escalation study. We are obviously also in an exploratory way looking at clinical activity and we also hope to be able to present that. And so looking forward to the opportunity to bring together and accumulate all of that data to share in a peer-reviewed setting. In terms of what our plans are for MALT1, I think we’ve been really clear in the past that we think MALT1 is a mechanism that will combine well with some pretty well-established drugs, BTK inhibitors, BCL2 inhibitors in a range of lymphomas.

And as a result of that, we think that actually the broadest potential development of MALT1 inhibitors will be in partnership with companies that own those assets or products actually on the market. And so while we will continue to study our molecule internally, we are always talking to partners about our projects and we think that there will be an opportunity with a more complete data set to socialize that with pharma and potentially partner the program as we conclude the Phase 1 study.

Operator: And we will take our next question from Matt Hewitt with Craig-Hallum Capital Group.

Matt Hewitt: Congratulations on the quarter. Maybe first up regarding the new predictive toxicology tools. Was this something where customers had come to you looking for help? Or was this you seeing an opportunity in the market given your strength and basically putting this together, I guess is the first part of the question? The second part is as this gets adopted, as you launch this, this gets adopted, do you see that changing the methods for the traditional tox studies? Will there be a cost benefit to utilizing your software versus the traditional methods? Or is this in addition to those traditional tox studies?

Ramy Farid: Thanks, Matt. Those are great questions. So first of all, with regard to your first question, absolutely both. This is widely known in the industry as a very serious problem. And customers and everybody that’s doing drug discovery, they’re complaining about this forever. And it’s a problem that everybody encounters in their drug discovery projects. So what happened is that our technology at a number of different pieces of the technology actually got to a certain level of performance that we recognize this seemingly impossible problems. It seemed like an impossible problem a few years ago, all of a sudden seem possible. It’s a very ambitious project, but we think, given what we’ve already done with a number of targets that it’s now the right time and we’re excited to not only — where the science has gotten to and not only the funding that we talked about before, about our partnership with NVIDIA, which is also really making this possible.

The performance of computers, even just a few years ago was not where it is now and where it is now is what’s making this even possible. So it’s bringing together obviously the interest and the demand for something like this with advances in the science and advances in computer hardware. Now with regard to the traditional methods. There are basically 2 methods that are used now. One is obviously experimental. And that’s what most people are doing. And of course, you can imagine that’s pretty slow, not very high throughput and obviously costly even to actually make the molecule and run it through panels of our targets one at a time to determine the sort of profile of the molecule with regard to off-target toxicity. The other approach has been one that involves using purely machine learning.

So this is basically collecting as much experimental data as possible and then just building ML models. The problem with that is that has a severe limitation, which is that you can only get reasonably reliable predictions for molecules that look a lot like the molecules that have already been synthesized. What that means is that these predictive computational methods only start to work very late in projects after you accumulated a ton of experimental data. And that obviously is missing a huge opportunity to catch these problems early in projects with the physics-based approaches that we developed and published that allows for any novel of any molecule that can be completely novel. There doesn’t need to be any experimental data existing yet to be able to accurately predict whether it binds to a panel of off targets.

So that’s what the sort of breakthrough is. There is never going to be a complete replacement for experimental methods. Eventually, you have to do the experiment. But the idea behind this technology is that it will significantly reduce the need to do those and it allow for prediction of this toxicology profile very early in projects and actually be able to fix it instead of just throwing your hands up and saying, “Oh, well, this is yet another series that’s dead.” Hope that makes sense.

Matt Hewitt: Absolutely. That’s very helpful.

Operator: And our next question comes from Chris Shibutani with Goldman Sachs.

Chris Shibutani: Two questions, if I may. In terms of how you think the gross margin is being improved and strengthened by some of the shifting of the allocation of employees that you described, can you give us a sense of where you are in terms of the rightsizing or the right distribution or shaping of the distribution, so to speak, so that we can think about where on the forward, this may continue to provide some gross margin benefit? And then secondly, I’m curious to know, particularly as Karen and her team continues to advance proprietary pipeline into the clinic, you guys, in essence, represent the sort of the leading end of the natural class or the ability to apply artificial intelligence type informed modalities. And we often observe and wonder whether or not in the clinical development stage, there is a smarter, better, leaner, quicker way of approaching it.

And if so, what is it that you guys are doing? And is it something that you’re even contemplating, expanding your own portfolio with? Just trying to understand if you could be the best-in-class example of utilizing the mix and that [ph], taking drug from discovery all the way through the clinic. What happens in the clinical development phase which you might get now [ph]? Thank you.

Ramy Farid: Thanks, Chris. Geoff, would you like to take the gross margin question first?

Geoff Porges: Sure. Chris, thanks for your question. The gross margin, as you kind of gathered from my prepared remarks, on the drug discovery side, there has been a reduction in the cost of services associated with the drug discovery revenue as some of the project teams that were working on collaborations have shifted over to work on proprietary programs. And so the expense associated with those employees, but also any CRO services or other services associated with their work has shifted from cost of services over to R&D. I don’t think that there is a lot more to shift there. And as I mentioned in my prepared remarks, the outlook very much depends upon what we do with respect to collaboration. So in future periods, if there is another collaboration that engages a number of those employees, then that shift might move back the other way.

I don’t think it’s going to be a seismic shift back the other way, but as it has been over the past few quarters, could it trend back the other way? Yes, somewhat. The larger question, of course, related to R&D spend. You saw R&D was flat Q2 over Q1. And I think we generally feel that we’re in a really good place in terms of the investment we’re making in the platform and the investment we’re making in proprietary medicines. Depending upon what happens with the clinical programs and their progress and what we decide to advance, et cetera, there could be a step-up associated with the clinical side. But I don’t see that as being an imminent step-up because we’ve given you a time line for seeing the clinical data. And so we’re continuing with those current trials to turn over those cards.

So I don’t think that there’s a big change in either the R&D line or in the cost of services. And then lastly, I think I alluded to the puts and takes on the gross margin line on software, where hopefully, I managed to explain that there will be a temporary reduction associated with the funding for the predictive tox initiative and that that will reverse itself at the conclusion of that funding. That’s just an anomaly of the different profitability between a software research project and a software contract with the customers. So hopefully, that’s clear. And Karen, do you want to talk about AI?

Karen Akinsanya: Certainly, yes. So you are quite right that obviously, as we move our programs into the clinic, our mindset is predict first and be as efficient as possible. We, ourselves, are not building those types of capabilities. As you know, our platform is very much chemistry-focused. However, we are engaging with companies who have built those kinds of platforms and evaluating whether they are additive to the sort of regulated way of doing things, can these help accelerate or improve efficiency in how we run trials, including things like patient selection and site selection. However, at this point, that’s not something that we will be focused on building internally. I just want to be clear with that.

Operator: [Operator Instructions] We will take our next question from Steven Mah with TD Cowen.

Steven Mah: Congrats on the quarter. Just 2 quick follow-up questions on the predictive toxicology initiative. One, does the Gates Foundation have any downstream right for anything you or your partners discover? And two, is this software applicable to both small molecules and biologics?

Ramy Farid: The short answer to the first question is no. With regard to the second question, the technology at its current stage being developed for prediction of binding to off targets by small molecules. So that’s where the focus is. Toxicity of antibodies is associated with — not with this sort of mechanism of off-target binding. The toxicity is often associated, for example, with immunogenetic responses. So it’s a different kind of problem, yes. That’s why the focus is also [indiscernible].

Steven Mah: Okay. Yes, that’s what I figured out. Thanks for your clarification. And then maybe staying on biologics; could you give us some color on the live design for biologics? What has the traction been since the launch in March? Thank you.

Ramy Farid: Sorry, what was the — yes, what was the question?

Steven Mah: The live design for biologics you guys launched that in March, the enterprise software.

Ramy Farid: Yes. Okay, yes. And how is it going? Yes. The reception has been fantastic. This is clearly a product that was — sort of had a lot of pent-up demand for it. And all of the demos that we’ve been doing in discussions with customers, again, the reception has been really, really great. As with enterprise software, the fact that it is so sticky once it’s in, also results in the time to adoption being a little slower. This is — it’s not expected that the customers just sort of hear about it and the next day, they’re using the software. The sort of sales cycle is a little bit longer with enterprise software. Again, the result of that is extreme stickiness once it’s developed. So what we can tell you is we have had customers using it. And again, the feedback is quite positive.

Steven Mah: Great. Thank you.

Operator: Thank you. I am showing no further questions at this time. That concludes today’s call. You may now disconnect.