For the management presentation for Twist Bioscience. My name is Matt Larew. I cover the company here at William Blair. Pleased to be joined this morning by Dr. Emily Leproust, who's the CEO, and CFO, Adam Laponis, as well as Angela Bitting from IR, and many other things at the company. Before we get to the presentation, I want to mention two things. First, the breakout is in Jenny B upstairs. You can follow us. Second, I have to inform you for a complete list of research disclosures or potential conflicts of interest, please visit williamblair.com. Again, thanks to Twist for being here today, and I'll turn it over to Emily.
Thank you very much, Matt, for the invitation and the introduction. It's my great pleasure to tell you more about Twist. I'll start by saying that I'll be making some forward-looking statements. At Twist, we are a company that builds DNA from scratch and provides customized solutions to companies on the biological continuum. We are a real company. We had almost $111 million of revenue last quarter in Q2. We have 1,000 employees, and we are serving a number of markets throughout the world. The key advantage, the key technology and differentiation that we have is that we're using a silicon-based technology to do what we do. We are grouped into two product groups. One is our DNA synthesis and protein solution, and the other is our NGS applications. They both use the same silicon platform.
The main market that we are serving is the drug discovery or therapeutics area, the production of chemicals in a sustainable way, the diagnosis of diseases, where we provide reagents and kits to our customers that are providing that diagnostics, mostly in oncology. Our products and services are being used for the development of traits to improve food security. We are trusted by some of the biggest diagnostic therapeutics, biotech, and universities in the world. As I mentioned early, the foundation is in the center. It's our silicon chip. On our silicon chip, we can synthesize, we can print 1 million oligos. An oligo is a small piece of DNA. In the same footprint, that 1 million oligo synthesis, our competition can synthesize 96 oligos. We have about a 10,000 times advantage compared to other technologies.
It's a center of our strategy, which is if you think of the silicon chip as a plane, we want to run full planes. We are launching more products that start having more and more applications, that are serving more customers and enabling more markets. That has been our strategy over the years. When we get to the financial slides and you see things going up and to the right, that is that strategy of loading more on the chip. This is a very important slide from us. On the bottom left, you have the silicon chip, so these oligos. From those oligos, we can assemble them, we can amplify them, we can use enzymes to make a large number of products out of those oligos. That picture has evolved over time.
A few years ago, we only had a few products, but we've been adding products after products. This slide is really a key piece of our strategy, which is the new product introduction strategy. In addition to that, we've kept improving on the silicon chip. In the middle here, you can see that it used to take us 51 L of reagents to make 1 million oligos. Right there, that was 99.8% less than the competition. Even though we were already great in 2023, actually, we've reduced the amount of reagents that we need down to 14.4 L, so that's 70% reduction in reagents. That has led to a 60% reduction in cost. At the bottom right, we've also accelerated the speed.
It used to take us 26 hours to build 100 bases of DNA. Now it takes us four hours, sorry, seven hours, which means that on the same machine, on a CapEx that have already been deployed, we've been able to increase the capacity by a factor of four. Not only we launch new product externally, but internally we keep improving. Another example of that is this is an example of the back end. If the picture looks blurry, it's on purpose. We don't want our competition to be able to see it. You're welcome to come see it by yourself. This is an example of from the oligos that we release from the chip at the top left, we can assemble them into fragment. If you go to the right, we can clone them.
If you go then to bottom, we can sequence them to make sure to find which one is perfect, and then moving to the right, we can do miniprep to send the exact flavor of the DNA that customers wanted. That is fully automated, and you can see at the bottom right that we keep adding more and more automation to support more and more of the products. Not only we can automate, we can automate the automation. This is a schematic of one lab. We created Gene Lab 1, where we used to have capacity to make 3 million fragments a year in two times 950 sq ft. We've automated the automation to integrate them together, and now we can do double the capacity in one-fifth of the space.
We increase the capacity, we increase productivity, we reduce headcount, and we have more space available for the next generation of products. It shows you the kind of innovations that we're putting in place to be able to serve more and more customers. At the same time, as we add capacity, we're also lowering the turnaround time because our customers have a need for speed, and we are very happy to provide to them. When we launch a product, over time, we're able to take time away from the production to provide those products and services faster and faster to our customers. All of that innovation has enabled us to grow our markets. In 2020, our markets were at the SAM of $2 billion.
By 2030, we think the SAM for the NGS application on the left will be $6 billion, and for DNA synthesis and protein solution on the right will be $7 billion. As we launch new products, as we improve our capabilities, we extend our SAM, and we're going to be able to grow faster than the overall market rates. Of particular note, two things. We are particularly excited on the left with the oncology diagnostic market. We think that that is going to be a great source of growth for us. On the right, we think that AI-driven drug discovery will add $1 billion to the markets that we served.
In addition to that, we think that there's another $1 billion in nucleic acid therapeutics, where every patient gets their own drug, so truly personalized medicine through gene and cell therapy, for instance, or for personalized mRNA therapeutics, and we think that will add another $1 billion to our opportunity. This is just a few examples of some of the products that we sell, and some of them are growing very, very fast. In this case, from the first half of 2025 to the first half of 2026. At the JP Morgan conference, I provided different numbers from other products. All of that to say is that we are not a one-trick pony. We have thousands of SKUs, and the combination of all those SKUs not only gives us top-line growth, but durable top-line growth because our business is very resilient.
From our customers' point of view, we're in the business of making happy customers. In the top left, we provide them solutions that they can't find from other providers with a customer experience that is exceptional. It's not only a question of selling, you have to deliver what they want. We have invested in operational excellence. We have invested in deploying automations to be able to serve more of our customers at great quality without adding headcount. That drives into the bottom right in our financial discipline. You see that over the last few years, we've increased our revenue, we increased our gross margin at a pretty flat OpEx, which means that by September of this year, we have a very strong commitment that we'll reach adjusted EBITDA breakeven. That adjusted EBITDA breakeven is really our North Star at this point, and it's in sight.
We're just a few months from being there. It's really driven by all of the work that we have done to leverage that semiconductor model that we have. The fixed cost is high, the variable cost is low. As we add more and more products, as we are able to sell more seats on the plane, the top-line growth gets improved, and we get the profitability. Looking at those numbers in more details in terms of a yearly view, over the last six years, we had a CAGR of 33%. Last year, we grew 20% as a business. We've been able, over the last few years, to go from a 32% growth margin to a 52% or more this year. That comes, again, from a very diversified revenue base.
As you can imagine, if you ramp revenue, you ramp growth margin, and you are fairly disciplined on your OpEx line, as you can see. That is what gives us the strong confidence that we'll achieve that adjusted EBITDA breakeven, and then we can deliver profitable growth beyond that. Clicking on and looking by quarters. Two big categories of revenue growth from us. One is the NGS application, and the other is the DNA synthesis and protein solution, both of them contributing very strongly to the growth. We are now in 13 consecutive quarters of sequential revenue growth, and we want to keep it that way. In terms of our industry segments, in January, we've introduced a new way of splitting the different industries that we serve to better reflect or to give more visibility to our investor base.
In previous years, at the top, healthcare was the dominant industry. It didn't show how good we are doing in both diagnostic and therapeutics. We've now split, and you can see that last year, diagnostic and therapeutics were the two biggest industry. Looking at the last quarter that we recently reported, therapeutics was almost $41 million of revenue, growing 55% year-over-year, very significant growth. Diagnostics was $40 million of revenue, only growing 14%. We've also signaled that the diagnostic growth is going to get back to 20% growth by Q4. All the way to the right, there is a global supply partner group segment. This is sometimes a little bit misunderstood, I have a slide on that. What we do is we enable, sometimes, our competitor to resell DNA that we make under their brand.
We make it, they sell it. That is a really great source of revenue growth for us at a very attractive financial results for us. On the right, we also have distributors. We are direct in many countries, but in APAC, distributors are actually very useful to us. Maybe digging in a little bit in the two different product groups that we have. The first one is DNA synthesis and protein solution. Last quarter had a 28% year-over-year growth. There we mostly serve companies that are doing drug discovery. It's either drug discovery where we sell products and they do the work themselves, or drug discovery where we sell a service to them. We have made a very strong effort in drug discovery to have a full menu.
What we know is drug discovery customers, they all want to do something different. Our strategy and approach of having a very broad menu where no matter how they do discovery, we'll be able to serve them, has been very successful. To give you a sense of the type of financial partnership that we have with our customers here, we are showing two different customers who, at the top, are large biotech, and at the bottom, a large pharma customers. We won't go into the details, but what's important to show is that the colors are different from one customer to the other. They are both discovering drug, and yet they are purchasing different products. Those colors change over time. At the top, that customers was using Gene Fragments in purple, and then over time added, in green, Clonal Genes.
At the bottom, another customer doing kind of the same thing was mostly doing the opposite, which is using Clonal Genes and then moving to Gene Fragments over time. You can see in yellow, when we launched our IgG product line, which is instead of selling DNA, we are selling protein, that customer embraced the switching to Twist to purchase protein. Those are two examples of customers. At our investor day a couple of weeks ago, we provided more data for five additional customers that are, in that case, only doing AI-enabled drug discovery. Again, you can see that every customer has a different approach to engaging with us. That's key to us, is we're able to meet customers where they are. We're not telling people how to do science.
Instead, we are asking people, "How do you want to do science?" Then we enable that. That is a great reflection on the platform that we have built, where no matter what, no matter how you're doing your science, no matter how you want to engage with us, all of those orders, they end up on the same silicon chip. We can serve you at scale and serve you your customized solution. AI and AI-enabled drug discovery has been a great wind in our sails. In 2025, out of the $66 million order growth that we had, $25 million came from AI drug discovery. We are very happy with AI drug discovery, in the next few slides, I'll give you more details. I also want to say that we're doing really well in traditional drug discovery, and actually are becoming a market leader in drug discovery.
Digging a little bit more into AI, what our customers are doing is, at the bottom, is the cycle of design, build, test, learn. They design some sequence. Someone has to build those sequences, someone has to test them, and then from the results from the data, they can learn, build a model, and then go back through that cycle to either improve the model over time or turn the crank. Say, "I have enough. I've built enough. Now I want to declare a new target", goes through that process of discovering drugs through AI. There's a number of workflows at the top, from Workflow 1 to Workflow 4. I won't go through all the details. Again, it's important that it's not a one-size-fits-all for all customers. In general, I'll use the example of Workflow 1 on the left and the example of Workflow 4 on the right.
In Workflow 1, customers want a pooled discovery. The AI algorithm that they use comes out with hundreds of thousands, sometimes millions of different sequence. On the silicon chip, those sequence get synthesized as a pool, then they get expressed in a protein as a pool, then they get tested, either through phage display or yeast display. Then on the other right, on the right with Workflow 4 is kind of the other opposite of how to do it. That is an approach where every sequence is studied one at a time. There, they give us a sequence, and we provide to them an end-to-end, where for every sequence, we give them a full characterization of the functionality, the affinity, the developability. There's pros and cons to both.
We're happy to serve, again, customers either way they want to do their data science. The key, though, is that for the building and the test, for the wet lab part of either way you want to discover drug, we are that partner to do the building and the testing for them. Moving on to the second product group, our NGS application. It grew 12% year-over-year last quarter. Again, we think that that will get back to 20% year over growth by Q4. There, we also have a full suite of products and services. All the reagents that you need from the sample to the sequencer, we're able to provide to you. Mostly, we are serving the oncology and diagnostic market with either screening, therapy selection, MRD.
The typical workflow to win one of those big diagnostic partner is to either start with a pilot, an R&D pilot, or even sometimes a service lab where they have samples, but they don't even have time in-house to try the technology. They send us samples. We do the work for them just to show you what the data could be like. Once they are convinced with the quality of the data, they can move up to validation and verification. They can do commercial studies. That can be a very substantial source of revenue for us. Once they get the sensitivity and the specificity that they want, and once they get approval, then they can go launch commercially. For every patient that gets tested, there is some reagents from Twist that is being used. Therefore, it's a very lasting relationship.
Those customers are pretty much locked in, because once they've done the clinical studies with us, it's very hard for them to move to a different supplier. The types of relationship that we have varies over time and by customers in terms of the flow of revenue. Here, showing you four different customers, from a tumor-informed MRD customers in the top left to an oncology diagnostic customers that has multiple tests with Twist on the top right to a rare disease customers in bottom left to an oncology diagnostic customers in the bottom right. You can see that also in this case, those customers buy different things from us. Some buy a fixed panel, some buy custom panels, some buy more library prep, some buy less.
Again, it's very important for us to meet customers where they are, because even in the area of cancer diagnostics, actually, customers do things differently. In general, we do like to sell library prep with panels, but sometimes it's the case, sometimes it's not. We're always there for the customers. What we are able to do is to support our customers in the continuum of cancer care that they're providing. If you follow the patient journey, starting from the left, there's a screening, and then there may be an early detection and profiling. There may be a therapy selection. After the initial treatment, when the tumor load goes back down, MRD becomes very important to decide whether that person needs chemotherapy after surgery.
There may be a therapy during that time as well, and you need more minimal residual disease to detect whether there's a relapse. During the patient journey, there's going to be many different needs, and that is why we have a large number of products to support all of that, both in the diagnostic side but also in the therapeutic side. I mentioned that new drug modalities such as cell and gene therapy, mRNA, are going to be key personalized therapies to patient. Twist will benefit because we'll be the payload in each of those patients. Looking forward, what are we going to do? Well, we're going to continue what has been working for us, which is continuing our strategy of loading more on the chip.
There's new products that we are launching, such as mRNA, such as more complex genes, MRD Express, also more product to support the discovery of antisense oligonucleotide or siRNA, which are new modalities. We will apply those products to enable more applications, reach out to more customers, and enter more markets. Again, it's all leveraging the investment that we've made in the silicon chip. The second thing that we'll do is really the three pillars of our success. The first pillar I mentioned in the past previously in my talk was the launch of new products. That NPI machine has been key, and we'll keep exercising that machine. The second thing is our operational excellence, being able to deliver the customized solutions that our customers want and ramp and grow with our customers.
At the same time, do the R&D to improve the speed, improve the capacity, lower the cost internally. The third is our commercial execution, as I often like to call it, our commercial violence, to make sure that all the potential customers get access to the Twist technology. With that, I think this is my last slide. I think we have a very compelling upside ahead of us. We have a highly differentiated platform. We are launching very differentiated products to serve a growing market. We are taking market shares in that market. We are extending into new market opportunities. We have existing customers, but we are landing more and more customers thanks to our commercialization approach.
We've been able to deliver so far revenue growth, but we want that revenue growth to be durable, and that is happening with everything we've been doing at more than 50% gross margin and with a path to get to 60%. We have a line of sight for adjusted EBITDA breakeven in Q4, so just a few months from now. We'll keep doing it. We will keep doing that NPI launch, operational execution, and commercial execution to have our numbers going up into the right. With that, I think I'm out of time and ready to walk to the first floor to answer some questions. Thank you very much