Basecamp Bio

Company’s multi-year biodiscovery programme, driven by strategic economic partnerships across 26 countries, has resulted in BaseDataTM, the largest and fastest growing database of new biological protein sequences - 9.8 billion - ever builtNew species data gives scientists new potential therapeutic solutions to areas including antimicrobial resistance, gene editing, and environmental sustainabilityScientific preprint demonstrates how BaseData breaks through biology’s data wall and now exceeds the size, growth and diversity of public datasets that serve as the foundations for AI applications in the life sciences industry LONDON and CAMBRIDGE, Mass., June 11, 2025 (GLOBE NEWSWIRE) -- Basecamp Research, an AI company dedicated to using nature to solve the most pressing challenges in the life sciences, today announced the discovery of over 1 million new species as part of BaseDataTM, the world’s largest biological protein sequence database, and the first to be purpose-built to power the next era of generative biology. “The rise of generative biology – using AI foundation models to design, generate, and annotate proteins, pathways and therapeutics – creates unprecedented demand for large, diverse biological sequence databases,” said Glen Gowers, Ph.D., CEO and Co-Founder of Basecamp Research. “The million plus new species we’ve identified to date gives us an unrivaled understanding of how life on Earth has evolved over billions of years. We believe it will help the life sciences sector overcome today’s critical data bottleneck and build a truly generative biology ecosystem with applications across nearly every area of human and planetary health.” Today, research relies heavily on public, open access biological sequence databases. Designed originally as academic collaboration tools, these databases are accessed over 100 million times per day and are the source of over 50% of all life science patents. This usage has grown by 10x in the last decade and continues to accelerate thanks in large part to the increasing use of data-hungry AI models in biopharma research. However, compared to all life on Earth, these databases are incomplete, riddled with redundancies, and highly-biased – 70% of all sequence data comes from just 10 species. Furthermore, the information in these databases is growing slowly, and companies using these databases for commercial purposes face increasing scrutiny on various legal fronts. This lack of growth and diversity holds back model performance and is a fundamental problem that slows research advancements in the life sciences: the data wall. Basecamp Research has broken through this data wall by pioneering an economic partnership-based model that incentivises the collection of samples across the planet’s most extreme and biodiverse environments, working with 125+ communities in 26 countries. In a milestone disclosure, the company is unveiling the results of its biodiscovery initiative — BaseData has identified 9.8 billion new protein sequences. Once redundant sequences are removed, BaseData is currently over 10x bigger than all public databases combined. It continues to grow rapidly, offering a fundamentally new understanding of life on Earth. “Biological foundation models are the key to continued progress in the life sciences, but their growth is slowing and performance is suffering,” said Oliver Vince, Ph.D., Co-Founder of Basecamp Research. “It’s this data wall that discourages life science teams in building and using huge biological foundation models. At Basecamp, we’ve demonstrated a new, scalable economic framework for expanding our knowledge of life on Earth and we show it is possible to overcome the data wall that’s limiting progress for others in bioAI. Training our own foundation models on this data is enabling collaborations across the life sciences sector to help solve pressing challenges in drug discovery.” Basecamp Research scientists have uncovered more than one million new microbial species to date in some of the planet’s most remote and extreme environments, expanding the view of the tree of life by over 10 times. The previously hidden biological insights in this data, combined with foundational AI models, will help shape everything from environmental sustainability and repair to therapeutics development. Examples of such newly discovered species include: On a World War II shipwreck, a new species of Burkholderia, a type of bacteria known for its ability to remove heavy metals from the environment. This could help improve pollution control and deepen our understanding of how bacteria resist antibiotics.In acidic hot springs near an active volcano, a new member of the Sulfolobaceae family that thrives in searing temperatures. Its specialized proteins, stress-response systems and stability near boiling point could help develop new ways to deliver medicine or preserve biological materials under harsh conditions.In Antarctic soil, a new species of Candidatus Eremiobacterota, a bacterium that survives by drawing nourishment from the air, generating its own water using hydrogen as an energy source — a finding that could inform novel gas-based drug delivery systems or therapeutic approaches. Basecamp Research is sharing further details in a pre-print paper and plans to offer early access to its unique data to life sciences researchers who express interest via its website www.basecamp-research.com to help them further their research. About Basecamp ResearchBasecamp Research is solving the most pressing challenges in the life sciences by exploring Beyond Known BiologyTM. We build foundational AI models on top of the world’s largest, ethically-sourced database of biological information to give AI the most complete understanding of biology ever. This allows us to design more complex biological systems with performance improving dramatically as AI sees more diversity and context. Basecamp Research collaborates with biopharma companies and academic research institutions to design novel protein sequences and biological systems that can transform therapeutic research and development. Our team of explorers, scientists and policy experts proudly work with more than 100 biodiversity partners across the globe, allowing us to deliver breakthroughs that can have a profound impact on healthcare and the lives of patients. Fast Company named Basecamp Research one of biotech’s top 10 Most Innovative Companies in 2025. For more information, visit basecamp-research.com. BaseDataTM and Beyond Known BiologyTM are brand names and technologies of Basecamp Research. Media Contact:Adam SilversteinSCIENT PRadam@scientpr.com Photos accompanying this announcement are available at https://www.globenewswire.com/NewsRoom/AttachmentNg/5597a489-463d-40e5-9fce-0064cc18a55e https://www.globenewswire.com/NewsRoom/AttachmentNg/98e6c5ce-da20-4bae-ad20-76687f9ddbc6 https://www.globenewswire.com/NewsRoom/AttachmentNg/61aeb561-580d-4009-bd50-95e4db6cfebf https://www.globenewswire.com/NewsRoom/AttachmentNg/fd0cde34-a7f2-4b1e-9d32-5d728a23e880 https://www.globenewswire.com/NewsRoom/AttachmentNg/44187ee1-3843-4ddf-91e7-2ce1979017b4 https://www.globenewswire.com/NewsRoom/AttachmentNg/ad51db49-1170-47da-b785-b96762ce198a

Fyodor Urnov had given a version of his speech dozens of times before. Yet as he leaned into the microphone to address the FDA’s top regulators late last year, he felt his heart racing. The agency officials were there to hear about a slow-burning crisis: What do you do when the root cause of a disease is known, the treatment is clear, but drug companies won’t touch it? It was November 2024, and in a Marriott conference room in Bethesda, MD, Urnov described a tongue-twisting and incredibly rare condition called familial hemophagocytic lymphohistiocytosis, or HLH. Caused by one of several genetic mutations, it overactivates the immune system in young children. Without a stem cell transplant, the children die. Even with one, the prognosis is poor. “The single thing I need you to understand,” Urnov told the FDA officials in the small, packed room, “is that 40% of infants with inherited HLH do not survive on standard of care.” Scientists have documented more than 7,000 rare diseases — the vast majority of which affect children, are caused by genetic defects, and have no good treatment. Since the invention of CRISPR gene editing in 2012, the toolbox for manipulating genes, and maybe treating those many diseases, has exploded. But the number of gene editing cures has not. Urnov, more than anyone, had been warning that there is a growing gap between what’s possible, and what drug companies find profitable enough to pursue. His alarm has only grown more urgent as biotech startups set up to go after those diseases have made layoffs or shut down. Many of the ones who have survived the sector’s ongoing downturn have pivoted to more common — and more lucrative — conditions. CRISPR’s potential, he argues, is being wasted. “I have lost all interest in making discoveries with potential ,” Urnov told Endpoints News . “The bleak truth is that we don’t need new gene editors, we don’t need new delivery modalities to treat 500 inborn errors of immunity. They’re completely treatable using existing tools and we have multiple capitalized for-profit biotechs that could be treating them — but they’re not.” This is the second in a series of stories from Endpoints News that looks at the future of CRISPR and its impact on patients, science and business. The first story in the series is “ Will CRISPR matter? ” For many genetic diseases, Urnov, a University of California, Berkeley professor, believes that science is no longer the limiting factor in making new medicines. In October 2021, he laid out the biotech industry’s mounting problems in an opinion piece for a scientific journal. He argued that it was technically possible to rapidly develop gene editing therapies for most genetic diseases, and outlined the steps it would take to go from diagnosing the cause of a condition, to making a custom therapy in two months. “This status quo is not acceptable. Patients dying of N = 1 genetic conditions cannot wait for the for-profit sector to work out its business model,” he wrote. His solution was to get regulators to start treating CRISPR as a “platform” — industry jargon for a technology that can be used to make treatments for many different diseases — instead of having to treat each use of the technology as a separate drug, with separate preclinical work, trials and regulatory applications, all of which add time and cost. The platform approach was baked into CRISPR from the start: A pair of molecular scissors called Cas9 can be reused for many purposes by easily reprogramming a synthetic molecule called a guide RNA that tells the scissors exactly where to cut. The gene editing toolbox has expanded beyond scissors, but the interchangeability of the guide RNA, the crux of CRISPR’s power, remains. Many biotech companies once boasted about the platform potential of these technologies. But those ambitions have narrowed amidst a tough funding climate and slow sales of other one-and-done cures with multimillion-dollar price tags. And when Urnov looked for a company to work with his vision of iterating treatments disease after disease, he was turned down by one after another. “The reaction was, ‘Fyodor, what you’re saying is great. Now is not the time in the life cycle of our business for us to take this on,’” he said. After a few years, he eventually found a collaborator in Danaher, a life sciences company that sells tools and manufacturing services, but doesn’t make drugs. Sadik Kassim, chief technology officer of Danaher Genomic Medicines who had previously worked at the gene editing company Vor Bio, sympathized with Urnov’s vision. “He’s really communicating the frustration of an entire field, particularly of translational researchers.” Kassim told me. In January 2024, Danaher and the Innovative Genomics Institute at UC Berkeley, where Urnov works, announced the partnership in which the company would provide expertise, equipment, reagents and an undisclosed financial contribution to develop therapies for two genetic immune diseases: HLH and a form of a devastating condition called severe combined immunodeficiency or SCID, two of the seemingly curable conditions that have triggered Urnov’s frustrations. It will take a couple years before the HLH therapy is ready to test in the first patient, but Urnov said he hoped that the treatment for a second patient, with the same disease but a different mutation, could be made in only four months. Along the way, he hopes to create a “CRISPR cures cookbook,” essentially standard operating procedures that others can follow to develop therapies for hundreds of other immune diseases caused by broken genes. Urnov isn’t trying to take traditional drugmakers out of the picture. Instead, he is trying to show academics and biotech companies that developing CRISPR cures for even the rarest of diseases is feasible. “If there are dozens of companies, large and small, that have taken our cookbook and built their own programs,” he said, “that will be the biggest imaginable win for us.” Urnov was not entirely alone in his crusade. Across the country in 2021, Kiran Musunuru had just returned to his lab at the University of Pennsylvania. After eight years working on CRISPR therapies for heart disease, including a brief stint at a biotech he co-founded in Boston, he was itching for a new challenge. He just needed a disease to cure. Inspiration came from across the street at the Children’s Hospital of Philadelphia. There, an old medical school friend, Rebecca Ahrens-Nicklas, was treating kids with rare metabolic diseases, who have small genetic typos that leave them with broken or missing enzymes. These infants are unable to break down harmful metabolites which can accumulate and cause liver failure, permanent brain damage, and even death. The challenge is that there are hundreds of ways to break a gene. Making separate therapies for those myriad mutations would mean hundreds of preclinical studies to satisfy regulators, costing an incalculable fortune. “The technology would work. What you run into is a regulatory challenge,” John Evans, CEO of the base editing company Beam Therapeutics, told me. At first, Musunuru and Ahrens-Nicklas didn’t worry about that. They simply picked a single mutation in a single disease, called PKU. The therapy worked so well in a mouse study, they began feeling an ethical obligation to develop treatments for more mutations in more diseases. The Children’s Hospital treats some of the sickest patients from around the world, so the mutations that Ahrens-Nicklas sees in her patients aren’t always well-documented in genetic databases, which skew heavily toward western European ancestries. They could pick one, or a few, mutations to work on, but they might never actually see children with those genetic glitches in the hospital. Musunuru wanted to avoid this “mutational discrimination.” “The most ethically sound way was to treat the patient in front of you,” Musunuru said. “Wherever they come from, whoever they are, whatever variant they have, whatever flavor of disease they have, let’s give them a fair shot.” Although PKU can be managed with strict diets and chronic therapies, some inherited metabolic diseases, including a cluster of six conditions called urea cycle disorders, can make babies sick from the day they are born. Even under constant supervision, they can suffer spontaneous metabolic crises that can cause irreversible brain damage. “The only chance they have at anything remotely resembling a normal life is a liver transplant,” Musunuru said. For nearly two years, Musunuru and Ahrens-Nicklas have quietly been running “dress rehearsals” attempting to make bespoke therapies for those urea cycle patients, as well as for a similar set of debilitating conditions called organic acidemias. They use Greek letters as code names for the patients. They worked through several different approaches before finding a possible solution for Patient “Alpha” but it came too late — they’re awaiting a liver transplant. Patient “Beta” was another complicated frustration as they realized that the quirks of the genome and technical limitations with CRISPR make some mutations harder to fix than others. “It was taking us a long time, but we were learning,” Musunuru said. Theoretical treatments for patients “Gamma,” “Delta” and “Epsilon” started to move faster. When Patient “Eta’ came along, they created a base editing solution within a month, “quicker than we expected,” Musunuru said. “That started us thinking, for the first time, maybe there’s still time left to actually try to manufacture this drug. And so that’s what we’re trying to do,” he said. Musunuru’s vision, like Urnov’s, goes beyond a single patient. Making and administering the first therapy could provide the initial proof points to show regulators a true platform to rapidly make custom cures for any infant with a urea cycle disorder. The bar for the first therapy might be high, but Musunuru eventually wants to build confidence with regulators so that new therapies can rapidly be made and tested in cells in a petri dish, without lengthy and expensive animal studies. “The clock is ticking with these kids, and you want to treat them as soon as possible. You can’t spend a couple of years doing the full set of studies. It’s gonna be way too late,” Musunuru said. That urgency has opened a door for Musunuru to move his custom CRISPR treatments toward the clinic faster than Urnov’s team. “The speed with which this moved is extraordinary,” Urnov told me, adding that the two groups are in close contact. “We’re frankly just throwing each other the supplies we’re missing. This is not a race.” There are indications the team is on the brink of acting: Musunuru said the first child could be treated this year. He isn’t sure if it will be Patient Eta, or someone who hasn’t been born yet. But he is pushing hard to make it happen, and very soon. “That would be important, and just show that it’s possible, and hopefully stimulate people’s imagination and get people thinking, OK, wow, you can do this. It’s not just science fiction.” Unsurprisingly, Urnov’s and Musunuru’s approaches have sparked a debate. “What Kiran and Fyodor are doing is incredible and I am rooting for them,” said John Finn, chief scientific officer of Basecamp Research, a biotech startup using AI to design new gene editing tools. “But what they are doing is taking a rare disease and turning it into rarer diseases. That is the opposite of where pharma is going.” Finn’s company, and several others, are trying to make new tools that allow scientists to completely replace a damaged gene, which would allow a single therapy to treat all forms of a genetic disease, regardless of what mutations cause it. But Musunuru doesn’t want to wait for this one-size-fits-all approach to materialize. “People have been a little too ambitious just trying to knock off the entirety of disease in one go,” he said. “This is about the art of the possible.” Musunuru also thinks that once scientists start addressing the root causes of conditions in the genome, the very concept of disease becomes “an artificial distinction.” “We should not really be thinking about it as a traditional medication; we should be thinking about it a bit more like an intervention or molecular surgery,” he said. “At the end of the day, it’s a genetic issue that occurs somewhere in the genome. And you can use the same tool and go after any of them.” Although CRISPR has often been compared to genetic surgery, Musunuru means it in a far more literal sense: viewing gene editing, at least for very rare diseases, as a procedure and not a drug, designed and administered by doctors trained as “interventional geneticists.” “No one asked the FDA, ‘Am I allowed to do the surgery?’ Once the basic principle has been demonstrated, then it’s up to the surgeon to decide with the patient,” Musunuru said. “I would like to see more decentralization,” he added, with “centers of excellence” credentialed for gene editing therapies, “rather than having to go back to the FDA over and over and over again.” “Maybe companies don’t have as much of a role as we have been envisioning them having in the long run,” Musunuru said. Some biotech leaders are even thinking about the concept. “There’s a point where if you’ve got 10 people with the disease, and you have to study all 10, the market disappears as a result. That’s a problem. It’s an area ripe for some innovation,” Intellia Therapeutics CEO John Leonard told me. “What’s the difference between a scalpel and CRISPR? It’s just what you’re cutting, as far as I’m concerned.” This is hardly the first time someone has tried to develop a genetic cure for a condition neglected by the biopharma industry. But Musunuru and Urnov are setting their sights far beyond a single child or a single disease, hoping to blaze a trail that can be walked again and again. “Changing a topping on a pizza does not cause a “back to beginnings” process of optimizing how the dough is made, how the pizza is shaped, or how it is baked,” Urnov wrote in an unusually glib analogy, part of a 6,000-word 2024 editorial in which he outlined the burdens caused by a regulatory system that makes drug developers redo the entire suite of preclinical and manufacturing tests every time they change the guide RNA. But getting the toppings wrong on a pizza probably won’t kill you. The wrong guide RNA might. So companies conduct safety studies in monkeys to see if it causes any unintended, potentially harmful edits. The question becomes, how safe is safe enough? Urnov thinks it’s possible to be cautious to a fault and says that most severe safety issues from gene therapies weren’t predicted from experiments in animals. “The number one thing our field needs is more clinical data. The only way to find out how to safely gene edit people is to edit more people,” he told me. The decision is almost certain to come down to the FDA, whose leadership has in recent weeks been gutted by the Trump administration. That includes Peter Marks, the head of the FDA’s Center for Biologics Evaluation and Research, who oversaw gene editing regulation at the FDA until his ouster in March. Marks had called CRISPR “a poster child for a platform technology” at a meeting last year, a comment that caused Urnov to almost fall out of his chair in excitement. It’s unknown if the new FDA shares the enthusiasm. But the agency’s new leader, Marty Makary, has proposed “a new pathway for drugs based on a plausible mechanism,” which could bode well for CRISPR drug developers trying to fix the root cause of disease in the genes. With fewer staff to implement such changes at the agency, the timing and details of that new pathway are unclear. But Urnov hopes that the November meeting with regulators has laid a foundation for change. “The biggest risk is if we don’t do this,” Urnov told the room that month. Looking back on the meeting, he described it as a step forward. “The feeling of gratification was profound, but it expired in about an hour and was replaced by an even greater sense of motivation to actually get this done,” he said. “The burden of action is on us.” In one of our interviews, Urnov surprised me by saying that he listens to Amon Amarth, a Viking-themed metal band, when he lifts weights. “And they have a wonderful line in one of their songs that goes, ‘We will find a way or make one.’ And so that is very much our mindset to this day,” Urnov told me. “The clinicians know what to do. The CRISPR people know what to do. And the regulators welcome their advances in doing it,” Urnov said. “If we don’t materially move the needle, if we don’t look back at 2024 or 2025 as the inflection point, shame on us.”

到目前为止，基于AI的设计仅限于现有的数据集，达不到生物学领域要求的智能水平。基因组学在训练AI模型方面具有最大的潜力，例如，AlphaFold2模型。然而，在勘探活动挖掘生物多样性数据库时，公司往往未能确保各方之间的经济利益分享，导致了对利益分配不公平的日益不满，因此，使得生物技术公司缩减此类研究项目。近日，英国南安普顿综合医院的Ben Johnson在Nature biotechnology上发表AI biotechs launch bioprospecting expeditions with Indigenous groups, agree to share benefits，说明了AI生物技术公司开始与原住民和国家政府合作，开展生物勘探活动，以搜寻全球热点地区，收集新的DNA和蛋白质序列，用于开发具有商业价值的分子。伦敦的Basecamp Research公司通过与全球25个国家合作，建立了世界上最大的用于训练AI的生物多样性数据库。该公司专注于从极端环境收集物种的DNA和蛋白质序列，以开发具有商业潜力的新蛋白质。Basecamp在开展勘探活动前同意分享蛋白质收入，以避免传统商业生物勘探中的不公正和生物剽窃等行为。Basecamp的生物勘探活动不仅注重数据收集，还强调与当地社区的合作，在不违反伦理的前提下收集生物数据，并已与目标地达成了利润分享协议。2024年8月，Basecamp不仅与喀麦隆政府签署了一项利润分享合作协议，承诺将对当地生物DNA进行测序，并分享使用这些序列制造AI工具的利润，而且为喀麦隆的科学家提供研究资金，以促进当地生物多样性研究的发展。Basecamp的BaseGraph数据库已经比最大的公共数据库大10倍，包含DNA、蛋白质、环境和地理信息，从而帮助预测蛋白质的功能。Basecamp利用深度学习工具不仅预测大型复杂蛋白质的三维结构和相互作用，还能发现新蛋白质并优化它们以满足客户在化妆品、治疗、食品和生物修复方面的需求。此外，Basecamp还与NVIDIA合作开发用于药物发现的生成性AI平台，并与巴塞罗那分子生物学研究所合作开发了ZymCTRL工具，用于生成性蛋白质设计。Basecamp还注重合作交流，它与15家公司和高校合作，共同开发新的生物技术产品。Basecamp带动了其他生物技术公司开展生物勘探探险，同意分享利益。随着生物勘探的加速，全球正在形成新的协议，以确保其可持续性和道德性，避免生物剽窃行为。2024年5月，世界知识产权组织签署了一项新条约，以打击生物剽窃行为，并确保原住民和其他当地社区从这些发现中获得利润分成。综上，以Basecamp Research为代表的AI生物技术公司与土著群体和国家政府合作、在不违反伦理的前提下收集生物数据，开展生物勘探活动，并促进利益共享。（郭苗苗 摘译）