For most of human history, when we needed a useful biological tool, we first looked for one in nature.<\/p>\n\n\n\n
Find a bacterium.<\/p>\n\n\n\n
Identify one of its enzymes.<\/p>\n\n\n\n
Then try to modify that enzyme so it performs the job we need a little better.<\/p>\n\n\n\n
Vilnius-based Biomatter is trying to reverse that logic.<\/strong><\/p>\n\n\n\n Instead of asking, \u201cWhich protein already exists in nature that we might be able to adapt?\u201d<\/em> the company wants to start with a different question:<\/p>\n\n\n\n \u201cWhat kind of protein do we need?\u201d<\/strong><\/p>\n\n\n\n And then design it.<\/p>\n\n\n\n Biomatter, legally registered in Lithuania as Biomatter Designs UAB<\/strong>, is building an AI- and physics-based platform for designing new enzymes. Its goal is to create proteins with specific desired properties, even if those proteins have never existed in nature.<\/p>\n\n\n\n This is one of the areas where generative artificial intelligence can mean something far more fundamental than generating text or images.<\/p>\n\n\n\n Here, it is generating biology.<\/p>\n\n\n\n To understand Biomatter, it helps to first understand enzymes.<\/p>\n\n\n\n Enzymes are proteins that accelerate chemical reactions.<\/p>\n\n\n\n Inside the human body, they help digest food, copy genetic information and carry out countless other processes.<\/p>\n\n\n\n But enzymes are just as important in industry.<\/p>\n\n\n\n They are used in food production.<\/p>\n\n\n\n Diagnostics.<\/p>\n\n\n\n Pharmaceuticals.<\/p>\n\n\n\n Agriculture.<\/p>\n\n\n\n Biofuels.<\/p>\n\n\n\n Chemical synthesis.<\/p>\n\n\n\n Gene therapy.<\/p>\n\n\n\n Sometimes an enzyme can replace a process that would otherwise require high temperatures, high pressure or aggressive chemicals.<\/p>\n\n\n\n That means the right enzyme is not simply a better biological component.<\/p>\n\n\n\n It can change the entire way a product is manufactured<\/strong>.<\/p>\n\n\n\n The problem is that nature did not evolve enzymes for our industrial processes.<\/p>\n\n\n\n Traditional protein engineering usually starts with a natural protein that already performs something close to the desired function.<\/p>\n\n\n\n Scientists then modify its amino-acid sequence.<\/p>\n\n\n\n Produce the new variant.<\/p>\n\n\n\n Test it in the laboratory.<\/p>\n\n\n\n See whether it performs better.<\/p>\n\n\n\n Then modify it again.<\/p>\n\n\n\n The process continues until the result is good enough.<\/p>\n\n\n\n But there is a fundamental limitation.<\/p>\n\n\n\n Natural enzymes evolved to help organisms survive, not to operate optimally inside biotechnology factories.<\/p>\n\n\n\n An enzyme may work perfectly inside a bacterial cell but become unstable at a higher temperature.<\/p>\n\n\n\n It may be too slow.<\/p>\n\n\n\n It may not tolerate the solvents used in manufacturing.<\/p>\n\n\n\n It may react with the wrong molecule.<\/p>\n\n\n\n Or the type of enzyme needed for a new product may simply not exist in nature at all.<\/p>\n\n\n\n From Biomatter’s perspective, continuously modifying a protein discovered in nature creates an artificial constraint.<\/p>\n\n\n\n Its Intelligent Architecture\u2122<\/strong> platform aims to build enzymes from the bottom up \u2014 starting from the desired function and physical requirements and working toward the molecular structure needed to achieve them.<\/p>\n\n\n\n Biomatter’s story is closely connected to Vilnius University.<\/p>\n\n\n\n Laurynas Karpus, Vykintas Jauni\u0161kis and Irmantas Rokaitis<\/strong> met at the university’s Institute of Biotechnology, where they began experimenting with the idea of applying generative AI methods to enzyme design.<\/p>\n\n\n\n Donatas Repe\u010dka<\/strong> and Vilnius University professor Rolandas Me\u0161kys<\/strong> also became part of the team. Together, they founded Biomatter in 2018.<\/p>\n\n\n\n At the time, the idea was fairly radical.<\/p>\n\n\n\n Generative neural networks were already showing that they could learn complex structures from data and produce new examples.<\/p>\n\n\n\n The Biomatter team asked a different question:<\/p>\n\n\n\n if an algorithm can learn what a realistic image looks like, can it learn what a functional protein looks like?<\/strong><\/p>\n\n\n\n One of the team’s most important early projects became ProteinGAN<\/strong>.<\/p>\n\n\n\n The model was trained on large numbers of natural protein sequences and generated entirely new sequences that had never previously appeared in nature.<\/p>\n\n\n\n But an amino-acid sequence that looks convincing on a computer is not necessarily a working enzyme.<\/p>\n\n\n\n The real test had to happen in the laboratory.<\/p>\n\n\n\n The team collaborated with researchers at Chalmers University of Technology in Sweden, who physically produced and experimentally tested some of the AI-generated proteins.<\/p>\n\n\n\n They worked.<\/p>\n\n\n\n In 2021, the results were published in Nature Machine Intelligence<\/strong> in the paper \u201cExpanding functional protein sequence spaces using generative adversarial networks.\u201d<\/em><\/p>\n\n\n\n That was an important moment.<\/p>\n\n\n\n The algorithm had not merely predicted what a protein might look like.<\/p>\n\n\n\n It had proposed new protein sequences that were later shown to function in real biological experiments<\/strong>.<\/p>\n\n\n\n There is, however, a huge distance between a successful scientific paper and a commercially useful enzyme.<\/p>\n\n\n\n An industrial customer does not merely need a protein that functions.<\/p>\n\n\n\n They may need an enzyme that:<\/p>\n\n\n\n works at a specific temperature;<\/p>\n\n\n\n survives a particular chemical environment;<\/p>\n\n\n\n reacts with exactly one target molecule;<\/p>\n\n\n\n remains stable during storage;<\/p>\n\n\n\n operates quickly enough;<\/p>\n\n\n\n and can be produced economically at scale.<\/p>\n\n\n\n Over time, Biomatter expanded the research behind ProteinGAN into its broader Intelligent Architecture\u2122<\/strong> platform.<\/p>\n\n\n\n It combines generative AI, bioinformatics, physics-based modelling and real laboratory experiments. The computer proposes proteins, the laboratory produces and tests them, and the experimental results feed back into the design process.<\/p>\n\n\n\n This is an important part of the Biomatter story.<\/p>\n\n\n\n The company is not simply a software business sending customers computer-generated sequences.<\/p>\n\n\n\n It operates its own laboratory infrastructure where the designed proteins are physically produced and validated.<\/p>\n\n\n\n AI designs. Biology decides whether it was right.<\/strong><\/p>\n\n\n\n Biomatter says this approach can dramatically reduce the number of experimental iterations required.<\/p>\n\n\n\n Traditional enzyme engineering can involve many rounds of design, production and testing.<\/p>\n\n\n\n The company’s ambition is to create a final enzyme in weeks rather than years<\/strong>.<\/p>\n\n\n\n That could be one of the most important economic consequences of the technology.<\/p>\n\n\n\n If every new biological component takes years to develop, many potentially useful products are never created.<\/p>\n\n\n\n Not because they are physically impossible.<\/p>\n\n\n\n But because finding the right enzyme would take too long and cost too much.<\/p>\n\n\n\n If protein-design time and cost fall significantly, an entirely new class of projects could become economically viable.<\/p>\n\n\n\n Biomatter’s technology is already being used with major international companies.<\/p>\n\n\n\n The company has named BASF, Thermo Fisher Scientific, Kirin and Neogen<\/strong> among its partners and customers. Biomatter says enzymes designed through its platform are already being applied to problems in health, food, agriculture and more sustainable manufacturing.<\/p>\n\n\n\n One example is its collaboration with Japan’s Kirin<\/strong> to develop enzymes for producing human milk oligosaccharides, or HMOs.<\/p>\n\n\n\n These are complex sugars naturally found in human breast milk and considered important in infant nutrition, but they are difficult to manufacture at industrial scale.<\/p>\n\n\n\n Another collaboration with ArcticZymes Technologies<\/strong> has involved enzymes used in molecular diagnostics, gene therapy, vaccines and other biopharmaceutical manufacturing.<\/p>\n\n\n\n In 2025, Biomatter also announced a collaboration with US-based Neogen<\/strong>, using Intelligent Architecture\u2122 to develop new enzyme-based products for food safety applications.<\/p>\n\n\n\n This is a very different stage from a university experiment.<\/p>\n\n\n\n The question is no longer simply:<\/p>\n\n\n\n \u201cCan AI design a functional protein?\u201d<\/em><\/p>\n\n\n\n The question is becoming:<\/p>\n\n\n\n \u201cCan commercially useful proteins be designed repeatedly for specific problems faced by global companies?\u201d<\/strong><\/p>\n\n\n\n In 2024, Biomatter raised a \u20ac6.5 million funding round<\/strong> led by Finnish investor Inventure<\/strong> and Germany’s UVC Partners<\/strong>.<\/p>\n\n\n\n Existing investors Practica Capital<\/strong> and Metaplanet<\/strong> also participated, alongside angel investors and industry specialists.<\/p>\n\n\n\n It was not the company’s first funding.<\/p>\n\n\n\n At an earlier stage, Biomatter had already raised a \u20ac500,000 round led by Practica Capital.<\/p>\n\n\n\n The newer capital is being used to expand the capabilities of Intelligent Architecture\u2122 and develop entirely new enzymes.<\/p>\n\n\n\n That matters in biotechnology because scaling does not simply mean hiring more software engineers and renting more servers.<\/p>\n\n\n\n It means laboratories.<\/p>\n\n\n\n Equipment.<\/p>\n\n\n\n Reagents.<\/p>\n\n\n\n Protein expression and purification systems.<\/p>\n\n\n\n Analytical chemistry.<\/p>\n\n\n\n Biochemists.<\/p>\n\n\n\n And large numbers of real-world experiments.<\/p>\n\n\n\n Biomatter remains a relatively small organisation.<\/p>\n\n\n\n Lithuanian company data in 2026 indicates a team of roughly 30\u201331 employees<\/strong>.<\/p>\n\n\n\n But the composition of that team is unusual.<\/p>\n\n\n\n Software engineers, machine-learning specialists, bioinformaticians, molecular biologists, biochemists and analytical chemists work inside the same company.<\/p>\n\n\n\n Recent Biomatter job openings also show continued expansion of its Vilnius laboratory operations, including roles for analytical chemists, biotechnology specialists, scientists and laboratory management.<\/p>\n\n\n\n This is a useful illustration of what an AI biotechnology company really looks like.<\/p>\n\n\n\n A model alone is not enough.<\/p>\n\n\n\n Someone has to take the idea generated by a computer and turn it into an actual molecule<\/strong>.<\/p>\n\n\n\n Biomatter is also beginning to generate meaningful commercial revenue.<\/p>\n\n\n\n Financial data filed in Lithuania shows that the company generated approximately \u20ac962,000 in sales in 2024<\/strong>, increasing to around \u20ac1.78 million in 2025<\/strong>.<\/p>\n\n\n\n That represents annual growth of roughly 85%.<\/p>\n\n\n\n The company is not yet profitable.<\/p>\n\n\n\n Its net loss in 2025 was approximately \u20ac991,000<\/strong>, compared with around \u20ac1.32 million in 2024.<\/p>\n\n\n\n For a deep-tech company, that is not an unusual stage. Biomatter is simultaneously developing proprietary technology, conducting laboratory research and expanding commercial operations.<\/p>\n\n\n\n The more interesting signal is that the scientific platform is already turning into paid customer projects.<\/p>\n\n\n\n In January 2026, Biomatter took another interesting step by joining the AstraZeneca BioVentureHub<\/strong> in Gothenburg, Sweden.<\/p>\n\n\n\n It became the first company from Lithuania \u2014 and from the Baltic region more broadly \u2014 to join the industrial life-sciences R&D ecosystem.<\/p>\n\n\n\n That creates an interesting full circle in the Biomatter story.<\/p>\n\n\n\n Some of the company’s early generative protein-design work was experimentally validated through collaboration with Chalmers University of Technology in Gothenburg.<\/p>\n\n\n\n Several years later, the Vilnius-built company returned to the same city as a growing biotechnology business working alongside one of the world’s largest pharmaceutical companies.<\/p>\n\n\n\n In 2025, Laurynas Karpus, Vykintas Jauni\u0161kis and Irmantas Rokaitis<\/strong> were selected among ten global finalists for the European Patent Office’s Young Inventors Prize<\/strong>.<\/p>\n\n\n\n They were chosen from more than 450 candidates for their AI- and physics-based enzyme-design technology.<\/p>\n\n\n\n Earlier, Karpus, Jauni\u0161kis and Rokaitis had also appeared in Forbes 30 Under 30 Europe<\/strong> in the Science & Healthcare category.<\/p>\n\n\n\n But at this point, perhaps more important than the awards is the sequence of stages the technology has already passed through.<\/p>\n\n\n\n It began in a university laboratory.<\/p>\n\n\n\n It became a scientific paper.<\/p>\n\n\n\n It became patentable technology.<\/p>\n\n\n\n It became a company.<\/p>\n\n\n\n And it is now becoming commercial enzymes used in projects with international customers.<\/p>\n\n\n\n The biggest idea behind Biomatter is not really about a single enzyme.<\/p>\n\n\n\n It is about how biological technology may be created in the future<\/strong>.<\/p>\n\n\n\n A software engineer does not begin a new application by searching nature for a piece of code that roughly does what is required.<\/p>\n\n\n\n An engineer does not search for a naturally occurring machine part and then spend years trying to reshape it.<\/p>\n\n\n\n They start with requirements.<\/p>\n\n\n\n What must the system do?<\/p>\n\n\n\n Under what conditions must it work?<\/p>\n\n\n\n What properties does it need?<\/p>\n\n\n\n Then they design it.<\/p>\n\n\n\n Biomatter is betting that protein engineering is moving in the same direction.<\/p>\n\n\n\n Not:<\/p>\n\n\n\n find an enzyme that almost fits.<\/em><\/p>\n\n\n\n But:<\/p>\n\n\n\n design the enzyme that is needed.<\/strong><\/p>\n\n\n\n If that becomes a reliable and repeatable process, the consequences could be extremely broad.<\/p>\n\n\n\n New pharmaceutical manufacturing processes.<\/p>\n\n\n\n More efficient diagnostics.<\/p>\n\n\n\n New food ingredients.<\/p>\n\n\n\n Cleaner chemical production.<\/p>\n\n\n\n Biological processes replacing energy-intensive industrial reactions.<\/p>\n\n\n\n And products that cannot be manufactured today simply because nature never had a reason to evolve the enzyme required to make them.<\/p>\n\n\n\n In 2018, Biomatter was a small team of Vilnius scientists trying to answer a fairly fundamental question:<\/p>\n\n\n\n Can artificial intelligence create a new functional protein?<\/strong><\/p>\n\n\n\n Today, the question has changed.<\/p>\n\n\n\n The company has international customers, close to \u20ac1.8 million in annual revenue, a team of more than 30 people, its own laboratory infrastructure and \u20ac6.5 million in fresh capital to expand the technology.<\/p>\n\n\n\n Its enzymes are already being used across health, food and industrial applications, while the company continues trying to shorten the path from a desired biological function to a real molecule.<\/p>\n\n\n\n That makes Biomatter more than simply another Lithuanian AI startup.<\/p>\n\n\n\n It is a company trying to do for biology what computer-aided design has done for many other areas of engineering:<\/p>\n\n\n\n learn to design something digitally before manufacturing it in the physical world.<\/strong><\/p>\n\n\n\n Only in this case, the final product is not a chip, a building or a machine.<\/p>\n\n\n\n It is a molecular machine of life that may never have existed before.<\/strong><\/p>\n","protected":false},"excerpt":{"rendered":" For most of human history, when we needed a useful biological tool, we first looked for one in nature. Find a bacterium. Identify one of its enzymes. Then try to modify that enzyme so it performs the job we need a little better. Vilnius-based Biomatter is trying to reverse that logic. Instead of asking, \u201cWhich protein […]<\/p>\n","protected":false},"author":1,"featured_media":183,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3],"tags":[],"class_list":["post-180","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-startups"],"_links":{"self":[{"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/posts\/180","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/comments?post=180"}],"version-history":[{"count":1,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/posts\/180\/revisions"}],"predecessor-version":[{"id":182,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/posts\/180\/revisions\/182"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/media\/183"}],"wp:attachment":[{"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/media?parent=180"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/categories?post=180"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/inpulse.lt\/en\/wp-json\/wp\/v2\/tags?post=180"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Enzymes are tiny factories<\/h3>\n\n\n\n
Nature did not optimise for our factories<\/h3>\n\n\n\n
The idea started at Vilnius University<\/h3>\n\n\n\n
ProteinGAN<\/h3>\n\n\n\n
From scientific experiment to product<\/h3>\n\n\n\n
Weeks instead of years<\/h3>\n\n\n\n
The enzymes are no longer just laboratory experiments<\/h3>\n\n\n\n
A \u20ac6.5 million bet on the idea<\/h3>\n\n\n\n
Around 30 people between code and the laboratory bench<\/h3>\n\n\n\n
Revenue nearly doubled<\/h3>\n\n\n\n
From Vilnius into the AstraZeneca ecosystem<\/h3>\n\n\n\n
Recognition for the underlying technology<\/h3>\n\n\n\n
Proteins could become programmable technology<\/h3>\n\n\n\n
Designing biology in Vilnius that never existed in nature<\/h3>\n\n\n\n