Synthetic Biology

The fourth ability is the most difficult: composing. Frances Arnold, Nobel Prize in Chemistry and trustee of the Bankinter Innovation Foundation, summed it up as follows:

“We can read, we can write, we can edit.

What we’re still dealing with is composing.”

A cell is more like a symphony than a line of code. We can read the score, change notes, and write new snippets. But we still do not know how to compose a complete biological work.

This idea avoids two mistakes:

• to think that synthetic biology is science fiction, when it already has real applications;
• believing that a cell is programmed like a computer. The cell is alive, evolving, and responding to an environment that we still only partially understand.

That evolutionary history complicates the design. One gene can influence several traits. A protein can serve different functions. A small modification can generate unexpected effects depending on the cell, the environment or the state of the organism.

That is why synthetic biology advances with ambition, but also with caution. Applications such as cancer drugs, modified animal organs for human transplants, synthetic aviation fuels or more drought-resistant crops are already being explored. But the challenge remains the same: to move from intervening specific parts to predicting how the entire system will behave.

As Woolfson summarizes, sequencing the human genome gave us a catalog of pieces, but not the complete manual for building an organism.

Synthetic biology aspires to do with living systems what engineering does with any other technology: design components that work together and produce a predictable result. To achieve this, it is necessary to understand how proteins, genetic circuits, metabolic pathways and cells work, and learn how to combine them in a stable and safe way.

Proteins and enzymes: the tools of life

Proteins perform a good part of the functions of life. Among them are enzymes, biological catalysts capable of accelerating chemical reactions with great precision.

Designing them is still difficult. Luis Serrano, director and group leader of the Centre for Genomic Regulation (CRG), recalled that just a decade ago thousands of variants had to be produced to find a useful one. AI and computational tools help shorten that path, but many engineered proteins still don’t work as expected.

That’s why synthetic biology combines two approaches. Rational design attempts to anticipate what sequence will produce a function. Directed evolution, the method for which Frances Arnold received the Nobel Prize in Chemistry, creates variants, tests them and selects the best ones. Miguel Alcalde, Research Professor at the CSIC and founder of EvoEnzyme, is also working along these lines.

The key is to combine calculation and evolution: using AI to search better, and biology to validate what works.

Genetic circuits: instructions within a cell

Biology can also be used to build circuits. In electronics, a circuit receives a signal, executes a logic, and produces an output. In biology, a genetic circuit does something similar, but inside a cell.

Jim Collins, professor of engineering and medical science at MIT and a pioneer of synthetic biology, explained that the field was born when some scientists began to wonder if they could build biological networks, rather than just studying natural ones.

His group developed one of the first genetic switches in E. coli: a kind of basic memory inside a cell. The idea had relevant implications: if a cell can receive information and respond, it can be designed to act under certain conditions.

The limit is in the context. A circuit that works in one model bacterium can fail in another organism. The cell is not an electronic board: it is alive, evolving, and responding in ways that are difficult to anticipate.

Metabolic pathways: turning cells into chemical factories

Another level of design is metabolic pathways: chains of reactions with which a cell transforms molecules into molecules. Synthetic biology can redesign them to make useful compounds.

Tobias Erb, director of Biochemistry and Synthetic Metabolism at the Max Planck Institute for Terrestrial Microbiology, is working on a key frontier: converting CO₂ into other materials. Technical carbon capture is still limited, while biology captures hundreds of billions of tons of CO₂ per year. The problem is that natural photosynthesis converts less than 1% of light energy into chemical energy.

Erb’s group designs artificial biological pathways to improve that efficiency. Some are already showing promising results: up to 2 times more efficiency and 10 times faster than natural photosynthesis in experimental settings.

The goal is for cells, enzymes or biological systems to convert CO₂ into raw materials to manufacture fuels, materials or chemicals in a more sustainable way. AI and automation accelerate that path. The next challenge will be to demonstrate that these processes also work on an industrial scale.

Cells as biological chassis

In synthetic biology, a cell can act as a “chassis”: a foundation on which to build new functions. It can be a bacteria, a yeast, a plant cell, or a modified human cell.

But a cell is never a passive structure. It has metabolism, evolutionary history, energy limits and defense mechanisms. That is why writing DNA is not enough. It must be introduced into the cell, made to work, kept stable and avoid unwanted effects.

Marc Güell, director of the Synbio Lab at the Universitat Pompeu Fabra, explained it with an example: bacteria from the skin microbiome designed as possible therapeutic platforms. They could produce a molecule locally to treat a particular problem. Even in an accessible environment such as the skin, the challenge remains enormous: that the bacteria are stable and behave predictably.

Living systems that learn and adapt

The most advanced frontier is not in creating fixed parts, but living systems capable of adapting.

Alfonso Jaramillo, director of the De Novo Synthetic Biology Lab at the I2SysBio Institute, works on cells that communicate through electrical signals and bacteria that modify their genetic activity when they receive stimuli.

One of its most striking examples was that of bacteria trained to play tic-tac-toe. The anecdote is surprising, but the important thing is the substance: a cell can change its behavior without modifying its DNA.

From gene to genome

A gene is a concrete instruction. A genome is the complete set of instructions for an organism. Moving from one to the other changes the scale of the challenge.

Patrick Cai, Professor of Synthetic Genomics at the University of Manchester, is working on that frontier: redesigning, fabricating, and transplanting synthetic chromosomes into recipient organisms. Writing ever-increasing fragments of life allows us to see to what extent we understand how it works.

The Sc2.0 project, focused on building a synthetic yeast genome, shows this leap well. Yeast is more complex than a bacterium, but much simpler than a human being. Redesigning its genome allows it to be reorganized, simplified, introduce new capabilities and see what changes are compatible with life.

Cai also underlined a key condition: these projects need international collaboration, standards and shared design plans. Genome writing is beginning to look less like an artisanal feat and more like an organized discipline.

What's missing to write life reliably

Writing a genome is still much more difficult than writing a computer program. There are three challenges:

1 – Design and manufacture long, accurate and affordable DNA;

2 – to ensure that these instructions produce the expected result;

3- and maintain that behavior in cells that change, evolve and respond to their environment.

Patrick Cai sums it up with a very illustrative comparison. In computer science, there is a compiler that allows you to check if a program will work before running it. In biology there is still no equivalent tool. We can write a DNA sequence, but we cannot reliably predict how the organism will behave.

The reason is that DNA is only one part of the equation. The organization of the genome, metabolism, the state of the cell and the environment in which it lives also play a role. That is why writing DNA is much easier today than programming life.

Designing organisms: a border, not a full reality

It is advisable to use expressions such as “write genomes”, “program cells” or “design organisms” carefully. They are directions of advancement, but they are not yet fully mastered capabilities.

Today we can edit genes, synthesize DNA, build circuits and modify cells. There are also organisms with dozens of genetic changes for very specific applications, such as some xenotransplantation programs.

What we cannot yet do in a general way is to design a complex organism from scratch, anticipate its behavior in different environments and guarantee its stability in the long term.

This challenge marks the scientific agenda of the field. Every synthetic chromosome, every stable circuit, every modified cell, and every failed attempt help to better understand how life is organized.

Synthetic biology is accelerated by the convergence of several capabilities: reading DNA on a large scale, synthesizing it, analyzing biological data, automating experiments, and using AI to propose new designs.

The core logic is the DBTL cycle: design, build, test, and learn. Before, each lap required a lot of manual work and long laboratory cycles. Today, models can suggest designs, biofoundries can test variants, and the data obtained feeds new rounds of improvement.

AI to design biology

AI helps to explore huge biological spaces: proteins, metabolic pathways, genetic circuits or candidate molecules. Anthony Costa, global chief strategy officer for Life Sciences at NVIDIA, placed this trend within a broader transition to more programmable biology.

But the use of generative AI has limits. In biology, data are more heterogeneous and difficult to interpret. A genetic sequence can tell how a molecule is written, but not always what it does or how it will behave inside a cell.

That is why an important part of the current career consists of expanding and organizing biological databases. Basecamp Research, for example, is sequencing organisms from multiple environments to capture a larger fraction of the planet’s genetic diversity. Phil Lorenz, its chief technology officer, described that diversity as follows:

The data that matters: function, not just sequence

The bottleneck is no longer in reading DNA, but in knowing what function a given sequence has. What is important is the functional data: what a cell produces, what activity a protein has, how a metabolic pathway changes or why a design fails.

The experts insisted on a basic condition for AI to be useful: quality, homogeneous, well-annotated and experimentally validated data. The frontier is in moving from accumulating biological data to generating useful data for design.

Biofoundries: Experiment Factories

Biofoundries are automated laboratories that combine robotics, software, molecular biology, data analysis, and scientific equipment. Its value lies in conducting experiments in a more systematic, traceable and reproducible way.

Automation isn’t just for a lot of testing. It serves, above all, to better choose what to try. The AI proposes designs, the biofoundry tests them, the results correct the model, and the cycle starts again.

The forum showed different models. Tong Si, from the Shenzhen Institutes of Advanced Technology, presented a large-scale biofoundry supported by public investment. Min Hao Wong, from A*STAR in Singapore, defended a more agile model, connected to companies. Marko Storch, Chief Operating Officer of the Biofoundry at Imperial College London, underlined the importance of common standards for different biofoundries to interoperate.

The conclusion is clear: a biofoundry is not just a laboratory with robots. You need specialized talent, software, protocols, connection to the industry, and a strategy for what problems you want to solve.

Case in point: antibiotics discovered with AI

The work of César de la Fuente, presidential associate professor at the University of Pennsylvania and director of the Computational Biology and Artificial Intelligence Group (Machine Biology Group), shows this acceleration well. His team uses AI models to search for antimicrobial peptides, optimize them, and test them in the lab. Some candidate molecules have advanced to preclinical stages.

The case shows the potential and limit of AI. It can greatly expand the search space, but discovering a molecule does not equate to having a drug. Then come toxicity analysis, stability, manufacturing, clinical trials, regulatory approval, and the business model.

Synthetic biology is beginning to transform sectors where living processes are already part of the value chain: health, food, agriculture, chemistry, energy or materials. Its pace of adoption will be different in each industry. In health, it can assume high costs if it solves a clear medical need. In food and agriculture, price, scale and social acceptance weigh more. In chemistry, energy and the environment, it must compete with highly optimized industrial processes. The key question is always the same: where does it bring a real advantage over what already exists?

Blood substitutes: a very specific medical need

Synthetic biology can also solve very practical problems. One of the examples presented at the Future Trends Forum was the development of blood substitutes for emergency transfusions.

Many serious hemorrhages occur outside the hospital, where compatible and cold blood is difficult. Allan Doctor, chief scientific officer and co-founder of KaloCyte, is working on an artificial substitute for red blood cells that can be transported in powder form, reconstituted at the site of use, and safely transported oxygen.

Its objective is to have a useful product in accidents, conflicts, catastrophes or remote areas. The case illustrates how synthetic biology can also innovate healthcare infrastructure, making it easier to store, transport and make treatments available where they are needed most.

From drug discovery to next-generation medicines

Synthetic biology can also improve drug discovery. It can provide more realistic cell models, new targets, engineered proteins, RNA therapies, and more flexible production systems.

The challenge is clear: a novel target is only valuable if it can be turned into a safe, effective and manufacturable treatment. In health, biological design must be thought of from the outset along with toxicology, regulation, manufacturing, and clinical practice.

The search for new antibiotics illustrates this well. César de la Fuente’s work shows how AI can identify peptides with antimicrobial activity and take some candidates to preclinical phases. In cancer, the same logic appears in more personalized therapies: better reading the biology of the tumor, designing more precise responses, and accelerating the move toward treatments tailored to the patient. In both cases, the decisive leap is in turning these advances into real treatments for patients.

Food and agriculture: producing better on a planet under pressure

Food and agriculture can benefit greatly from synthetic biology, but they are difficult sectors to transform. They compete in huge markets, very sensitive to price, taste, safety and trust.

The need, however, is clear. The challenge of sustainable food is to produce more and better, with less pressure on soil, water, energy and ecosystems. At the same time, the world’s population is growing and the demand for protein is increasing. Rodrigo Ledesma-Amaro, director of the Bezos Center for Sustainable Protein and director of the Center for Microbial Foods, summed it up like this: feeding more people without intensifying the pressure on the planet.

Synthetic biology can help in two ways: producing food and ingredients in a different way, and developing tools for more efficient and resilient agriculture.

Some companies in the sector are exploring a path less focused on imitating meat and more oriented towards creating foods with differentiated properties: better nutritional profile, new textures, possible benefits for the microbiome or designs adapted to specific needs.

Synthetic biology as a food security tool

Synthetic biology can also bolster food security. Singapore is a good example. The lack of agricultural land and dependence on imports have led the country to bet on new forms of food production. He was the first to approve cultured meat and has turned this need into a technological strategy. As Min Hao Wong explained, the pandemic highlighted the risk of over-reliance on global supply chains.

Edited crops and crop protection

In agriculture, synthetic biology combines gene editing, beneficial microorganisms, biostimulants, pheromones, sensors, AI and new agronomic practices. The goal is to produce more with less pressure: less fertilizer, fewer broad-spectrum pesticides, and fewer losses from drought, salinity, pests, or disease.

New genomic techniques can accelerate crop improvement. Pierre Larrieu, head of Bayer Crop Science for Iberia and North Africa, pointed out that European farmers have fewer and fewer chemical tools available and need new options to maintain productivity and resilience.

Crop protection is also moving towards more specific solutions. An example is pheromones to control pests in fruit trees: instead of applying broad-spectrum insecticides, the target insect is attracted and the impact on other species is reduced.

Barbara Nave, senior project leader for White Biotechnology at BASF, placed this transition in a practical framework: reducing inputs, developing biological products and improving resistance to drought or salinity. The challenge is to move solutions from the laboratory to the field and make them work in real conditions.

Synthetic biology usually presents itself through its discoveries: a cell that produces a new molecule, an improved enzyme, a gene therapy, or a fermented food. But the real filter comes later: manufacturing it repeatably, safely, stably and at a competitive cost.

Scaling biology means ensuring that a living system continues to behave the same when everything around it changes: volume, temperature, nutrients, process times or regulatory requirements.

A strain can operate in a test tube and behave differently in a fermenter of thousands of liters. A protein can be expressed, but it can be difficult to purify. A metabolic pathway can produce well at first and lose performance over time.

That is why biomanufacturing is the point where synthetic biology ceases to be design and becomes industrial engineering.

Manufacturing with quality: the case of DNA

Manufacturing biology requires volume, quality, purity, speed and traceability. In cell and gene therapies, this is especially critical: the DNA needed to make or modify therapies can become a bottleneck.

Julen Oyarzabal, Chief Scientific Officer (CSO) and partner at Columbus Venture Partners, explained the problem with DNA for advanced therapies. Traditional methods can involve lengthy processes, bacterial impurities, and lead times of several months for GMP-grade material.

Its approach seeks to produce cleaner DNA in much shorter timeframes, with applications in cell and gene therapies, RNA, cancer vaccines or rapid responses to pandemics.

The case shows a key idea: an intermediate component can condition an entire industry. If DNA takes months, personalized medicine slows down. If a protein is not purified well, the cost skyrockets. If a strain does not maintain its yield, the plant loses viability.

That is why the manufacturing infrastructure – GMP, fermentation, purification, analytics, quality control and documentation – is strategic in synthetic biology.

Contracts, customers and industrial risk

Scaling biology involves taking technical risks: that the strain loses yield, that purification is too expensive, that the process is not stable, or that quality is not maintained batch after batch.

Odd Erik Hansen cited the case of Glycom to illustrate a possible strategy: combining commercial agreements, project finance, and industrial capacity to move from pilot scales to much larger volumes. He also showed a common logic in biomanufacturing: accept negative margins at the beginning to enter the market, gain volume and then approach the necessary cost.

The lesson is clear: scaling is not just about producing more. It consists of reducing technical uncertainty, validating the process and demonstrating that it can be repeated with quality, safety and controlled costs.

What to do inside and what to outsource

Scaling doesn’t mean building everything from scratch. In synthetic biology there are CDMOs(companies that develop and manufacture products for third parties), CROs(organizations that provide research services by contract), biofoundries, shared pilot plants and specialized suppliers that allow progress to be made with less of their own infrastructure.

But outsourcing everything has risks. Toni Glieder insisted that a company can outsource equipment, technologies or infrastructure, but not the talent and knowledge that allow it to continue innovating.

This knowledge is what allows us to adjust processes, diagnose failures, interpret data, redesign strains, comply with regulations and learn from each batch.

Shared infrastructures reduce barriers to entry, but they only add value if they have specialized talent, common standards and a real connection to the needs of the industry.

Data to scale

AI can help design better proteins, strains, or processes, but scaling up needs quality experimental data. It is not enough to have data: it must be comparable, well measured and useful for training models.

Data on fermentation, productivity, metabolites, culture conditions, batch failures, purification and quality control are needed. Toni Glieder pointed out a frequent problem: many AI projects find out late that they don’t have the necessary data. Odd Erik Hansen agreed on the importance of combining laboratory and computational platform.

Digital biology only works if it maintains a close link with physical biology. The models help reduce the search space, but the fermenter, pilot plant, and final product are still the ultimate test.

Synthetic biology is advancing faster than some of its business models. New capabilities emerge in the laboratory, but the market applies another filter: what problem they solve, who needs it, how much it costs to scale them and what advantage they offer compared to what already exists.

Many projects must translate a scientific possibility into a viable business proposition: sell a product, provide a service, license technology, manufacture for third parties, or partner with a large company.

The investment reflects that maturation. According to SynBioBeta, venture capital funding in synthetic biology reached $12.2 billion in 2024, up from $10.7 billion in 2023, following the post-peak cooling of 2021. Capital continues to arrive, but in a more selective way.

Arturo Urrios, partner at Ysios Capital, summed up the change: in Europe it is increasingly difficult to finance a platform in the abstract. Investors are asking for a clearer route to product, medical indication or market.

Tools, Platforms Products

In synthetic biology, several business models coexist. Some sell tools: design software, DNA synthesis, automation, databases, biofoundries, or laboratory services. Others develop platforms capable of generating product families. A third group is committed to end products: therapies, ingredients, enzymes, materials, cultures, diagnostics or environmental solutions.

Each model has a different logic. Tools must demonstrate recurring use. Platforms can generate a lot of value, but they need time and credibility. The products have a clearer history for the market, although they take on more technical, regulatory and commercial risk.

Javier García Cogorro, managing partner of Columbus Venture Partners, pointed to a frequent warning: the disconnection with the market. At Columbus, the evaluation starts there: real demand, differential value compared to existing alternatives, and the team’s ability to turn science into a marketable solution.

Funding technologies that are slow to scale

Venture capital has been key to the growth of synthetic biology, but its timing does not always match those of the sector. Developing a therapy, a manufacturing platform, a fermentation process, or a food can take years, sustained investment, and costly scale-ups.

Arturo Urrios explained an important difference: a corporate investor can accept longer horizons if the technology fits with his industrial strategy. A Venture Capital fund usually needs returns in shorter terms.

In synthetic biology, each phase needs a different type of capital: early science, technical validation, pilot plant, manufacturing, market arrival, and industrial expansion all have different risks and metrics.

The case of KaloCyte illustrates the difficulty of opening a new product category. When a technology arrives before its own market, there is no ready-made industrial chain, no fully defined demand, and no established payment system. Allan Doctor highlighted three key uncertainties: manufacturing cost, future price and regulatory timing. That combination makes it difficult to attract private investment, even though the medical need is evident.

Large companies and public capital

Large companies are key players in synthetic biology. They can act as customers, development partners, investors, startup buyers, or go-to-market channels. In pharmaceuticals, food, chemicals, agriculture or materials, many startups need corporate alliances to validate technology, manufacture at scale or reach global customers.

In the case of the pharmaceutical industry, external innovation is especially relevant, but it must be connected to real needs, reasonable times and possibilities for clinical development. In ecosystems like Cambridge, that interaction occurs early on through accelerators, incubators, business schools, corporate mentors, and venture capital.

Public capital also plays an important role. It can finance research, support infrastructure, co-invest or reduce risks in phases where private capital does not reach. A clear tension emerged at the forum: public funding can help build markets, but it must be well connected to business logic.

The case of Bpifrance shows a model in which public capital helps to mobilise private investment without replacing it, as Javier García Cogorro pointed out.

Spain as a destination for investment in synthetic biology

Spain appeared in the debate as an ecosystem with the capacity to attract synthetic biology projects thanks to the combination of scientific talent, infrastructure and competitive costs.

The case of Viralgen illustrates this opportunity well. The company chose Spain because of the availability of specialized talent, proximity to experts in viral technologies and the ability to quickly build a facility aimed at international customers, according to Javier García Cogorro.

Added to this is another advantage: Spain has top-level researchers and can develop high-quality science at lower costs than the United States, as Adrian Woolfson pointed out.

Synthetic biology is not just biology. Genetics, chemistry, engineering, data science, AI, automation, regulation, manufacturing, intellectual property, business and scientific communication can coexist in the same project.

That is changing the professional map of the sector. The biological researcher is still central, but more and more hybrid profiles are appearing: bioprocess engineers, fermentation experts, computational scientists, biological data specialists, quality professionals, regulatory profiles, plant operators and scientific entrepreneurs.

Hybrid profiles for a hybrid discipline

The convergence of biology, AI, and automation requires teams capable of speaking multiple languages. Models and data must be connected from the outset to biological, clinical, or industrial applications, as Phil Lorenz pointed out.

This mix of disciplines also changes the way we work. César de la Fuente’s laboratory brings together computer scientists, chemists, engineers, biologists, physicists and biochemists. Part of his role, he explained, is to act as a “translator” between different ways of thinking.

Training is beginning to reflect this change. Synthetic biology is advancing as a fast-paced and multidisciplinary field, while many universities continue to be organized into more stable disciplines, as Wilfried Vanhonacker pointed out. But there are already programmes that combine biology, engineering, computing and entrepreneurship, such as the master’s degree in Integrative Synthetic Biology at UIMP-CSIC in Spain or the specialised programmes at Imperial College London and the University of Edinburgh.

Operate, manufacture and scale

Industrial talent has a weight of its own. Fermentation, purification, quality control, documentation, GMP production and transfer to the plant require diverse teams: researchers, engineers, vocational training profiles and specialized technicians capable of operating equipment, applying quality protocols and working in organic production plants.

Toni Glieder explained it from industrial experience: a biological technology needs professional equipment capable of maintaining and improving the production system for years. In biomanufacturing, knowledge is in how the organism behaves, how the process responds, and how quality is preserved batch after batch.

Communicating a difficult technology

Science communication is also part of the necessary talent. Synthetic biology needs to explain what problem it solves, what evidence supports it, what limits it has, and what value it brings to existing alternatives.

Internally, this communication helps biologists, chemists, engineers, data scientists, clinicians and industrial profiles to work with a common language. Externally, it allows dialogue with regulators, investors, customers and citizens without confusing real capabilities with promises that are still open.

Synthetic biology is no longer just a scientific field. It is increasingly part of the industrial and security strategies of major economies. Its relevance does not depend only on discovering new treatments or materials, but also on mastering capabilities such as biological design, DNA synthesis, biomanufacturing, biological data or biosafety.

Whoever controls these capacities will have an easier time developing medicines, producing food, manufacturing advanced materials or responding to future health crises with greater autonomy. Synthetic biology is beginning to be seen as a strategic infrastructure, comparable to other emerging technologies such as artificial intelligence or semiconductors.

New forms of governance

The regulatory framework is beginning to move from an almost exclusively risk-focused approach to models that also incorporate potential benefit, early learning, and dialogue with developers.

This change is especially visible in healthcare, where Advanced Therapy Drugs (ATMs) can benefit from more agile regulatory procedures.

The evolution is part of a broader trend toward anticipatory forms of governance, according to Douglas Robinson, Public Policy Advisor and Strategic Intelligence Leader at OECD and Senior Research Scientist in Emerging Technologies and Innovation Policy at CNRS. These include tools such as regulatory sandboxes , scenario evaluation or early dialogue between regulators, researchers, companies and other stakeholders.

In food and agriculture, approval times and evaluation criteria directly influence the speed with which new technologies reach the market.

Social trust

Public acceptance will be one of the factors that condition the development of synthetic biology, especially in food and agriculture.

Pierre Larrieu noted that many technologies are already part of the food system, although they are often poorly understood or incompletely communicated. Douglas Robinson expanded on this idea by recalling that trust also depends on aspects such as governance, transparency, access to technology, intellectual property, competition or the ability of consumers to choose.

The experience of other technologies shows that social acceptance rarely depends on scientific evidence alone. The perception that innovations respond to real needs, are developed with clear safety criteria and offer understandable benefits for society also play a role.