Hubbry Logo
Biological computingBiological computingMain
Open search
Biological computing
Community hub
Biological computing
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Biological computing
Biological computing
Biological computing
View on Wikipedia
from Wikipedia

Biological computers use biologically derived molecules — such as DNA and/or proteins — to perform digital or real computations.

The development of biocomputers has been made possible by the expanding new science of nanobiotechnology. The term nanobiotechnology can be defined in multiple ways; in a more general sense, nanobiotechnology can be defined as any type of technology that uses both nano-scale materials (i.e. materials having characteristic dimensions of 1-100 nanometers) and biologically based materials.[1] A more restrictive definition views nanobiotechnology more specifically as the design and engineering of proteins that can then be assembled into larger, functional structures[2][3] The implementation of nanobiotechnology, as defined in this narrower sense, provides scientists with the ability to engineer biomolecular systems specifically so that they interact in a fashion that can ultimately result in the computational functionality of a computer.

Scientific background

[edit]

Biocomputers use biologically derived materials to perform computational functions. A biocomputer consists of a pathway or series of metabolic pathways involving biological materials that are engineered to behave in a certain manner based upon the conditions (input) of the system. The resulting pathway of reactions that takes place constitutes an output, which is based on the engineering design of the biocomputer and can be interpreted as a form of computational analysis. Three distinguishable types of biocomputers include biochemical computers, biomechanical computers, and bioelectronic computers.[4]

Biochemical computers

[edit]

Biochemical computers use the immense variety of feedback loops that are characteristic of biological chemical reactions in order to achieve computational functionality.[5] Feedback loops in biological systems take many forms, and many different factors can provide both positive and negative feedback to a particular biochemical process, causing either an increase in chemical output or a decrease in chemical output, respectively. Such factors may include the quantity of catalytic enzymes present, the amount of reactants present, the amount of products present, and the presence of molecules that bind to and thus alter the chemical reactivity of any of the aforementioned factors. Given the nature of these biochemical systems to be regulated through many different mechanisms, one can engineer a chemical pathway comprising a set of molecular components that react to produce one particular product under one set of specific chemical conditions and another particular product under another set of conditions. The presence of the particular product that results from the pathway can serve as a signal, which can be interpreted—along with other chemical signals—as a computational output based upon the starting chemical conditions of the system (the input).

Biomechanical computers

[edit]

Biomechanical computers are similar to biochemical computers in that they both perform a specific operation that can be interpreted as a functional computation based upon specific initial conditions which serve as input. They differ, however, in what exactly serves as the output signal. In biochemical computers, the presence or concentration of certain chemicals serves as the output signal. In biomechanical computers, however, the mechanical shape of a specific molecule or set of molecules under a set of initial conditions serves as the output. Biomechanical computers rely on the nature of specific molecules to adopt certain physical configurations under certain chemical conditions. The mechanical, three-dimensional structure of the product of the biomechanical computer is detected and interpreted appropriately as a calculated output.

Bioelectronic computers

[edit]

Biocomputers can also be constructed in order to perform electronic computing. Again, like both biomechanical and biochemical computers, computations are performed by interpreting a specific output that is based upon an initial set of conditions that serve as input. In bioelectronic computers, the measured output is the nature of the electrical conductivity that is observed in the bioelectronic computer. This output comprises specifically designed biomolecules that conduct electricity in highly specific manners based upon the initial conditions that serve as the input of the bioelectronic system.

Network-based biocomputers

[edit]

In networks-based biocomputation,[6] self-propelled biological agents, such as molecular motor proteins or bacteria, explore a microscopic network that encodes a mathematical problem of interest. The paths of the agents through the network and/or their final positions represent potential solutions to the problem. For instance, in the system described by Nicolau et al.,[6] mobile molecular motor filaments are detected at the "exits" of a network encoding the NP-complete problem SUBSET SUM. All exits visited by filaments represent correct solutions to the algorithm. Exits not visited are non-solutions. The motility proteins are either actin and myosin or kinesin and microtubules. The myosin and kinesin, respectively, are attached to the bottom of the network channels. When adenosine triphosphate (ATP) is added, the actin filaments or microtubules are propelled through the channels, thus exploring the network. The energy conversion from chemical energy (ATP) to mechanical energy (motility) is highly efficient when compared with e.g. electronic computing, so the computer, in addition to being massively parallel, also uses orders of magnitude less energy per computational step.

Engineering biocomputers

[edit]
A ribosome is a biological machine that uses protein dynamics on nanoscales to translate RNA into proteins.

The behavior of biologically derived computational systems such as these relies on the particular molecules that make up the system, which are primarily proteins but may also include DNA molecules. Nanobiotechnology provides the means to synthesize the multiple chemical components necessary to create such a system.[citation needed] The chemical nature of a protein is dictated by its sequence of amino acids—the chemical building blocks of proteins. This sequence is in turn dictated by a specific sequence of DNA nucleotides—the building blocks of DNA molecules. Proteins are manufactured in biological systems through the translation of nucleotide sequences by biological molecules called ribosomes, which assemble individual amino acids into polypeptides that form functional proteins based on the nucleotide sequence that the ribosome interprets. What this ultimately means is that one can engineer the chemical components necessary to create a biological system capable of performing computations by engineering DNA nucleotide sequences to encode for the necessary protein components. Also, the synthetically designed DNA molecules themselves may function in a particular biocomputer system. Thus, implementing nanobiotechnology to design and produce synthetically designed proteins—as well as the design and synthesis of artificial DNA molecules—can allow the construction of functional biocomputers (e.g. Computational Genes).

Biocomputers can also be designed with cells as their basic components. Chemically induced dimerization systems can be used to make logic gates from individual cells. These logic gates are activated by chemical agents that induce interactions between previously non-interacting proteins and trigger some observable change in the cell.[7]

Network-based biocomputers are engineered by nanofabrication of the hardware from wafers where the channels are etched by electron-beam lithography or nano-imprint lithography. The channels are designed to have a high aspect ratio of cross section so the protein filaments will be guided. Also, split and pass junctions are engineered so filaments will propagate in the network and explore the allowed paths. Surface silanization ensures that the motility proteins can be affixed to the surface and remain functional. The molecules that perform the logic operations are derived from biological tissue.

Economics

[edit]

All biological organisms have the ability to self-replicate and self-assemble into functional components. The economical benefit of biocomputers lies in this potential of all biologically derived systems to self-replicate and self-assemble given appropriate conditions.[4]: 349  For instance, all of the necessary proteins for a certain biochemical pathway, which could be modified to serve as a biocomputer, could be synthesized many times over inside a biological cell from a single DNA molecule. This DNA molecule could then be replicated many times over. This characteristic of biological molecules could make their production highly efficient and relatively inexpensive. Whereas electronic computers require manual production, biocomputers could be produced in large quantities from cultures without any additional machinery needed to assemble them.

Notable advancements in biocomputer technology

[edit]

Currently, biocomputers exist with various functional capabilities that include operations of Boolean logic and mathematical calculations.[5] Tom Knight of the MIT Artificial Intelligence Laboratory first suggested a biochemical computing scheme in which protein concentrations are used as binary signals that ultimately serve to perform logical operations.[4]: 349  At or above a certain concentration of a particular biochemical product in a biocomputer chemical pathway indicates a signal that is either a 1 or a 0. A concentration below this level indicates the other, remaining signal. Using this method as computational analysis, biochemical computers can perform logical operations in which the appropriate binary output will occur only under specific logical constraints on the initial conditions. In other words, the appropriate binary output serves as a logically derived conclusion from a set of initial conditions that serve as premises from which the logical conclusion can be made. In addition to these types of logical operations, biocomputers have also been shown to demonstrate other functional capabilities, such as mathematical computations. One such example was provided by W.L. Ditto, who in 1999 created a biocomputer composed of leech neurons at Georgia Tech which was capable of performing simple addition.[4]: 351  These are just a few of the notable uses that biocomputers have already been engineered to perform, and the capabilities of biocomputers are becoming increasingly sophisticated. Because of the availability and potential economic efficiency associated with producing biomolecules and biocomputers—as noted above—the advancement of the technology of biocomputers is a popular, rapidly growing subject of research that is likely to see much progress in the future.

In March 2013. a team of bioengineers from Stanford University, led by Drew Endy, announced that they had created the biological equivalent of a transistor, which they dubbed a "transcriptor". The invention was the final of the three components necessary to build a fully functional computer: data storage, information transmission, and a basic system of logic.[8]

Parallel biological computing with networks, where bio-agent movement corresponds to arithmetical addition was demonstrated in 2016 on a SUBSET SUM instance with 8 candidate solutions.[6]

In July 2017, separate experiments with E. Coli published on Nature showed the potential of using living cells for computing tasks and storing information. A team formed with collaborators of the Biodesign Institute at Arizona State University and Harvard's Wyss Institute for Biologically Inspired Engineering developed a biological computer inside E. Coli that responded to a dozen inputs. The team called the computer "ribocomputer", as it was composed of ribonucleic acid. Harvard researchers proved that it is possible to store information in bacteria after successfully archiving images and movies in the DNA of living E. coli cells.[9]

In 2021, a team led by biophysicist Sangram Bagh realized a study with E. coli to solve 2 x 2 maze problems to probe the principle for distributed computing among cells.[10][11]

In 2024, FinalSpark, a Swiss biocomputing startup, launched an online platform enabling global researchers to conduct experiments remotely on biological neurons in vitro.[12]

In March 2025, Cortical Labs unveiled CL1, the world's first commercially available biological computer integrating lab-grown human neurons with silicon hardware.[13] Building on earlier work with DishBrain, CL1 uses hundreds of thousands of neurons sustained by an internal life-support system for up to six months, enabling real-time learning and adaptive computation within a closed-loop environment. The system operates via the Biological Intelligence Operating System (biOS), allowing direct code deployment to living neurons. CL1 is designed for applications in drug discovery, disease modeling, and neuromorphic research, offering an ethically preferable alternative to animal testing and consuming significantly less energy than traditional artificial intelligence systems.[14][15][16][17]

Future potential of biocomputers

[edit]

Many examples of simple biocomputers have been designed, but the capabilities of these biocomputers are very limited in comparison to commercially available inorganic computers.

The potential to solve complex mathematical problems using far less energy than standard electronic supercomputers, as well as to perform more reliable calculations simultaneously rather than sequentially, motivates the further development of "scalable" biological computers, and several funding agencies are supporting these efforts.[18][19]

See also

[edit]

References

[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Biological computing

View on Grokipedia
from Grokipedia
Biological computing, also known as biocomputing, is an interdisciplinary field that employs biological materials and living systems—such as DNA, proteins, and cells—to execute computational tasks, offering a bio-inspired alternative to conventional silicon-based electronics. This approach integrates principles from synthetic biology, molecular engineering, and computer science to design circuits that process inputs through biochemical reactions, enabling functions like logic operations, pattern recognition, and data storage within dynamic, self-sustaining environments. Unlike traditional computing, which relies on binary digital logic and rigid hardware, biological computing leverages the inherent parallelism, stochasticity, and adaptability of living systems to handle complex, noisy, or evolving problems more efficiently in certain domains. The foundational concept of biological computing emerged in 1994 when Leonard Adleman demonstrated that DNA molecules could solve the directed Hamiltonian path problem, a combinatorial optimization task, by encoding graph vertices and edges as nucleotide sequences and using biochemical reactions to explore solution paths. This proof-of-concept highlighted the massive parallelism of molecular interactions, where billions of DNA strands can simultaneously test possibilities, though early implementations faced scalability limits, such as requiring vast quantities of material for larger problems. Subsequent milestones in the early 2000s included the engineering of synthetic genetic circuits in bacteria, such as the bistable genetic toggle switch, which maintains one of two stable states based on input signals, and the repressilator, a ring oscillator that generates rhythmic gene expression patterns. Key approaches in biological computing span molecular, cellular, and multicellular scales. At the molecular level, uses strand hybridization and enzymatic manipulations for operations like logic and arithmetic, while protein-based systems exploit enzymatic cascades for . Cellular computing engineers or with genetic circuits—such as those designed via the software—to implement digital logic gates, including all 16 two-input functions, often visualized through fluorescent outputs. Multicellular consortia enable , where populations of engineered cells communicate via diffusible molecules () to solve problems like or environmental sensing, as demonstrated in spatial arrangements that form gradients for multi-input logic. These systems move beyond Turing-complete models by incorporating biological features like , noise tolerance, and self-repair, leading to the concept of "cellular supremacy" in tasks such as adaptive or real-time diagnostics in unpredictable settings. Applications of biological computing are particularly promising in and , where its and autonomy shine. For instance, engineered can act as living sensors for pollutants or pathogens, processing signals to trigger therapeutic responses or diagnostic readouts, as seen in colony-based systems for pollution testing and . In , cellular computers could autonomously adjust outputs based on disease biomarkers, while in synthetic biology-driven bio-production, they optimize metabolic pathways for sustainable manufacturing. Despite challenges like metabolic burden, circuit crosstalk, and the need for precise control , ongoing advances in tools like and high-throughput design are accelerating the field toward practical, scalable implementations. As of 2025, emerging developments in intelligence and bio-hybrid systems with artificial biological neurons are enhancing integration with traditional computing paradigms.

Overview

Definition and principles

Biological computing refers to the utilization of biological molecules, cells, or organisms to execute computational tasks, encompassing the storage, processing, and output of through biochemical reactions or cellular dynamics. This approach leverages the inherent information-processing capabilities of , such as DNA strands for data encoding or enzymatic reactions for logical operations, to perform calculations that mimic or surpass aspects of electronic computing. Unlike traditional silicon-based systems, biological computing operates in aqueous environments at ambient temperatures, drawing on molecular interactions to handle complex problems with high parallelism and specificity. Key principles of biological computing include massive parallel processing, as exemplified by DNA strands simultaneously exploring combinatorial solution spaces in experiments solving problems. Self-assembly enables spontaneous organization of molecular components into functional structures, while built-in error correction mechanisms, akin to in , maintain computational fidelity by discarding erroneous intermediates. Additionally, energy efficiency arises from ATP-driven reactions that power reversible binding and catalytic cycles, achieving thermodynamic advantages over electronic counterparts in terms of operations per unit energy. At its core, biological computing conceptualizes information in molecular terms: bits are represented by discrete states, such as the presence or absence of specific DNA base pairs (A-T or C-G pairings) that encode binary values. Logic gates emerge from enzyme cascades, where sequential biochemical activations implement Boolean operations like AND or OR, propagating signals through substrate conversions. Theoretical foundations, such as chemical reaction network (CRN) theory, demonstrate Turing completeness, allowing universal computation via networks of interacting species. A fundamental CRN dynamic is captured by the rate equation for species concentration: d[X]dt=k[A][B]−l[X]\frac{d[X]}{dt} = k [A][B] - l [X] where [X][X] is the concentration of product species, kk is the forward reaction rate constant for reactants AA and BB, and ll is the degradation rate constant, illustrating how concentrations evolve to compute outputs over time.

Historical development

The roots of biological computing trace back to the mid-20th century, influenced by and early theoretical models of . In the 1940s and 1950s, explored self-reproducing automata within frameworks, laying foundational ideas for machines capable of universal construction and replication inspired by biological systems. These concepts highlighted the potential for computational processes to mimic biological growth and adaptation, bridging digital logic with organic-like . Building on this, in the 1960s, Howard Pattee proposed biochemical mechanisms for processing and error correction in , suggesting that molecular interactions could function as rudimentary logic gates for reliable biological computation. A pivotal experimental breakthrough occurred in the 1990s with the advent of DNA-based computing. In 1994, Leonard Adleman demonstrated the first molecular solution to a combinatorial problem by using DNA strands to encode and solve the directed Hamiltonian path problem in a small graph, leveraging the massive parallelism of biochemical reactions to generate and select valid paths. This experiment marked the shift from theoretical speculation to practical biocomputation, illustrating how DNA's hybridization properties could perform search operations infeasible for conventional computers at the time. The 2000s saw significant advances in engineering biological components for computational purposes. In 2000, Michael Elowitz and Stanislas Leibler engineered the first synthetic gene circuit, known as the repressilator, in Escherichia coli, which produced sustained oscillatory behavior through cyclic repression of three genes, demonstrating tunable, clock-like dynamics in living cells. Complementing this, in 2006, Paul Rothemund introduced DNA origami, a technique for folding long single-stranded DNA with short staple strands to assemble precise two-dimensional nanostructures, enabling programmable scaffolds for future computational assemblies. These developments underscored the feasibility of designing genetic and molecular circuits with predictable outputs. In the , biological computing matured with innovations in programmable genetic tools and multi-cellular systems. By 2016, researchers utilized /dCas9 systems to construct robust digital logic circuits in eukaryotic cells, including multi-layer repression cascades that executed complex operations with up to seven inputs, expanding the toolkit for intracellular . In 2015, synthetic bacterial consortia were engineered to perform tasks, where populations of specialized strains collaborated via chemical signaling to solve problems like and decision-making, mimicking in microbial communities. By the late 2010s, these efforts coalesced into established fields like , which gained formal recognition around the early 2000s through initiatives integrating engineering principles with genetic redesign, and neuromorphic bioelectronics, which emerged in the mid-2010s with organic devices emulating neural for energy-efficient interfacing with biological tissues. The 2020s brought further breakthroughs in organoid intelligence and biocomputing using living neural tissues. In 2022, the DishBrain system demonstrated lab-grown organoids learning to play the through electrophysiological feedback, showcasing adaptive learning in biological neural networks. By 2025, Cortical Labs released the CL1, the world's first commercially available biological computer powered by human neurons integrated with silicon chips, enabling efficient processing for AI-like tasks in a biocompatible manner. These advancements highlight the shift toward hybrid bio-electronic systems capable of real-time learning and environmental interaction.

Fundamental principles

Biological mechanisms for computation

Biological computing leverages molecular and cellular processes inherent in living systems to perform information processing tasks. At the molecular level, transcription and translation serve as fundamental computational primitives, where DNA sequences are transcribed into messenger RNA (mRNA) and subsequently translated into proteins that execute specific functions based on input signals such as environmental cues or ligand binding. This process enables the encoding, storage, and execution of computational instructions analogous to software in traditional systems, with regulatory elements like promoters and enhancers acting as conditional logic operators. RNA molecules further contribute to computation through aptamers, which are short, single-stranded RNA sequences that fold into specific three-dimensional structures capable of binding target molecules with high affinity and specificity. These aptamers function as sensor-actuator modules, detecting inputs (e.g., small molecules or ions) and triggering downstream conformational changes or interactions that propagate signals, thereby implementing sensing and response logic in biological circuits. For instance, RNA aptamers can be engineered to modulate activity, creating allosteric switches that couple input detection to output generation in a modular fashion. At the cellular level, gene regulatory networks (GRNs) model logic operations through the combinatorial control of by transcription factors, where activators and repressors interact to produce , and NOT gates. In these networks, the presence or absence of transcription factors serves as binary inputs, determining whether target genes are expressed (output 1) or silenced (output 0), as demonstrated in models of the in . Quorum sensing in bacteria exemplifies parallel decision-making within GRNs, where cells collectively assess population density via diffusible autoinducers, enabling coordinated behaviors such as in Vibrio fischeri only when a threshold concentration is reached, thus implementing a population-level for group computation. Network dynamics in biological systems are governed by feedback loops and bistability in signaling pathways, which allow for robust input-output mapping and memory storage. For example, the mitogen-activated protein kinase (MAPK) cascade processes extracellular signals through sequential phosphorylation events, creating layered amplification and decision points that map diverse inputs to specific cellular responses like proliferation or apoptosis. Bistability arises from positive feedback, where an active component reinforces its own activation, enabling switch-like behavior that maintains states even after input removal, as seen in the competence development pathway in Bacillus subtilis. A key example of these dynamics is found in phosphorylation cascades, which operate as AND/OR logic gates by requiring multiple kinase activations (AND) or allowing alternative pathways (OR) for signal propagation. The kinetics of these enzymatic reactions are described by the Michaelis-Menten equation: v=Vmax⁡[S]Km+[S]v = \frac{V_{\max} [S]}{K_m + [S]} where vv is the , Vmax⁡V_{\max} is the maximum rate, [S][S] is the substrate concentration, and KmK_m is the Michaelis constant representing substrate affinity. This model captures how substrate saturation leads to switch-like transitions in cascade outputs, enhancing computational precision in pathways like the yeast pheromone response. Biological systems incorporate error handling through redundancy and mechanisms to ensure computational reliability. In , for instance, multiple DNA polymerases provide backup fidelity, while proofreading exonucleases remove mismatched with an error rate as low as 10−710^{-7} per , mitigating propagation of computational errors in genetic information storage. Such mechanisms parallel error correction in silicon-based but rely on stochastic molecular interactions for robustness.

Comparison to traditional computing

Biological computing architectures fundamentally differ from traditional electronic computing, which relies on the von Neumann model characterized by sequential , centralized control, and a clear separation between and units. In contrast, biological systems employ distributed, architectures where emerges from interconnected networks of molecules or cells, enabling simultaneous operations across vast scales without a central processor. Additionally, biological is inherently analog and , relying on continuous chemical gradients and probabilistic reactions rather than the discrete, deterministic binary logic of digital electronics. Performance metrics highlight stark trade-offs between the two paradigms. Biological operations, such as DNA strand hybridization, achieve exceptional energy efficiency, consuming approximately 5 × 10^{-20} J per reaction compared to around 10^{-15} J for a typical . However, biological speeds are significantly slower, with gate-like operations taking seconds or minutes due to diffusion-limited reaction kinetics, versus nanoseconds in electronic systems. Scalability in biological computing excels in volumetric parallelism, operating effectively at attomolar concentrations (10^{-18} M) to enable billions of simultaneous computations in microliter volumes, though this is constrained by molecular crowding and error propagation in larger networks. density further favors , with DNA storing up to 1 bit per nm³, orders of magnitude higher than silicon-based at approximately 10^{-6} bits per nm³. Key advantages of biological computing include inherent adaptability through dynamic molecular interactions and self-repair mechanisms, such as enzymatic error correction in DNA systems, which maintain functionality without external intervention. These systems also demonstrate robust tolerance to noise, leveraging stochastic processes akin to evolutionary algorithms to converge on solutions amid environmental variability. Despite these strengths, biological computing faces notable disadvantages, including a lack of precise control over reaction outcomes due to inherent variability in biological environments, leading to higher rates than the near-perfect reliability of electronic determinism. is particularly challenging, often requiring full redesign of molecular components or cellular pathways, in contrast to the rapid reconfiguration possible in software-defined electronic systems.

Types of biocomputers

Biochemical computers

Biochemical computers harness chemical reactions involving biomolecules, such as DNA strands and enzymes, to perform computational operations through processes like hybridization, catalysis, and reaction cascades. These systems operate in solution-phase environments, leveraging the specificity and parallelism of molecular interactions to execute logic gates, arithmetic functions, and problem-solving algorithms. Unlike electronic or mechanical biocomputers, biochemical variants rely solely on wet chemistry without physical movement or electrical interfaces. DNA computing represents a foundational approach in biochemical computation, where oligonucleotides encode information and hybridize to form solutions to complex problems. In a seminal experiment, Leonard Adleman demonstrated the feasibility of this paradigm by solving an instance of the directed Hamiltonian path problem—an NP-complete combinatorial challenge—using DNA molecules to represent graph vertices and edges in a test tube. The process involved generating all possible paths via ligation and polymerase chain reaction amplification, followed by selective extraction of valid solutions through gel electrophoresis and affinity purification, yielding the correct path within seconds for a seven-vertex graph. This parallel search capability arises from the massive molecular population, enabling exhaustive exploration of solution spaces that are intractable for classical computers. Subsequent advances utilized DNA strand displacement reactions to construct dynamic logic circuits, where input strands competitively displace output strands from complexes, propagating signals without enzymatic intervention. For instance, researchers implemented reversible strand displacement to build digital circuits, including a four-bit square-root calculator comprising 130 DNA strands that executed over 100 parallel reaction steps with high yield. These cascades enable Boolean operations like AND, OR, and XOR gates, as well as more complex functions such as neural network simulations, by cascading displacement events that amplify weak signals through seesaw mechanisms. Additionally, algorithmic self-assembly of DNA tiles has enabled patterned computation, where triple-crossover DNA molecules assemble into lattices that perform cumulative XOR operations tile-by-tile, computing binary patterns during crystallization. This 2000 demonstration by Seeman and colleagues marked a key step in using DNA nanotechnology for parallel, deterministic computation via error-correcting tile designs. Enzyme-based biochemical computers extend this paradigm by incorporating catalytic biomolecules for signal amplification and processing within reaction networks. DNAzymes—catalytically active DNA strands selected —form logic gates by cleaving RNA substrates in response to specific inputs, enabling operations like NOT, AND, and XOR without external enzymes. For example, modular deoxyribozyme designs have constructed half-adders and full adders, where substrate cleavage outputs fluorescent signals proportional to input combinations, achieving outputs with sensitivities down to nanomolar concentrations. , RNA-based catalysts, similarly support computational circuits; evolutionary selection has yielded ribozyme gates that perform YES, AND, and OR functions by modulating ligation or cleavage rates based on input binding. These systems amplify weak inputs through multiple turnover , allowing scalable circuits that process multiple signals in parallel, as seen in libraries of DNAzyme subunits forming autonomous diagnostic networks. Metabolic engineering principles have been applied to in vitro enzymatic reaction networks to realize arithmetic computations, where substrate concentrations serve as inputs and product yields as outputs. Immobilized enzyme cascades, such as those combining hexokinase, glucose-6-phosphate dehydrogenase, and phosphatase, perform addition by summing glucose and fructose inputs into a shared ATP pool, with outputs measured via absorbance changes. Subtraction and multiplication emerge from competitive inhibition and sequential kinetics; for instance, a network using invertase and glucose oxidase subtracts input ratios by depleting shared substrates, while multiplier circuits scale outputs quadratically through coupled dehydrogenases. These cell-free systems demonstrate arithmetic precision over input ranges of 0.1–10 mM, with reaction times under 30 minutes, highlighting the potential for continuous-flow biochemical processors. Brief historical DNA experiments, like Adleman's, laid groundwork for these networks by inspiring parallel molecular processing. Despite these advances, biochemical computers face limitations inherent to molecular fidelity, particularly in DNA hybridization where sequence mismatches lead to erroneous strand associations. Hybridization errors arise from partial complementarity or kinetic trapping, reducing specificity by up to 10% in long-strand assemblies and necessitating error-correcting mechanisms like domains. In , such infidelity propagates through cascades, limiting circuit depth to tens of gates before signal loss exceeds 50%, as observed in tile assemblies where branch migration failures disrupt patterns. These challenges underscore the need for thermodynamic optimization and toehold designs to enhance factors beyond 1000-fold.

Biomechanical computers

Biomechanical computers harness the mechanical properties of biological structures, such as translocation, contraction, and deformation, to perform computations at the molecular scale. These systems leverage force generation and movement driven by biological motors and polymers, enabling parallel processing and pattern-based logic without relying on electronic signals. Unlike biochemical approaches that depend on static chemical equilibria, biomechanical computing emphasizes dynamic physical interactions for information processing and storage. Molecular motors like kinesin enable track-based logic by propelling microtubules along predefined paths in nanofabricated networks, where the geometry encodes computational problems such as the subset sum. In these setups, kinesin motors, fueled by ATP hydrolysis, drive microtubules at speeds of 0.5–10 µm/s through junctions that split or direct traffic, allowing parallel exploration of solution spaces; for instance, solving a three-element subset sum problem required 179 microtubules over 180 minutes with high accuracy at pass junctions (97.9–99.7%). This ATP-fueled transport functions as a binary state machine, where motor attachment and detachment represent state transitions, facilitating directed cargo movement and logic operations in engineered environments. Actin-myosin systems utilize contractile networks to perform force computations and generate emergent patterns, mimicking cellular cytoskeletal dynamics for parallel problem-solving. motors slide filaments, producing contractile forces that reorganize networks into asters or clusters, with velocities of 2–5 µm/s proportional to activity and filament density. In computational models, swarms of filaments on myosin-coated surfaces execute pseudo-random walks to navigate mazes, solving graph-based problems through collective motion and in simulations with 900 filaments covering a 450×450 unit area. These networks can propagate contraction pulses or form stable structures under tuned active stress, enabling force-dependent analogous to mechanical logic gates. A seminal example is the kinesin-microtubule chips developed by Viola Vogel's group in the , which demonstrate directed assembly for nanoscale manufacturing. These microfluidic platforms feature kinesin-propelled as carriers in segmented canals (30 µm wide), sequentially loading and assembling biomolecules like NeutrAvidin and DNA strands at five stations to form complexes, powered by ATP in a proof-of-concept system. Such chips integrate biomechanical transport with hybrid electronic controls for precise cargo positioning. Force-output computation in biomechanical systems exploits piezoelectric-like responses in , where mechanical stress generates charge separation for sensing and feedback. Collagen and exhibit this effect due to their non-centrosymmetric structures, producing voltages under deformation that can signal environmental changes or computational states; for instance, in scaffolds, these responses enable self-sensing actuators with outputs proportional to applied strain. Hybrid designs briefly incorporate to read these mechanical signals, enhancing integration with traditional .

Bioelectronic computers

Bioelectronic computers represent hybrid systems that integrate biological components, such as cells or biomolecules, with electronic hardware to enable and at the interface of living and synthetic materials. These systems leverage the and adaptability of biological elements alongside the precision and speed of , facilitating applications in neural and sensory augmentation. Unlike purely biochemical approaches, bioelectronic designs emphasize electronic readouts and transduction mechanisms to bridge biotic and abiotic domains. Biohybrid interfaces form a core component of these systems, particularly through neuron-silicon junctions that emulate . In these setups, living neurons are cultured directly on substrates, allowing bidirectional communication where neuronal action potentials modulate transistor currents, and vice versa, mimicking and signal integration. Such interfaces have demonstrated stable synaptic-like responses, with neuronal firing rates influencing electronic gate voltages over extended periods. Similarly, ion channels embedded in lipid bilayers serve as biomolecular memristors, exhibiting history-dependent conductance changes analogous to synaptic weight updates. For instance, voltage-activated channels like alamethicin in droplet-interface bilayers display pinched current-voltage , enabling functions with switching energies below 1 fJ per event. Organic electronics enhance biocompatibility in these hybrids via conducting polymers such as poly(3,4-ethylenedioxythiophene) (PEDOT), which form the basis for flexible, bio-compatible transistors. PEDOT-based organic electrochemical transistors (OECTs) operate by ionic doping in aqueous environments, matching the conductivity and mechanical properties of biological tissues while supporting and proliferation. These transistors achieve transconductances up to 100 mS and operate at low voltages (<1 V), making them suitable for interfacing with excitable cells without eliciting immune responses. Neuromorphic bioelectronics further advance these systems by mimicking neural spikes using protein-based nanowires, harvested from electroactive bacteria like Geobacter sulfurreducens. These nanowires, composed of β-sheet pilin proteins, conduct electrons over micrometer scales with metallic-like properties (conductivity ~0.015 S/cm) and enable diffusive memristive behavior that replicates spike-timing-dependent plasticity. Devices incorporating such nanowires respond to biological voltages (10-100 mV) and exhibit adaptive conductance changes, facilitating energy-efficient neuromorphic processing with power consumption in the nW range. A prominent example of evolving bioelectronic integration is seen in retinal prosthetics from the 2010s, such as the Argus II system, which transitioned from basic generation to computational nodes by incorporating onboard for image preprocessing and stimulation patterning. By the 2020s, advancements in bacterial electroactive films have extended this paradigm, with Shewanella or Geobacter biofilms forming conductive matrices on electrodes that perform distributed computation through collective . These films generate sustained currents (up to 1 mA/cm²) and adapt to environmental signals, serving as living sensors or logic gates in biohybrid circuits. Signal transduction in these systems often relies on voltage-gated ion channels, whose dynamics are modeled by the Hodgkin-Huxley framework to predict current flows across membranes. The core equation for channel current is given by I=g(V−E)I = g (V - E) where II is the ionic current, gg is the time- and voltage-dependent conductance, VV is the , and EE is the reversal potential for the species. This formulation captures the nonlinear gating kinetics essential for bioelectronic signal amplification and has been adapted to simulate hybrid device responses in neural interfaces.

Cellular and organoid-based computers

Cellular and organoid-based computers leverage intact living cells or multicellular assemblies as computational substrates, harnessing biological processes for information processing without relying on synthetic molecular components alone. These systems exploit cellular machinery for parallel, adaptive computation, often integrating to engineer specific functions. Key examples include prokaryotic and eukaryotic cells programmed for logic operations and sensing, as well as advanced models that mimic neural networks for learning tasks. Bacterial computing has advanced through the engineering of to implement logic circuits using , enabling light-inducible control of for operations. For instance, optogenetic s in E. coli respond to blue light to activate split systems, facilitating AND in genetic circuits. Similarly, the OptoLacI system engineers the LacI repressor for light-controlled , constructing versatile platforms like the OptoE.coliLight system that toggle promoter activity with high precision and low crosstalk. These approaches allow to process environmental signals in real time, demonstrating scalability in microbial consortia for distributed . In mammalian systems, networks of induced pluripotent stem (iPS) cell-derived neurons form topologically controlled circuits for distributed sensing and integration. Protocols generate arrays of interconnected iPSC-derived neurons that self-assemble into functional networks, enabling the mapping of connectivity and response to stimuli across the . These networks support collective sensing of chemical or electrical inputs, with applications in modeling where individual cells contribute to emergent . For example, iPSC-derived neuronal spheroids on microelectrode arrays exhibit synchronized activity patterns, allowing distributed detection of environmental cues through network-wide propagation. Organoid intelligence (OI) represents a frontier in using brain organoids—3D cultures of human stem cell-derived neurons—as computational substrates for biocomputing. These organoids, typically comprising up to 100,000 cells with myelinated axons and spontaneous activity, emulate cortical structures and support learning via plasticity mechanisms. In OI, bio-inference occurs through synaptic plasticity in the living neural tissue, enabling computation via adaptive changes in synaptic connections and neural dynamics rather than traditional digital logic gates. Advancements in 2024 include microfluidic platforms for organoids, which provide precise spatiotemporal chemical gradients and nutrient delivery to enhance viability and induce adaptive responses, as well as feedback-driven systems that enable automated closed-loop of organoids by integrating and imaging for real-time adaptation. Researchers are connecting these organoids to the cloud, allowing them to learn and solve tasks in real-time through bio-digital feedback loops that integrate biological responses with digital processing. Integrated with AI, these systems enable tasks; for instance, organoids interfaced with multielectrode arrays (MEAs) decode electrophysiological signals using to classify stimuli and predict outputs in simulated environments. Recent 2025 developments, supported by NSF funding, focus on cellular computers using live organoids for energy-efficient AI, achieving computations with power consumption orders of magnitude lower than silicon-based systems. A $2 million NSF grant to supports engineering organoids on 3D-printed scaffolds to mimic cortical organization, where live neurons process encoded visual inputs via optical stimulation. serves as the primary interface: fluorescent imaging captures activity, while algorithms like detect motion patterns, demonstrating proof-of-concept for hybrid bio-AI architectures that leverage biological efficiency for tasks such as image recognition. A notable 2024 example from efforts illustrates potential for beyond-Turing computation through in living cellular systems. Researchers engineered genetic circuits in such as E. coli to process spatial information via reaction-diffusion mechanisms, forming self-organizing patterns that exceed classical limits by incorporating developmental dynamics. These morphogenetic computations, inspired by Turing's theory, enable cells to evaluate complex environmental states—such as gradients—through emergent spatial logic, paving the way for adaptive biocomputers that evolve structures in response to inputs.

Engineering and design

Synthetic biology approaches

Synthetic biology approaches to biological computing involve engineering genetic and molecular components to create programmable circuits that perform computational operations within living or cell-free systems. These methods draw on principles of to design logic gates and networks that mimic digital , enabling cells to process inputs and generate outputs based on biological signals. Key advancements focus on , , and to build reliable biocomputers capable of executing logic and more complex algorithms. Genetic circuit design forms the foundation of these approaches, utilizing promoters, repressors, and inverters to implement logic operations such as NOT, AND, and OR gates. Promoters act as input sensors that drive transcription, while repressors inhibit to create inverters, allowing the of basic logic elements that can be combined into larger networks for in cells. For instance, repressor-promoter pairs have been engineered to function as NOT gates, where the presence of an input represses output , enabling the realization of universal logic functions in synthetic biochemical circuits. Standardization of these components is facilitated by repositories like the iGEM Registry, which catalogs genetic parts such as promoters and repressors for reuse and interoperability across designs, promoting community-driven development of modular biocomputing systems. CRISPR tools have revolutionized circuit design by enabling precise and dynamic control over gene expression. The Cas9 nuclease, when adapted for interference (CRISPRi) or activation (CRISPRa), allows for tunable repression or activation of target genes, facilitating the rewiring of existing genetic networks without permanent DNA modification. Base editing variants of CRISPR further enhance precision by introducing single-nucleotide changes to fine-tune promoter strength or repressor binding sites, supporting the construction of robust logic gates in diverse cellular contexts. In vitro assembly methods, particularly cell-free expression systems, accelerate prototyping of these circuits by decoupling computation from cellular constraints. These systems use crude cell extracts to transcribe and translate DNA templates into functional proteins, allowing rapid testing of genetic designs in hours rather than days, which is essential for iterating on biocomputer architectures before in vivo implementation. A key technique for modular computation involves orthogonal transcription factors, such as those based on the T7 RNA polymerase system, which operate independently of host cellular machinery to create isolated channels for information processing. The T7 system, derived from bacteriophage, uses specific promoters recognized only by engineered T7 polymerases, enabling the construction of multi-layered circuits where outputs from one module drive inputs in another without crosstalk, thus supporting scalable Boolean computations. Scalability in approaches extends from single-gene modifications to genome-scale engineering, where multiplexed tools like arrays or recombineering enable the simultaneous alteration of hundreds of loci to embed complex computational networks across entire genomes. This progression allows biocomputers to handle larger datasets and more intricate logic, transitioning from simple prototypes to systems-level designs.

Integration with and AI

Integration with electronics and AI in biological computing involves developing interfaces that bridge the analog, ionic signaling of biological systems with the digital, electronic processing of conventional hardware, enabling hybrid systems with enhanced computational capabilities. Microelectrodes are widely used for neural recording in these setups, allowing high-density, minimally invasive capture of action potentials from neurons or organoids. For instance, thin-film microelectrode arrays facilitate chronic in vivo recordings over extended periods, such as 365 days in mouse models as of November 2025, by providing stable electrical contact with neural tissues. Complementing this, nanowire arrays enable single-cell addressing by penetrating cell membranes to record intracellular signals with high spatial resolution, as demonstrated in silicon nanowire field-effect transistor arrays interfaced with acute brain slices for mapping neural circuits. These methods support precise readouts from biological components, forming the foundation for bioelectronic hybrids. Data flow between biological and electronic domains is managed through specialized sensors that perform analog-to-digital conversion, with iontronic devices playing a key role in translating ionic currents into electrical signals. Iontronic sensors, inspired by biological synapses, convert chemical ionic inputs into multiplexed electrical outputs, enabling efficient in low-power environments. In hybrid architectures, these interfaces integrate into bio-AI chips designed for , where biological elements like organoids provide adaptive processing alongside electronic circuits for real-time decision-making. Recent neuromorphic integrations, such as those combining organoids with silicon-based neuromorphic hardware in 2025, leverage these chips to mimic -like efficiency, achieving energy savings over traditional von Neumann architectures while supporting tasks like . AI augmentation enhances the design and operation of these systems by optimizing biological circuits through machine learning techniques. algorithms, for example, train organoid responses by rewarding specific neural activity patterns, improving performance in goal-directed tasks such as simulations. These hybrid approaches not only amplify biological signals but also allow AI to iteratively refine circuit parameters, fostering scalable bio-AI systems for applications in adaptive . Biohybrid neural interfaces, incorporating soft bio-inspired , have advanced in brain-computer interfaces as of 2025.

Applications

Biomedical and therapeutic uses

Biological computing has significant potential in biomedical applications, particularly for systems that respond to physiological signals. In , synthetic biology enables the engineering of cells with logic gates that sense elevated glucose levels and trigger insulin release only when specific conditions are met, mimicking natural pancreatic function. For instance, a glucose-blue light system in engineered cells ensures insulin production occurs solely in the presence of high glucose and an external stimulus, demonstrating precise control in preclinical models of . Similarly, genetically encoded synthetic beta cells have been developed to autonomously detect and secrete insulin via programmed genetic circuits, offering a bioengineered alternative to mechanical pumps. In diagnostics, biological computing facilitates in vivo pathogen detection through engineered gene circuits that process molecular inputs to produce detectable outputs. Synthetic biology devices, such as those using toehold switches in bacteria or mammalian cells, can sense pathogen-specific nucleic acids or proteins and activate reporter genes for real-time identification within the body, expanding beyond traditional in vitro assays. Engineered live bacteria serve as implantable or ingestible sensors that detect disease biomarkers, including viral or bacterial signatures, and relay signals via bioluminescent or therapeutic outputs, enabling noninvasive monitoring of infections like those caused by SARS-CoV-2 or antibiotic-resistant strains. Tissue engineering benefits from biological computing via organoid-based platforms that simulate human physiology for drug testing and disease modeling. Brain organoids integrated with computational frameworks, known as organoid intelligence (OI), replicate neural circuits to test therapeutics for neurological disorders, providing more accurate predictions of efficacy and toxicity than 2D cell cultures. Recent advancements in OI, as of 2025, have enabled organoids to perform learning tasks and model neurodegenerative diseases like Alzheimer's, supporting the evaluation of personalized treatments. Therapeutic computing employs engineered cells with apoptotic circuits to target cancer precisely, activating cell death pathways only in malignant environments. Engineered suppressor T cells incorporate logic gates to integrate multiple inputs and execute conditional , reducing off-target effects in tumors. Synpoptosis systems, using synthetic protein circuits, enable programmable control of cell death modes like and , overriding resistance mechanisms and enhancing outcomes in preclinical cancer models. Clinical trials in 2025 have advanced implantable bio-computers for real-time monitoring, incorporating elements like logic-gated cells for chronic disease management. For example, trials of off-the-shelf CAR natural killer cells with AND-gate logic for blood cancers demonstrated complete remissions by ensuring only upon dual recognition, highlighting safety in relapsed patients. These developments underscore the transition from bench to bedside, though economic barriers such as high manufacturing costs remain a challenge for widespread adoption.

Environmental and industrial applications

Biological computing has shown promise in environmental applications through engineered microbial systems that act as biosensors for detecting pollutants in real-time. Bacterial networks, such as those in engineered , utilize genetic circuits to sense and respond to heavy metals like and , converting chemical inputs into electrical outputs via extracellular pathways. These systems employ redox-potential-dependent algorithms to multiplex detection, achieving sensitivities below EPA limits (0.1 µM for and 0.045 µM for ) in complex water samples. In , engineered microbes leverage computational design to optimize degradation pathways for environmental contaminants. Synthetic microbiomes, constructed using tools like Super Community Combinations (SuperCC), model metabolic interactions among to enhance breakdown, such as bromoxynil octanoate in soils, achieving over 80% degradation efficiency through predicted flux distributions and cross-feeding mechanisms like hypoxanthine exchange. approaches integrate to direct microbes toward specific catabolic routes, improving the transformation of recalcitrant into harmless byproducts. Industrial biotechnology employs biological computing in metabolic factories for biofuel production, where optimized gene regulatory networks (GRNs) fine-tune expression to maximize yields. In , computational pipelines like ecFactory predict targets for overexpression or , reducing candidates from hundreds to a minimal set that boosts production of biofuels such as alcohols by enhancing pathway fluxes. tools, including CRISPR-Cas9, have enabled up to 3-fold increases in yields in engineered species by redesigning GRNs for inhibitor tolerance and substrate efficiency. Cellular computers facilitate on-site environmental by performing distributed computations in living networks for monitoring ecosystems. In developments, bio-computational systems using spatial in microbial populations enable real-time processing of environmental data, such as pollutant gradients, for applications in and monitoring without external power. These leverage emergent logic gates in cellular consortia to analyze variables like and toxin levels directly . A notable example is the development of synthetic microbial consortia for plastic breakdown, where computational optimization designs cross-feeding networks among to degrade mixed plastic monomers like those from , enabling efficient of waste plastics into usable compounds. Scalability remains a challenge, as deploying these consortia at industrial scales requires addressing stability in diverse environments.

Challenges

Technical and scalability issues

One major technical challenge in biological computing is ensuring reliability amid inherent stochasticity and unintended interactions. Stochastic noise arises from the random nature of biochemical reactions, such as transcriptional bursting and , leading to cell-to-cell variability in levels that can disrupt circuit performance. In synthetic genetic circuits, this intrinsic noise reduces predictability, with protein abundance fluctuations scaling inversely with copy number, often resulting in unreliable outputs for computational tasks. exacerbates these issues, where off-target interactions between pathways cause signal interference; for instance, in engineered E. coli ROS-sensing circuits, introduces relative errors up to 23.5% in detection without compensation. Such errors propagate in multi-component systems, limiting the fidelity of logic operations. Scalability poses further hurdles due to physical and chemical constraints in assembling complex biocomputers. In three-dimensional environments, limits reaction rates, as molecules must rely on for collision and binding, often requiring tens of hours for computations involving hundreds of DNA strands at low concentrations. Confinement strategies like can reduce times to minutes, but scaling to larger networks increases the risk of non-specific bindings, constrained by the finite number of available molecules—far fewer than the exponential solution spaces in problems like the traveling salesman. Multi-step assemblies suffer from yield drops, where crosstalking interactions lead to in the production of desired structures; for heterogeneous systems like protein complexes, yields plummet beyond a critical component threshold without precise concentration tuning. This "yield catastrophe" restricts practical implementations to small-scale prototypes. Speed and throughput in biological computing lag behind electronic systems, with reaction times typically spanning seconds to hours compared to nanosecond clock cycles in silicon-based processors. Biochemical gates, such as those in DNA strand displacement, operate at rates limited by hybridization kinetics, hindering real-time applications. Recent advancements in address these limitations by enabling precise control and parallelization; for example, droplet-based systems facilitate high-throughput analysis of synthetic gene circuits in E. coli, accelerating dynamic testing and reducing diffusion bottlenecks through compartmentalization. By 2025, integrated microfluidic platforms have improved throughput for strain engineering and , achieving scalable automation that mitigates slow reaction propagation. Interoperability remains a barrier, as the lack of standardized biological parts leads to inconsistent integration across circuits and platforms. Efforts like the Global Biofoundry Alliance promote modular workflows using ontologies such as SBOL for DNA designs and LabOp for operations, enabling reproducible assembly of genetic components without custom redesign. These standards facilitate exchange of parts like promoters and ribosome binding sites, reducing variability in multi-lab collaborations. An illustrative case is , where error rates in strand displacement gates—stemming from leakage and incomplete reactions—range from 1-10% per operation, necessitating redundancy to achieve reliable .

Ethical and economic considerations

The development of biological computing raises significant ethical concerns, particularly regarding risks associated with engineered organisms. Engineered biological systems, such as synthetic cells or used for computation, could inadvertently escape containment and disrupt ecosystems or human health if they exhibit unintended behaviors or mutations. Additionally, the dual-use potential of these technologies—where research intended for beneficial computing applications could be repurposed for bioweapons—poses a substantial threat, as tools might enable the creation of novel pathogens with enhanced or resistance. For instance, advancements in gene editing and organoid intelligence could lower barriers to designing biological agents, prompting calls for stricter oversight to mitigate risks. Furthermore, the use of organoids in computing has sparked debates on the potential for emergent , necessitating ethical frameworks for the welfare of such systems. In the context of organoid intelligence and synthetic cognition, ethical considerations extend to obtaining informed consent from donors of biological materials, such as stem cells used to generate organoids. Project-specific consent procedures are recommended to ensure donors are fully informed about the potential uses, including computational applications and risks of emergent sentience. Additionally, frameworks for monitoring the potential emergence of consciousness in bio-AI systems involve assessing neural activity patterns and implementing oversight mechanisms, such as ethical review boards, to evaluate moral status and prevent unintended suffering. Equity issues further complicate the ethical landscape, with access disparities in biotechnology exacerbating global inequalities. High costs and concentrated development in wealthier nations limit the benefits of biological computing to privileged regions, potentially widening the gap in healthcare and computational resources for underserved populations. Intellectual property (IP) frameworks on living systems, such as patented synthetic genomes or designs, intensify these concerns by restricting to foundational biological components, hindering collaborative innovation in developing countries. Economically, biological computing demands substantial due to the of R&D, including the cultivation and integration of biological components like organoids. For example, establishing facilities for organoid-based bioprocessing can require multimillion-dollar setups for controlled environments, contributing to overall biotech R&D expenditures that often exceed billions annually across related fields. Market projections indicate strong growth potential, with the biological computers sector valued at USD 18.7 billion in 2024 and expected to expand significantly, driven by applications in and neuromorphic computing, though precise 2030 figures vary by subfield. Regulatory challenges include navigating IP hurdles in synthetic genomes, where traditional patent laws struggle to address the reproducibility and evolution of living computational elements, leading to disputes over ownership and licensing. In 2025, the FDA issued draft guidance on using artificial intelligence to support regulatory decisions for biological products, which indirectly applies to bio-implants and hybrid bioelectronic systems by emphasizing data integrity and risk assessment in development pipelines. Sustainability considerations highlight the environmental trade-offs of biological computing, as lab production of engineered organisms generates significant eco-footprints through energy-intensive bioreactors and chemical waste. However, these systems offer green benefits, such as lower energy consumption compared to silicon-based computing—potentially reducing carbon emissions in data processing—while promoting bio-based alternatives that align with circular economy principles. Initiatives like green lab practices are increasingly adopted to minimize waste in organoid cultivation, balancing innovation with ecological responsibility.

Future directions

Emerging technologies

Recent advances in intelligence have introduced microfluidic- hybrid systems that enable learning capabilities in organoids, merging biological neural tissues with for enhanced computational processing. These hybrids utilize to sustain organoid cultures while algorithms decode and encode electrophysiological signals, allowing organoids to perform adaptive tasks such as . In 2025, such systems demonstrated rudimentary learning through -mediated training protocols on 3D interfaces, where organoids responded to stimuli with improved accuracy over repeated exposures. Complementing these developments, the allocated nearly $32 million in August 2025 to accelerate -driven approaches in and , strengthening the U.S. through innovations in . Fusion of DNA logic with neurochemistry represents a promising direction, where DNA strand displacement reactions emulate neural signaling pathways to create molecular neural networks. Researchers have developed DNA-based winner-take-all encoders that process multiple inputs to mimic synaptic competition, as demonstrated in 2018 work applying DNA logic circuits to , with potential extensions to modulate cell surface receptors and influence neural-like signaling cascades. These systems enable programmable control over biological signals, such as autonomous activation of pathways in response to environmental cues, bridging synthetic with endogenous neurochemical processes. A notable example includes DNA neural networks trained on example datasets to solve optimization problems, demonstrating error-tolerant computation at the molecular scale. Morphogenetic computing in cellular automata offers models beyond traditional Turing machines, leveraging self-organizing biological patterns for non-von Neumann architectures. A 2024 Communications of the ACM paper outlined how living cellular computers, inspired by , exploit to achieve through continuous state evolution and spatial dynamics. These automata simulate developmental phenotypes using elementary rules that generate complex, adaptive structures, enabling computations unattainable by discrete Turing equivalents, such as solving halting problems via biological growth processes. Quantum-bio hybrids are exploring entanglement-inspired states in DNA to enhance information processing in biological systems. In 2024, theoretical models positioned DNA as a quantum computer capable of maintaining entangled states during base pairing and replication, facilitating coherent energy and . These approaches suggest potential for entanglement-driven signaling in DNA nanostructures, where correlated states across molecules enable parallel processing akin to quantum bits. A key trend in 2025 involves AI-optimized for sustainable computing, where accelerates the design of bioengineered circuits that minimize energy use compared to silicon-based systems. AI tools streamline editing and pathway optimization in synthetic organisms, enabling biocomputers that operate on ambient biological substrates for eco-friendly . This convergence supports scalable, low-carbon alternatives, with market analyses projecting biological computing growth driven by such AI integrations. Additionally, the 2025 (iGEM) competition showcased innovations in , including AI-guided designs for biocomputational applications in therapeutics and environmental sensing.

Potential societal impacts

Biological computing holds transformative potential to revolutionize by leveraging the energy efficiency and parallel processing capabilities inherent in biological systems, potentially reducing the environmental footprint of data centers that currently consume vast amounts of . Projections indicate that by 2030, synthetic biology-derived products, including those enabling biological computation, could permeate most industrial sectors, fostering innovations in materials, fuels, and therapeutics that underpin a bio-based . This shift may usher in ubiquitous , where bio-computational devices enable real-time, patient-specific diagnostics and treatments, enhancing healthcare accessibility worldwide. However, these advancements carry risks, including job displacement in traditional computing and sectors as through bio-foundries streamlines and manufacturing processes, potentially reducing demand for manual roles. Additionally, the release of engineered biological systems poses threats to , as self-replicating organisms could disrupt ecosystems through unintended or competitive exclusion of . On the societal benefits side, biological computing could advance solutions by enabling carbon-neutral , such as engineered microbes for efficient carbon capture and sustainable production, contributing to global decarbonization efforts. It also promises greater by democratizing access to advanced diagnostics and therapies in underserved regions, bridging gaps in global healthcare disparities through scalable, low-cost biological platforms. Looking toward 2030 and beyond, visions include implantable biocomputers that interface directly with human neural systems for cognitive enhancement, potentially augmenting and capabilities. Such developments will spark ethical debates surrounding "living machines," questioning the moral status of biohybrid entities that blur lines between artificial and biological life, raising concerns over , , and human augmentation inequities. Addressing these implications necessitates robust policy frameworks, including international treaties to govern the development and deployment of biological computing technologies, ensuring , equitable access, and mitigation of dual-use risks through collaborative oversight. Current regulatory inadequacies highlight the urgency for specialized that addresses unique challenges like ownership and environmental safeguards.

References

  1. https://www.nature.com/articles/s41467-019-13232-z
  2. https://cacm.acm.org/research/biocomputation-moving-beyond-turing-with-living-cellular-computers/
  3. https://www.science.org/doi/10.1126/science.7973651
  4. https://www.nature.com/articles/35002131
  5. https://www.nature.com/articles/35002125
  6. https://www.nature.com/articles/s41467-024-49264-3
  7. https://www.nature.com/articles/s12276-025-01530-0
  8. https://www.sciencedirect.com/science/article/pii/S294992162500002X
  9. https://viterbischool.usc.edu/news/2025/10/artificial-neurons-developed-by-usc-team-replicate-biological-function-for-improved-computer-chips/
  10. https://www.pnas.org/doi/10.1073/pnas.1119911109
  11. https://theconversation.com/biological-computers-could-use-far-less-energy-than-current-technology-by-working-more-slowly-245962
  12. https://www.dna.caltech.edu/Papers/winfree_thesis.pdf
  13. https://pubs.acs.org/doi/10.1021/acssynbio.3c00474
  14. https://www.dna.caltech.edu/Papers/sCRN_computation_TR2007.pdf
  15. https://inria.hal.science/hal-01519828/file/FLBP17cmsb.pdf
  16. https://apps.dtic.mil/sti/tr/pdf/AD0655232.pdf
  17. https://www.nature.com/articles/nature04586
  18. https://www.nature.com/articles/ncomms15459
  19. https://royalsocietypublishing.org/doi/10.1098/rsta.2014.0218
  20. https://www.nature.com/articles/nrmicro3239
  21. https://www.cambridge.org/core/journals/mrs-bulletin/article/organic-neuromorphic-devices-past-present-and-future-challenges/A641A09A984E4D18AED39AD90B70CA3F
  22. https://www.nature.com/articles/s41586-022-04515-7
  23. https://newatlas.com/brain/cortical-bioengineered-intelligence/
  24. https://www.sciencedirect.com/science/article/abs/pii/S0958166902002811
  25. https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004795
  26. https://cs.stanford.edu/people/eroberts/courses/soco/projects/2003-04/dna-computing/evaluation.htm
  27. https://pmc.ncbi.nlm.nih.gov/articles/PMC1315819/
  28. https://www.pnas.org/doi/10.1073/pnas.2220022120
  29. https://www.sandia.gov/app/uploads/sites/210/2023/10/Comet23-slides-topost.pdf
  30. https://arxiv.org/pdf/2106.07236
  31. https://personal.ntu.edu.sg/mjjshu/JACS23.pdf
  32. https://www.newscientist.com/article/mg25834422-100-the-unique-promise-of-biological-computers-made-from-living-things/
  33. https://pmc.ncbi.nlm.nih.gov/articles/PMC4946074/
  34. https://pmc.ncbi.nlm.nih.gov/articles/PMC4923711/
  35. https://www.science.org/doi/10.1126/science.1200520
  36. https://pubs.acs.org/doi/10.1021/ja016756v
  37. https://academic.oup.com/nar/article/26/22/5203/1023766
  38. https://www.nature.com/articles/s41467-025-57016-0
  39. https://pmc.ncbi.nlm.nih.gov/articles/PMC10720954/
  40. https://www.osti.gov/servlets/purl/1468075
  41. https://pubs.acs.org/doi/10.1021/jacs.1c12373
  42. https://www.nature.com/articles/s41467-021-23744-2
  43. https://www.mdpi.com/1424-8220/23/13/5782
  44. https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2022.906363/full
  45. https://onlinelibrary.wiley.com/doi/full/10.1002/aisy.202200312
  46. https://www.frontiersin.org/journals/science/articles/10.3389/fsci.2023.1017235/full
  47. https://www.biorxiv.org/content/10.1101/2024.03.15.585237v5
  48. https://eleks.com/research/organoid-intelligence/
  49. https://www.sciencedirect.com/science/article/pii/S2542660525001854
  50. https://www.cell.com/cell-systems/fulltext/S2405-4712(24)00311-9
  51. https://pmc.ncbi.nlm.nih.gov/articles/PMC4571725/
  52. https://www.nature.com/articles/s41467-020-20656-5
  53. https://parts.igem.org/Main_Page
  54. https://www.cell.com/trends/biotechnology/pdf/S0167-7799(15)00274-7.pdf
  55. https://pmc.ncbi.nlm.nih.gov/articles/PMC7609024/
  56. https://pmc.ncbi.nlm.nih.gov/articles/PMC6632179/
  57. https://pmc.ncbi.nlm.nih.gov/articles/PMC6182181/
  58. https://pmc.ncbi.nlm.nih.gov/articles/PMC3564264/
  59. https://www.researchgate.net/publication/397208746_A_subnanolitre_tetherless_optoelectronic_microsystem_for_chronic_neural_recording_in_awake_mice
  60. https://pmc.ncbi.nlm.nih.gov/articles/PMC2808222/
  61. https://pubs.acs.org/doi/10.1021/acsnano.5c04885
  62. https://archive.aps.org/smt/2025/mar-s54/4/
  63. https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2023.1052607/full
  64. https://onlinelibrary.wiley.com/doi/full/10.1002/adfm.202111271
  65. https://pmc.ncbi.nlm.nih.gov/articles/PMC4664311/
  66. https://jbioleng.biomedcentral.com/articles/10.1186/s13036-023-00379-z
  67. https://aapsopen.springeropen.com/articles/10.1186/s41120-025-00109-3
  68. https://www.science.org/doi/10.1126/science.adl4793
  69. https://www.cell.com/cell/fulltext/S0092-8674(24)00348-6
  70. https://www.aacr.org/about-the-aacr/newsroom/news-releases/off-the-shelf-car-natural-killer-cell-therapy-with-logic-gates-elicits-complete-remissions-in-relapsed-or-refractory-blood-cancers/
  71. https://www.nature.com/articles/s41423-024-01153-x
  72. https://www.nature.com/articles/s41467-025-62256-1
  73. https://www.nature.com/articles/s41467-024-49098-z
  74. https://pubmed.ncbi.nlm.nih.gov/30198342/
  75. https://www.pnas.org/doi/10.1073/pnas.2417322122
  76. https://pubs.rsc.org/en/content/articlehtml/2025/ya/d5ya00118h
  77. https://www.nature.com/articles/s41467-025-58492-0
  78. https://data.mendeley.com/datasets/zyk37sxsbt/1
  79. https://pmc.ncbi.nlm.nih.gov/articles/PMC7611518/
  80. https://www.nature.com/articles/s41467-019-12021-y
  81. https://link.springer.com/chapter/10.1007/978-981-19-9891-1_3
  82. https://www.nature.com/articles/s41467-023-43168-4
  83. https://link.springer.com/article/10.1007/s11047-018-9686-x
  84. https://www.nature.com/articles/s41587-025-02709-6
  85. https://www.nature.com/articles/s41467-025-61263-6
  86. https://www.pnas.org/doi/10.1073/pnas.1806859115
  87. https://www.sciencedirect.com/science/article/pii/S2588933823000201
  88. https://www.ncbi.nlm.nih.gov/books/NBK11496/
  89. https://carnegieendowment.org/research/2024/10/mitigating-risks-from-gene-editing-and-synthetic-biology-global-governance-priorities?lang=en
  90. https://www.nature.com/articles/d41586-025-03633-0
  91. https://pmc.ncbi.nlm.nih.gov/articles/PMC10844458/
  92. https://pmc.ncbi.nlm.nih.gov/articles/PMC10796793/
  93. https://pmc.ncbi.nlm.nih.gov/articles/PMC11882709/
  94. https://www.who.int/publications/i/item/9789240071818
  95. https://ecoevopublisher.com/index.php/gmo/article/html/3835/
  96. https://www.nsf.gov/news/nsf-invests-14m-bioengineered-systems-ethical-biocomputing
  97. https://www.emergenresearch.com/fr/industry-report/biological-computers-market
  98. https://www.einfolge.com/blog/decoding-synthetic-biology-patents-ip-challenges-in-dna-programming
  99. https://www.fda.gov/media/184830/download
  100. https://pmc.ncbi.nlm.nih.gov/articles/PMC11078267/
  101. https://www.green-algorithms.org/assets/publications/2023_GREENER_NatCompSci.pdf
  102. https://www.astrazeneca.com/content/astraz/what-science-can-do/topics/sustainability/creating-culture-sustainability-across-labs.html
  103. https://www.researchgate.net/publication/388493358_Organoid_Intelligence_and_Biocomputing_Advances_Current_Steps_and_Future_Directions
  104. https://www.sciencedirect.com/science/article/pii/S2950489925000521
  105. https://www.nsf.gov/news/nsf-invests-nearly-32m-accelerate-novel-ai-driven-approaches
  106. https://www.sciencedirect.com/science/article/abs/pii/S0893608025003661
  107. https://www.nature.com/articles/s41593-018-0173-z
  108. https://advanced.onlinelibrary.wiley.com/doi/10.1002/advs.202507060
  109. https://www.caltech.edu/about/news/dna-based-neural-network-learns-from-examples-to-solve-problems
  110. https://www.researchgate.net/publication/379838250_Elementary_Morphogenetic_Cellular_Automata_for_Modeling_Developmental_Phenotypes
  111. https://www.nature.com/articles/s41598-024-62539-5
  112. https://pubmed.ncbi.nlm.nih.gov/40850465/
  113. https://www.nature.com/articles/s44385-025-00021-1
  114. https://www.startus-insights.com/innovators-guide/synthetic-biology-trends/
  115. https://www.consegicbusinessintelligence.com/press-release/biological-computers-market-growth-insights-2025
  116. https://www.labiotech.eu/trends-news/igem-2025-synthetic-biology-highlights/
  117. https://phys.org/news/2025-06-biology-future.html
  118. https://pmc.ncbi.nlm.nih.gov/articles/PMC11458441/
  119. https://pmc.ncbi.nlm.nih.gov/articles/PMC9641226/
  120. https://www.nature.com/articles/s41467-024-46865-w
  121. https://www.pnas.org/doi/10.1073/pnas.2310458121
  122. https://www.weforum.org/stories/2020/06/prevent-cyber-bio-security-threats-covid19-governance/
  123. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5239551
Add your contribution
Related Hubs
User Avatar
No comments yet.