Building biological computers from bacterial populations

Project Information

SynBioBrain

Grant agreement ID: 770835

DOI

10.3030/770835

Project closed

EC signature date 23 April 2018

Start date 1 May 2018

End date 31 October 2024

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 1 998 025,00

EU contribution

€ 1 998 025,00

Coordinated by

UNIVERSITY COLLEGE LONDON
United Kingdom

Periodic Reporting for period 5 - SynBioBrain (Building biological computers from bacterial populations)

Reporting period: 2024-05-01 to 2024-10-31

Summary of the context and overall objectives of the project

Biosensor development is undergoing a transformative advance with applications in healthcare diagnostics, environmental monitoring, bioprocessing, and agriculture, all of which have significant implications for public health, environmental safety, and sustainable bio-industries.

Synthetic biology now enables the engineering of whole cell biosensors in which sensing, signal transduction, and output all occur within living bacterial cells. However, current biosensor systems often face a critical limitation: they are typically confined to detecting single signals and executing simple analog responses. This restricts their utility in complex real-world scenarios where simultaneous multi-analyte detection and advanced decision-making are required.

To address this challenge, the project developed a novel strategy by constructing digital biological computers: engineered bacterial communities that perform logical operations through spatial patterning and inter-colony communication. The aim was to overcome the genetic and metabolic constraints of single-cell logic circuits, which limit their scalability to multi-input functions.

Overall objectives:
• Engineer modular bacterial colonies capable of executing simple digital logic functions, via diffusion-based morphogen signaling and spatial arrangement.
• Assemble larger computational networks by combining multiple colonies to perform complex, multi-input logic without further genetic modifications.
• Introduce sender colonies that emit biochemical inputs—such as metabolites—allowing the system to integrate multiple environmental signals.
• Demonstrate computational output through spatially patterned fluorescent colonies, clearly visible under normal light, proving the feasibility of naked-eye biosensor readouts.
• Validate proof-of-principle multi-input biosensors for metabolite monitoring.

This approach transforms living bacterial populations into small-scale biological computers that analogously function like digital circuits—detecting, integrating, and computing multiple signals to generate readable outputs. The result is a versatile platform for multi-analyte biosensing and in vivo decision-making, with the potential for broad societal impact.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

We implemented a digital spatial computing approach based on morphogen gradients, where engineered bacterial colonies act as computational units. Each colony was programmed with a genetic circuit implementing a transfer function, focusing initially on high-pass and band-pass operations. Early prototypes used IPTG-inducible systems responsive to liquid droplets to validate the concept. Once feasibility was established, we transitioned to AHL-based quorum sensing circuits for inter-colony communication.

A key theoretical advance was demonstrating that any Boolean logic function can be decomposed into a set of output colonies, where the overall output is ON if any designated colony fluoresces. This was achieved by developing an algorithm that maps truth tables to spatial colony arrangements, enabling modular design of complex logic without additional genetic engineering.

To support design and optimization, we built two complementary mathematical models:
• A fast, reduced model for rapid exploration of candidate designs.
• A full reaction-diffusion model for detailed simulation and verification. These were integrated into a computational design platform, as described in Fedorec et al. (2024), allowing automated decomposition and simulation of logic circuits.

On the experimental side, we established automated protocols using an OpenTrons liquid-handling robot for precise colony placement on six-well agar plates. In collaboration with Loopbio, we developed a custom imaging system capable of real-time GFP fluorescence monitoring inside an incubator, enabling dynamic observation of spatial patterns.

We successfully constructed multi-input biosensors for metabolites including lactate, acetoacetate, arabinose, and propionate, integrating them into logic-based computational frameworks. These proof-of-principle systems demonstrated visible spatial patterning of fluorescent colonies under normal light, validating the concept of living biocomputers for multi-analyte sensing.

Exploitation and Dissemination:
The results have been disseminated through the publication in Nature Communications (Fedorec et al., 2024) and presentations at synthetic biology conferences. The design platform and protocols provide a foundation for future applications in diagnostics, environmental monitoring, and bioprocess control, with potential for open-source release to accelerate adoption.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

We have developed a modular spatial computing platform that enables the implementation of arbitrary digital logic functions using engineered bacterial colonies arranged on a “biological chip.” Unlike conventional single-cell genetic circuits, which are constrained by metabolic burden and limited scalability, our approach leverages distributed computation across multiple colonies, each performing a simple transfer function. This architecture allows complex logic to be achieved without additional genetic modifications, simply by reconfiguring colony positions on the agar substrate.
This represents a significant advance beyond the state of the art in synthetic biology and biosensing:
• Scalability and flexibility: Any Boolean logic function can be realized through spatial arrangement, enabling rapid adaptation to new sensing tasks.
• Integration of multi-input biosensors: Our system can incorporate diverse bacterial sensors for metabolites (e.g. lactate, acetoacetate, arabinose, propionate, pH), supporting multi-analyte detection and decision-making.
• Automated design and deployment: The integrated computational platform combines algorithmic decomposition of truth tables with reaction-diffusion modeling, streamlining the design process for complex biocomputations.
• Visible outputs: Spatial patterning of fluorescent colonies provides an intuitive, naked-eye readout, eliminating the need for specialized equipment.

Ultimately, this work establishes a foundation for programmable living materials and distributed biological computing, with transformative potential across healthcare, environmental monitoring, and smart biomaterials.

figure1.png