Description

Abstract

Biological computing is a promising field with potential applications, yet past engineered genetic circuits for logic gates have suffered from excessive complexity, cumbersome operation, and reliance on multiple input substances.

To overcome these issues, we decided not to confine computation within cells. Therefore, we created the LOGIC project, proposing an in vitro logic computing system based on bacterial quorum sensing and spatial diffusion, including lightweight adders and a light-induced QS molecular degradation system.

We have also developed LOGIC Toolkit, which provides new methods for the next generation of iGEMers and researchers to explore biological computing.

Introduction

Engineering biological systems capable of computation has long been a goal of biology, which can be applied in fields such as biosafety or environmental monitoring, or as a basic component in computer construction. [1]

Background

In binary addition, the circuit consists of several basic logic gates:

AND gate:

Outputs 1 only when both inputs are 1.
Input A Input B Output (A AND B)
0 0 0
0 1 0
1 0 0
1 1 1
OR gate:

Outputs 1 as long as one input is 1.
Input A Input B Output (A OR B)
0 0 0
0 1 1
1 0 1
1 1 1
XOR gate:

Outputs 1 when the two inputs are different.
Input A Input B Output (A XOR B)
0 0 0
0 1 1
1 0 1
1 1 0

A binary adder consists of a half-adder and full-adders (Figure 1). The half-adder is used to process the first bit calculation, while the full-adder is used to process the second and higher bit calculations. By connecting the half-adder and full-adder in series, binary addition can be achieved.

Half Adder
Figure 1. Half-Adder & Full-Adder

Biological computing is the process of using cells to simulate the above logic gates and adders.

Current biological computing scheme

To date, the most common paradigm in biological computing has been to engineer logic circuits into gene regulatory networks within cells.[2][3][4]

However, this method requires complex genetic circuits to achieve logic gate functions, and it's complexity is constrained by the number of available orthogonal components and metabolic burden. In addition, each new function requires extensive genetic engineering, which makes the system inflexible.

Moon et al., 2012
Figure 2. Constructing multiple logic gates in single cells using regulatory elements of the bacterial type III secretion system (T3SS). [2]
BIT2024
Figure 3. Constructing hierarchical AND gates within a single cell to mimick the electronic combination lock.[3]
ETH_Zurich2014
Figure 4. Implementing the Sierpinski Triangle in bacterial colonies using logic gates (XOR gates). [4]

Our Solution

Logic Gates & Adders

Through computational and experimental evidence, it has been proven that a single modular colony can perform simple digital logic operations.[1] By combining multiple colonies, complex adders can be constructed, thereby eliminating the need for further genetic editing of bacteria. We utilized quorum-sensing systems to implement dual- or multi-input logic gates, which were then assembled into half-adders and full-adders for binary addition, ultimately producing the computational output.

The computational system constructed using this method is highly lightweight and flexible, offering a new approach to biological computing. There are two basic genetic circuits used to perform logical calculations: high-pass and band-pass.[1]

High-pass: Downstream gene expression is activated when the concentration of the signal molecules exceeds the threshold.

Band-pass: Downstream gene expression is activated when the concentration of the signal molecules is within a specific range.

High-pass and Band-pass

We will use these two basic circuits to construct the half-adder and full-adder.

Spatial Computation

OR and AND gates
Figure 5. OR gate and AND gate implementation through spatial configuration

With "high-pass" and other engineered bacteria, we can now proceed to construct logic gates. Unlike conventional logic gates, LOGIC operates as a form of spatial computation: the input is defined by the local concentration of AHL molecules, while the output is reported through the fluorescence intensity of sfGFP.

The characteristic of a “high-pass” colony is that it requires the AHL input concentration to reach a certain threshold before producing an output. It is easy to imagine that the closer the AHL input point is to the processor colony, the higher AHL concentration produced at the high-pass colony.

For easy understanding, we now roughly classify AHL input points by their distance from the engineered bacteria into “nearby AHL input points” and “far AHL input points”. For a “high-pass” engineered processor bacterium, if we provide two “nearby AHL input points”, then a signal at just one input point is sufficient to activate the “high-pass” bacterium—clearly, this constitutes an OR gate. Similarly, if we provide the “high-pass” with two “far AHL input points”, it is easy to see that only when both input points carry AHL signals will the “high-pass” bacterium be activated, yielding an AND gate (Figure 6)!

Full Adder
Figure 6. Full-adder implementation with three inputs and two outputs

By employing "high-pass" and "band-pass" composite module combined with spatially configured input, we can achieve a series of logic gates including OR, AND and XOR, among others. Under multi-input and multi-output conditions, more complex logic functions can be realized. For example, a full-adder

By employing "high-pass" and "band-pass" composite module combined with spatially configured input, we can achieve a series of logic gates including OR, AND and XOR, among others. Under multi-input and multi-output conditions, more complex logic functions can be realized. For example, a full-adder can be readily implemented using three inputs and two outputs, with the results mediated by one “band-pass” and one “high-pass” composite module (Figure 7).

This provides the foundational components for computing arbitrary logic functions and building biological computers. What's more, effective signal transmission between colonies is also needed. After constructing the logic gate colonies and the connector colonies, we can then build a binary adder.

Full Adder

Orthogonal Systems

Orthogonal AHL System
Figure 7. The role of orthogonal AHL molecular systems

Perhaps you have already noticed that spatial computation imposes strict requirements on the distances and spatial arrangement between AHL signal input points and downstream engineered bacteria—for example, preventing spatial crosstalk among inputs and outputs, and ensuring that the distance between the engineered bacteria and the AHL input sources is appropriate.

As the complexity of logic computation increases, the system complexity rises sharply. To raise the upper limit of our biological circuits' complexity within limited space while maintaining signal accuracy, we employed orthogonal AHL molecular systems.[5]

Degradation Modules

Light Control Part
Figure 8. Schematic diagram of light-controlled components

After constructing the computation in space, we further aimed to achieve temporally sustained and repeatable computation.

Revisiting our project LOGIC, the current design cannot support our adder to perform multiple rounds of addition. To enable multi-round operations, the existing computational state must be reset—i.e., residual AHL in the agar and the fluorescence used as the output signal need to be cleared.

To provide controllable and timely clearing and resetting of computational results, we designed an opto-degradation module. To degrade residual AHL molecules, we employed the quorum-sensing-molecules-degrading enzyme AiiA[6], and fused it with the blue-light-responsive photoreceptor domain VVD, enabling light-controlled regulation of its enzymatic activity.[7] For the fluorescence outputs, we add a protein degradation tag based on the AsLOV2 protein to fluorescent reporters, enabling light-inducible degradation of the fluorescent signal.[8]

LOGIC Toolkit

To facilitate understanding and application of LOGIC, we digitized experimental data and performed curve fitting for both diffusion dynamics and cellular responses. By characterizing the diffusion field and cellular response, we can design input spatial patterns that yield specific logic outputs. We also developed visualization software based on the fitted curves to facilitate understanding of LOGIC and predict outcomes. Additionally, to minimize operational artifacts, we designed and and build a software-driven precision positioning hardware device. Finally, we integrated the components and tools developed during experimentation into the LOGIC toolkit.

Toolkit Component 1
Figure 9. (A) Orthogonal AHL biosensors and diffusion-response mapping. (B) Logic gate modules. (C) Light-induced degradation. (D) Modeling method of spatial computation. (E) Software for visualization and debugging. (F) Hardware for easier colony localization.
Learn more about our design

References

  1. Fedorec, A. J. H., Treloar, N. J., Wen, K. Y., Dekker, L., Ong, Q. H., Jurkeviciute, G., Lyu, E., Rutter, J. W., Zhang, K. J. Y., Rosa, L., Zaikin, A., & Barnes, C. P. (2024). Emergent digital bio-computation through spatial diffusion and engineered bacteria. Nature Communications, 15(1). https://doi.org/10.1038/s41467-024-49264-3
  2. Moon, T. S., Lou, C., Tamsir, A., Stanton, B. C., & Voigt, C. A. (2012). Genetic programs constructed from layered logic gates in single cells. Nature, 491(7423), 249–253. https://doi.org/10.1038/nature11516
  3. Team:BIT-China/Project - 2014.igem.org. (n.d.). https://2014.igem.org/Team:BIT-China/Project
  4. Team:ETH Zurich/data - 2014.igem.org. (n.d.). https://2014.igem.org/Team:ETH_Zurich/data#Gene_Circuit
  5. Jonkergouw, C., Savola, P., Osmekhina, E., Van Strien, J., Batys, P., & Linder, M. B. (2023). Exploration of chemical diversity in intercellular quorum sensing signalling systems in prokaryotes. Angewandte Chemie International Edition, 63(2). https://doi.org/10.1002/anie.202314469
  6. Fetzner, S. (2014). Quorum quenching enzymes. Journal of Biotechnology, 201, 2–14. https://doi.org/10.1016/j.jbiotec.2014.09.001
  7. Han, T., Chen, Q., & Liu, H. (2016). Engineered photoactivatable genetic switches based on the bacterium phage T7 RNA polymerase. ACS Synthetic Biology, 6(2), 357–366. https://doi.org/10.1021/acssynbio.6b00248
  8. Tague, N., Coriano-Ortiz, C., Sheets, M. B., & Dunlop, M. J. (2023). Light-inducible protein degradation in E. coli with the LOVdeg tag. eLife, 12. https://doi.org/10.7554/elife.87303