Week 7 HW: Genetic Circuits Part II: Neuromorphic Circuits

Assignment Part 1: Intracellular Artificial Neural Networks (IANNs)

1) What advantages do IANNs have over traditional genetic circuits, whose input/output behaviours are Boolean functions?

The main limitation of traditional genetic circuits is their restricted control over the strength, timing, and cellular context of therapeutic effects, as their input–output behaviour is usually constrained to simple Boolean logic. In contrast, IANNs can provide finer control of gene expression and cellular behaviour by tuning promoters, repressors, and other genetic components, and they can also sum and weight multiple inputs within a single network, rather than relying on many individually wired logic gates, which become complex and error‑prone.

References:

  • Carneiro, D.C., Rocha, C., Patrícia K. F. Damasceno, Barbosa, V. and Soares, P. (2024). Therapeutic applications of synthetic gene/genetic circuits: a patent review. Frontiers in Bioengineering and Biotechnology, 12. doi:https://doi.org/10.3389/fbioe.2024.1425529.
  • Seak, L.C.U., Lo, O.L.I., Suen, W.C.-W. and Wu, M.-T. (2021). Next-generation biocomputing: mimicking artificial neural network with genetic circuits. bioRxiv. doi:https://doi.org/10.1101/2021.03.12.435120.

    • 2) Describe a useful application for an IANN; include a detailed description of input/output behaviour, as well as any limitations an IANN might face to achieve your goal.

    Exploring my Final Project 2 idea (Bacillus subtilis biofilm paint), creating IANN genetic circuits in this spore-forming bacterium could transform the living paint into a smarter system for VOC pollution mapping. Rather than only showing simple surface darkening, the paint could form responsive biofilm patterns that record not just the presence of pollution, but also the environmental context in which exposure occurred. This would create a richer “pollution memory map,” making the project more useful for long-term environmental monitoring, scientific analysis, and the development of an educational and artistic installation.

    I was inspired by the paper “A single-layer artificial neural network with engineered bacteria” by Sarkar et al. (2020). In this work, the authors designed an IANN in which extracellular chemical signals acted as inputs. These signals were linearly combined and processed through a nonlinear log-sigmoid activation function generated by synthetic genetic circuits. Each artificial neuron was encoded in bacteria, with weight and bias values tuned by engineering molecular interactions within the cell. This allowed bacterial “neurons” to perform specific logical functions, effectively creating an ANN using living cells.

    For this living paint system, the IANN would integrate several environmental inputs. Different VOCs could be detected through separate promoter pathways, where each promoter contributes a weighted signal to the final pigmentation response. For example, VOC1 and VOC2 could each activate distinct sensory promoters whose outputs are combined inside the circuit. Additional contextual inputs could include airflow-sensitive signalling, which would influence spore germination and help distinguish stagnant pollution from outdoor exposure, and UV exposure, which would act as an indicator of spore viability and environmental stress. The output would be a nonlinear pigmentation or biofilm pattern response, where the intensity, shape, or layering of darkening reflects the combined history of the environment’s pollution.

    A major limitation of this system is the dormant nature of Bacillus subtilis spores. Because IANN function depends on active transcription and translation, the circuit can only “compute” during short wet periods when spores germinate. During prolonged dry weather or strong UV exposure, the spores remain dormant, meaning the paint cannot continuously process inputs. This could lead to inconsistent activation and patchy or faded biofilm patterns over time. Another important challenge is mutation and genetic instability. A multi-gene IANN containing several nodes alongside VOC sensing pathways would be at high metabolic risk of creating mutations, especially over months of outdoor exposure. There are also regulatory and public acceptance limitations. Engineered spore-forming bacteria with multi-input decision-making circuits would likely face greater biosafety scrutiny than simpler VOC-degrading systems. Finally, the nonlinear nature of the IANN output may make interpretation difficult. Because the visible darkening is based on multiple weighted inputs, complex shade gradients may be hard for scientists to decode into a clear pollution history. While this complexity improves sensing, it may reduce interpretability unless paired with computational decoding tools.

    References:

  • Sarkar, K., Bonnerjee, D. and Bagh, S. (2020). A single-layer artificial neural network with engineered bacteria. doi:https://doi.org/10.48550/arXiv.2001.00792.

    • 3) Draw a diagram for an intracellular multilayer perceptron where layer 1 outputs an endoribonuclease that regulates a fluorescent protein output in layer 2.

    cover image cover image
    Assignment Part 2: Fungal Materials

    1. What are some examples of existing fungal materials, and what are they used for? What are their advantages and disadvantages over traditional counterparts?

    Fungal materials have been an increasing area of development for the production of sustainable alternative materials, in particular, mycelium, the underground root network of fungi. There are two primary ways of processing mycelium into a material.

    1. Composite mycelium materials (CMM): An organic growing material (medium) like woodchips or agricultural waste is pressed inside a mould and is then inoculated with mycelium, allowing for the network to grow and fill in the cracks of the medium, creating a solid material. In order to stop the mycelium from growing, the material is dehydrated through heat or compression. The outcome is a solid foam-like material whose characteristics can be manipulated depending on the selected growing medium. All the parts of this processing technique are composed of biological elements, resulting in its biodegradability over a short period of time.

    2. Pure mycelium materials (PMM): The mycelium is grown without a medium, but still has a nutrient source to allow its growth. Once grown. It can be combined with additional processing techniques, allowing greater control over the final product. For example, the material can be coated with a chemical to increase its durability, but this can make it non-biodegradable. In comparison to CMM, it allows the production of flexible materials and its characteristics can be altered depending on the strain of fungi used.

    Examples of mycelium materials:

    Mycelium has been expanding in the world of textiles thanks to a 2018 patent that enabled the use of fungi in textile production. Mycelium leather has been tested and proven to, in some cases, have a higher tensile strength and resistance to tearing than synthetic leather (Gandia et al., 2021). However, it may degrade more quickly unless treated with a plastic coating. The company MycoWorks has been exploring how to replicate woven fabric with mycelium by directing the orientation of its growth. In theory, this would allow to create a stronger woven material. Another example of mycelial materials includes packaging such as CMM styrofoam (hemp and mycelium), developed by a company called Ecoactive (Ecoactive LLC, 2026). Furthermore, it is being explored as a paper-like material produced by growing pellets of fungi that are agitated into a pulp. This is a fast process, but it is prone to contamination. In construction, mycelium has a lot of advantageous properties, such as high thermal insulation and fire resistance, whilst still allowing for good airflow. Currently, it is being used for carbon-neutral architecture (Myceen, 2024) in a variety of materials (bricks, insulation, pastes…).

    Lastly, I discovered that mushroom material can be grown to carry out an electrical wiring pattern. In Ecovative’s patent, a fungus is grown on a nutrient medium (like potato dextrose agar or broth) that contains metal salts such as copper sulfate or copper chloride. As the mycelium spreads, it absorbs these metals and forms thin sheets that grow in the shape of the desired circuit pattern, creating conductive “wires” within the fungal material (Cerimi et al., 2019). In comparison to standard circuits, it would essentially be compostable as well as having a lower carbon footprint in production. In terms of functionality, being a ‘living electronic’ could, in theory, be combined with self-regenerative repair and sensing (Adamatzky, 2022).

    Advantages

  • They are biodegradable, so they can break down over a relatively short time.
  • Their processing generally involves lower toxicity and energy use than that produced from petrochemicals.
  • By changing the fungal strain, substrate and post-processing, the same system can be adapted for many different applications.
  • They can have a lower overall carbon footprint than conventional materials, especially when they replace high‑impact products.
  • In some cases, they can be designed as “living” or responsive materials (e.g. self‑healing structures or conductive), which is not possible with inert synthetic materials.

    • Disadvantages

  • When kept fully bio-based and uncoated, they are often less durable and more sensitive to biological decay than traditional materials.
  • Their mechanical properties can be more variable and sometimes lower than standardised industrial materials, which makes engineering and certification more challenging.
  • They usually take longer to produce because the mycelium needs time to grow and colonise the substrate, slowing down manufacturing cycles.
  • At the moment, they are often more expensive than mass‑produced plastics, foams or leathers.
  • They can be sensitive to environmental conditions such as humidity and contamination, which can limit certain outdoor or high‑load uses unless additional protection is added.
  • Being a relatively new class of materials, they come with more uncertainty around regulation, long‑term performance and risk assessment compared to well‑known conventional materials.

    • References:

  • Adamatzky, A., Ayres, P., Beasley, A.E., Chiolerio, A., Dehshibi, M.M., Gandia, A., Albergati, E., Mayne, R., Nikolaidou, A., Roberts, N., Tegelaar, M., Tsompanas, M.-A., Phillips, N. and Wösten, H.A.B. (2022). Fungal electronics. Biosystems, 212, p.104588. doi:https://doi.org/10.1016/j.biosystems.2021.104588.
  • Ecoactive (n.d.). Ecovative. [online] Ecovative. Available at: https://ecovative.com [Accessed 26 Mar. 2026].
  • Cerimi, K., Akkaya, K.C., Pohl, C., Schmidt, B. and Neubauer, P. (2019). Fungi as source for new bio-based materials: a patent review. Fungal Biology and Biotechnology, 6(1). doi:https://doi.org/10.1186/s40694-019-0080-y.
  • Gandia, A., van den Brandhof, J.G., Appels, F.V.W. and Jones, M.P. (2021). Flexible Fungal Materials: Shaping the Future. Trends in Biotechnology, 39(12). doi:https://doi.org/10.1016/j.tibtech.2021.03.002.
  • Myceen (2024). Materials for carbon-neutral architecture. [online] Myceen.com. Available at: https://myceen.com/about [Accessed 26 Mar. 2026].
  • Pale Blue dots (2026). Fungi-based materials. [online] Notion. Available at: https://pbdvc-research.notion.site/Fungi-based-materials-3b088667784f416e90169be831fb6105 [Accessed 26 Mar. 2026].

    • 1. What might you want to genetically engineer fungi to do and why? What are the advantages of doing synthetic biology in fungi as opposed to bacteria?

    For my project idea of developing an IBD treatment via psilocybin's effects on gut microbiota and serotonin signalling, a key challenge is confirming whether microbial reshaping is a direct effect of psilocybin (Fung et al., 2019),(Robinson, et al., 2023). I would therefore propose to start with the engineering of one psilocybin-producing fungal strain (Psilocybe cubensis) and genetically engineer variants that differ by only one feature, for example: one with active PsiH enzyme (IPsiH catalyses the 4-hydroxylation of tryptamine to 4-hydroxytryptamine, which is an essential and unique part of psilocybin biosynthesis that allows for the production of psilocin) (Huang et al., 2025), one with PsiH knocked out, one with tuned production levels (via promoter strength), and one producing an inactive analogue. These variants could be applied to ex vivo gut models to compare microbiome composition and immune markers, isolating whether psilocybin itself and not other mushroom components drives changes. Additionally, a reporter circuit with fluorescence under a gut condition-responsive promoter would track fungal activity timing and location. This genetic engineering approach would provide a controlled platform to test the hypothesis of psilocybin's gut therapeutic potential.

    References:

  • Dragos Ciocan and Eran Elinav (2023). Engineering bacteria to modulate host metabolism. Acta physiologica, 238(3). doi:https://doi.org/10.1111/apha.14001.
  • Fung, T.C., Vuong, H.E., Luna, C.D.G., Pronovost, G.N., Aleksandrova, A.A., Riley, N.G., Vavilina, A., McGinn, J., Rendon, T., Forrest, L.R. and Hsiao, E.Y. (2019). Intestinal serotonin and fluoxetine exposure modulate bacterial colonization in the gut. Nature Microbiology. doi:https://doi.org/10.1038/s41564-019-0540-4.
  • Gregory Ian Robinson, Li, D., Wang, B., Rahman, T., Gerasymchuk, M., Hudson, D., Kovalchuk, O. and Kovalchuk, I. (2023). Psilocybin and Eugenol Reduce Inflammation in Human 3D EpiIntestinal Tissue. Life, 13(12), pp.2345–2345. doi:https://doi.org/10.3390/life13122345.
  • Huang, Z., Yao, Y., Di, R., Zhang, J., Pan, Y. and Liu, G. (2025). De Novo Biosynthesis of Antidepressant Psilocybin in Escherichia coli. Microbial biotechnology, [online] 18(4), p.e70135. doi:https://doi.org/10.1111/1751-7915.70135.