Week 7 HW: Genetic Circuits Part II: Neuromorphic Circuits

Q1) What advantages do IANNs have over traditional genetic circuits, whose input/output behaviors are Boolean functions?

• IANNs can be rearranged, remodelled, or changed into a new form, layout, or function after their initial creation in comparison to boolean genetic circuits which are rigid/fixed, and its reconfiguring ability means it can be used for analytical devices integrating biological recognition elements (enzymes, microorganisms, antibodies) with transducers to detect bioavailable pollutants (microplastics, heavy metals) to check biotoxicity
• Less metabolic stress in cells with IANNs compared to boolean genetic circuits
• IANNs can do Complex pattern recognition and analog computation - IANNs can look at many variables like temperature, Wind Direction, Traffic Speed and Humidity and the logic derived from this is relationships and biological signatures for example, it is context aware and gives a probability based result (risk in percentage or category).

This is the opposite for traditional genetic circuits, the same processing step is a conditional if/then statement, as Boolean circuits are limited to 2^n states. For example, if a city sensor receives an input that is “partly true” (e.g., moderate congestion but high humidity), a Boolean circuit may fail to trigger / provide an incorrect binary output.

• In a Boolean circuit, every input typically has equal weighting to flip a gate comparatively to IANNs which allow for synaptic weighting. You can tune the promoter strengths / molecular affinities so that a “wind speed” input has a 70% influence on the output, while “humidity” only has 30%. This is critical for the “smart city” use case where variables have different levels of importance.
• IANNs can cluster many signals before reaching an activation threshold, so IANNs are natural filters and can filter out noise which trigger false positives in traditional genetic circuits. They require a specific consensus of molecular inputs to fire.

Q2) Describe a useful application for an IANN; include a detailed description of input/output behavior, as well as any limitations an IANN might face to achieve your goal.

Chemical biosensor

• Variable 1 - concentration of nitrogen dioxide can be a primary input, binding to specific receptors.
• Variable 2 (VOCs): Volatile Organic Compounds (like benzene from car exhaust) act as a secondary input.
• Variable 3 (Temperature): Temperature-sensitive proteins act as a “contextual weight.”
• The IANN “recognizes” a pattern. For example: [High nitrogen dioxide] + [High VOCs] + [High Temperature] indicates a stagnant smog event. Because the signals are analog, the network can distinguish between a “passing truck” (short spike) and a sustained pattern.
• Output: Proportional response: The output is not a binary “On/Off” but a graduated biological response proportional to the severity of the recognized pattern.
• Remediation (The action): The organism secretes a specific enzyme (like nitroreductase) to break down nitrogen dioxide into non harmful byproducts. In the IANN, if the pollution pattern is “moderate,” the plant produces a moderate amount of enzyme. If it is critical, it triggers a maximum secretion.
• The signal output - The system produces a Green Fluorescent Protein (GFP). The intensity of the “glow” represents the recognized pollution level, which can be mapped by city drones or satellite imaging to update a City Digital Twin in real-time.

Limitations

• Metabolic stress, running a neural network inside a cell is energy intensive, so the organism secreting the enzyme for breaking down chemicals may grow slower or die untimely compared to natural counterparts.
• IANNs tuned in laboratory settings under controlled conditions in comparison to a Smart City, a Bio-Sentinel might face natural disasters or physical damage. These outside stresses may modify the weights of the neural network unexpectedly, causing the system to lose its calibration.

Q3) Draw a diagram for an intracellular multilayer perceptron where layer 1 outputs an endoribonuclease that regulates a fluorescent protein output in layer 2.

INPUT X1                          INPUT X2
    (DNA: pCsy4)                      (DNA: pOutput)
         |                                 |
         | [Tx]                            | [Tx]
         v                                 v
      mRNA_Csy4                        mRNA_Output
         |                          (contains Csy4 site)
         | [Tl]                            |
         v                                 |
     [ Csy4 ]----------------------------> X (Cleavage/Inhibition)
    (Protease)                             |
                                           |
                                           v
                                     [ NO PROTEIN ]
                                           or
                                    [ FP OUTPUT ] 
                                   (if X1 is absent)

How does the above diagram work?

In a neural network context, this biological setup functions as an NOT gate or a weighted input where X1 negatively regulates X2.

Transcription (Tx) & Translation (Tl): These represent the biological computation steps.

How does it work: The mRNA for the fluorescent protein (X2) contains a specific hairpin sequence. When the Csy4 protein (X1) is present, it recognizes and cleaves that hairpin.

Output: Cleavage usually destabilizes the mRNA or separates the ribosome-binding site from the coding sequence, effectively turning off the output.

Logic Table

Assignment Part 2: Fungal Materials

Q1) What are some examples of existing fungal materials and what are they used for? What are their advantages and disadvantages over traditional counterparts?

• Mycelium for leather jackets, bags, coats - advantages are that they are renewable, biodegradable and sustainable. Disadvantages are that they are expensive, and production process is not scalable, not as durable (reported by some users) and easily repairable as manual leather, mycelium cultivation if not done with renewable energy can be environmentally costly.
• Mycelium and enzymes like laccases and peroxidases can degrade, sequester, or detoxify environmental toxins like heavy metals, plastics, oil and dyes (Dinakarkumar et al., 2024).

Q2) 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?

I’d like to engineer fungi for medicinal purposes: fungi are naturally produce secondary metabolites (like penicillin). Engineering these allows us to produce high-value drugs like psilocybin (for depression) or cannabinoids (for pain) more sustainably than traditional farming (Keller, 2019). Fungi are eukaryotic comparatively to bacteria, and can do complex protein folding for stability as well as do post translational modifications (glycolysation, methylation), can secrete extracellular fluid outside cells quickly, can metabolise complex chemicals comparatively to bacteria which can break down simple chemicals and can form complex networks of filaments comparative to bacteria which are unicellular.

Assignment Part 3: First DNA Twist Order

Insert: T7 Phage gp17 Receptor Binding Domain (RBD).

Coordinates: UniProt P03748, Residues 371–553. C-terminal receptor-binding region (residues 371–553), including the distal tip domain (466–553) containing specificity-determining loops.

Logic: This domain is responsible for host recognition. In my project, this is the “Variable” component that I am optimizing with ProteinMPNN to target P. aeruginosa.

1 AA Sequence for my Insert I will use residues 371-553 from UniProt entry , https://www.uniprot.org/uniprotkb/P03748/entry#sequences. This is the portion of the protein which actually chooses which bacteria to kill.

GHVLQLESASDKAHYILSKDGNRNNWYIGRGSDNNNDCTFHSYVHGTTLTLKQDYAVVNKHFHVGQAVVATDGNIQGTKWGGKWLDAYLRDSFVAKSKAWTQVWSGSAGGGVSVTVSQDLRFRNIWIKCANNSWNFFRTGPDGIYFIASDGGWLRFQIHSNGLGFKNIADSRSVPNAIMVENE

2 Reverse translation from protein to DNA sequence

Reverse Translate results
Results for 183 residue sequence "Untitled" starting "GHVLQLESAS"

>reverse translation of Untitled to a 549 base sequence of most likely codons.
ggccatgtgctgcagctggaaagcgcgagcgataaagcgcattatattctgagcaaagat
ggcaaccgcaacaactggtatattggccgcggcagcgataacaacaacgattgcaccttt
catagctatgtgcatggcaccaccctgaccctgaaacaggattatgcggtggtgaacaaa
cattttcatgtgggccaggcggtggtggcgaccgatggcaacattcagggcaccaaatgg
ggcggcaaatggctggatgcgtatctgcgcgatagctttgtggcgaaaagcaaagcgtgg
acccaggtgtggagcggcagcgcgggcggcggcgtgagcgtgaccgtgagccaggatctg
cgctttcgcaacatttggattaaatgcgcgaacaacagctggaacttttttcgcaccggc
ccggatggcatttattttattgcgagcgatggcggctggctgcgctttcagattcatagc
aacggcctgggctttaaaaacattgcggatagccgcagcgtgccgaacgcgattatggtg
gaaaacgaa

3 Codon optimised sequence

GGCCACGTGCTGCAGCTGGAAAGCGCGAGCGATAAAGCGCATTATATTCTGAGCAAAGATGGCAATCGTAATAACTGGTACATCGGCCGCGGCAGCGATAATAACAATGATTGCACCTTCCATAGCTACGTGCACGGCACCACCCTGACCCTGAAACAGGATTATGCGGTGGTGAACAAACATTTCCACGTGGGCCAGGCAGTGGTCGCGACCGATGGCAACATTCAGGGCACCAAATGGGGCGGCAAATGGCTGGATGCGTATCTGCGCGATAGCTTTGTGGCGAAAAGCAAAGCCTGGACCCAGGTGTGGAGCGGCAGCGCCGGCGGCGGCGTGTCCGTGACCGTGAGCCAGGATCTGCGCTTTCGCAATATCTGGATTAAATGCGCGAATAATAGCTGGAACTTCTTCCGCACCGGCCCGGATGGCATTTACTTTATTGCCAGCGATGGCGGTTGGCTGCGCTTTCAGATTCACTCGAACGGCCTGGGCTTCAAAAACATTGCCGATAGCCGCAGCGTGCCGAACGCGATTATGGTGGAAAACGAA

4

For phage protein expression, I will use the pET-28a(+) vector. I chose this vector because:

• It has a robust T7 promoter for high-yield protein production.

• It includes a Kanamycin resistance marker (standard for selection).

• It features an N-terminal His-tag, which is essential for the proteomic validation.

Sequence insert - annotated link: https://benchling.com/s/seq-LYTiB6CS3kC0MuzSbY4s?m=slm-w9Gc9uoWLYMnC4Q5CcGe

Plasmid construct: https://benchling.com/s/seq-n18sRhXq6Yz6uol2epHg?m=slm-W6czjeckW1A2aaynxjBC

Genetic circuits prelab:

References

Dinakarkumar, Y., Ramakrishnan, G., Gujjula, K. R., Vasu, V., Balamurugan, P., & Murali, G. (2024). Fungal bioremediation: An overview of the mechanisms, applications and future perspectives. Environmental Chemistry and Ecotoxicology, 6, 293–302. https://doi.org/10.1016/j.enceco.2024.07.002

Keller, N. P. (2019). Fungal secondary metabolism: Regulation, function and drug discovery. Nature Reviews Microbiology, 17(3), 167–180. https://doi.org/10.1038/s41579-018-0121-1