Week 1 HW: Principles and Practices

1. Biological engineering application

Living in London, I was exposed to the reality of black mold after moving into my first flat. It is a recurring issue in damp housing, and the main accessible “solution” most tenants use is repeated cleaning with bleach or similar harsh chemicals, which is unsustainable, damaging to surfaces, and itself a health concern for people with respiratory conditions or sensitive skin. UK health and housing guidance now explicitly recognises damp and mold as a serious risk factor for asthma, respiratory infections and other illnesses, with the burden falling disproportionately on people in poorer quality or overcrowded accommodation.

For this homework, I am interested in developing a biological approach to this problem: an engineered host strain that produces enzymes capable of breaking down black mold. In more technical terms, I imagine using a non‑pathogenic microbial chassis, such as a safe bacterial or yeast strain, engineered to secrete mold‑degrading hydrolases like chitinases and glucanases that target key structural polysaccharides in fungal cell walls. These enzymes could disrupt the integrity of the mold, making it easier to remove and potentially reducing regrowth compared to purely cosmetic cleaning that leaves viable hyphae embedded in surfaces.

There are at least two deployment modes I would consider. One is a cell‑free product, where the engineered host is grown in a controlled facility and only the purified or partially purified enzyme cocktail is applied in homes, similar to an “enzymatic cleaner”. The other, higher‑risk mode would be a live “biocleaner” formulation where the engineered microbes themselves are applied to contaminated surfaces for a limited time to compete with and degrade mold before being inactivated. A key appeal of both approaches is the potential to design more targeted, biodegradable interventions that reduce reliance on broad‑spectrum chemical biocides and repeated repainting, while ideally offering better long‑term control of the underlying fungal growth.

2. Governance policy goals

Because this project directly interacts with people’s homes, health, and indoor environments, I think it is essential to define clear governance policy goals before imagining its commercial use. My first goal is to ensure safety and security, by preventing uncontrolled spread of engineered strains, avoiding harmful effects on non‑target fungi or indoor microbiomes, and reducing the risk that these enzymes or strains could be misused (for example, to damage infrastructure or interfere with beneficial fungal processes). This also includes enabling rapid detection and response if something goes wrong, such as unexpected allergic reactions, ecological impacts, or accidental release beyond the intended setting.

A second goal is to promote beneficial uses of this technology. This means applications that genuinely improve health and housing quality: reducing respiratory illness associated with damp and mould, decreasing the need for repeated bleach use, and making remediation more effective and longer‑lasting. Governance should steer development towards evidence‑based, clinically and environmentally evaluated products, and away from purely cosmetic or marketing‑driven uses that add risk without real benefit.

A third goal is to promote equity and autonomy. Damp and mould problems in the UK are more common in poorer quality and overcrowded housing, and tenants often have limited power to demand remediation. Any use of engineered strains or enzymes in housing should not turn low‑income tenants into test subjects, and affected residents should have clear information and the ability to consent or opt out of experimental uses. Governance should also try to ensure that if the technology proves safe and effective, it is accessible to those most affected by unhealthy housing, not just to landlords or homeowners who can pay a premium.

Finally, another important goal is managing environmental impact. Policies should aim to minimise long‑term persistence of engineered microbes outside controlled settings and protect beneficial fungi in soils, building materials, and surrounding ecosystems. On the research side, governance should not unnecessarily block basic and applied research on mould-degrading enzymes, but instead channel it into safer designs and transparent evaluation, so that the technology can develop responsibly over time.

3. Governance actions

Option 1 – Prioritise cell‑free enzyme products over live strains

Purpose
The current situation relies mainly on bleach and repainting, which can be harmful to occupants and often fails to remove underlying mould growth. This option shifts deployment away from live engineered microbes in homes and towards cell‑free enzyme formulations: engineered strains are only used in contained facilities, and residents receive products that contain purified or stabilised mould‑degrading enzymes but no living GMOs.

Design

• Regulators classify live black‑mould‑targeting strains as “contained use only” and restrict consumer‑facing products to cell‑free preparations.
• Manufacturers must demonstrate that engineered hosts are not present in the final product and that enzymes degrade safely after use.
• Guidance is developed for safe application and disposal (for example, avoiding aerosolisation in ways that could trigger allergies).

Key actors include national or regional health and environmental regulators, institutional biosafety committees, and companies producing cleaning products.

Assumptions

• Cell‑free enzymes can be produced at scale and remain active long enough on surfaces to be effective.
• Removing live strains from homes meaningfully reduces ecological and biosecurity risk compared with live “biocleaner” formulations.
• Regulators can reliably verify absence of viable engineered organisms in products.

Risks of failures and of “success”

• If enzymes are unstable or hard to formulate, products may be expensive, ineffective, or require additives with their own health impacts.
• A perceived “green” label for enzymatic cleaners could encourage overuse without proper ventilation or PPE, potentially increasing allergen exposure.
• If this model is too restrictive, it may push DIY or black‑market sharing of live strains without any safeguards.

Option 2 – Licensing and standards for professional live‑strain use

Purpose
This option accepts that, in some cases, live engineered strains might be more effective at penetrating porous materials and competing with mould, but tries to confine such use to trained professionals operating under strict standards. It aims to improve remediation outcomes while keeping higher‑risk interventions out of casual consumer hands.

Design

• Only licensed remediation companies can purchase and deploy live mould‑degrading strains.
• Licences require staff training, use of engineered safety features (for example, auxotrophy, kill switches), documentation of where and when strains are applied, and post‑treatment monitoring.
• Inspection regimes and reporting requirements are established, with penalties for misuse or failure to follow protocols.

Actors include regulators, professional bodies, insurers, and employers in the mould‑remediation industry.

Assumptions

• Professional users will comply with licensing conditions and keep good records.
• Built‑in genetic safeguards and application protocols can keep strains from persisting or spreading beyond treated areas.
• Regulators have enough capacity to issue licences, audit companies, and respond to incidents.

Risks of failures and of “success”

• Safety features could fail in real‑world conditions (for example, evolution of escape mutants, unexpected environmental niches).
• Smaller landlords or unlicensed contractors might copy or source strains informally, bypassing controls.
• If this approach is seen as highly effective but expensive, it could widen inequalities between people whose housing is thoroughly remediated and those left with unsafe conditions.

Option 3 – Community‑engaged pilots with local consent and oversight

Purpose
Because damp and mould disproportionately affect low‑income and marginalised tenants, this option focuses on equity and autonomy. It aims to ensure that early field trials or deployments happen only with meaningful community involvement and consent, rather than quietly using vulnerable residents as test cases.

Design

• Any pilot use of engineered strains or novel enzyme formulations in residential buildings requires:
  • Transparent, accessible information about risks, benefits, and alternatives.
  • Collective or individual consent from residents, depending on context.
  • A local oversight body including tenants, public‑health officials, independent scientists, and tenant‑rights organisations.
• Pilots must include mechanisms for feedback, complaint, and withdrawal, plus public reporting of outcomes (including negative results).

Key actors are city or municipal health departments, housing authorities, tenant associations, NGOs, and academic or industrial partners.

Assumptions

• Residents will have the time, resources, and trust to participate in decision‑making.
• It is possible to design consent processes that are not coercive, even when people feel stuck in unhealthy housing.
• Local oversight bodies can be set up without excessive delay or capture by particular interests.

Risks of failures and of “success”

• Engagement could become tokenistic, where decisions are effectively made in advance and consultation is performative, undermining trust.
• The additional process overhead might slow deployment of helpful technologies in situations where mould is already causing serious harm.
• Successful pilots in a few communities might be used rhetorically to justify broader roll‑out without reproducing the same level of engagement elsewhere.

4. Scoring table (policy goals vs options)

5. Preferred governance option(s) and trade‑offs

Looking across my scores, I would prioritise a combination of Option 1 (cell‑free enzyme products) and Option 3 (community‑engaged pilots) as my main recommendation. Option 1 performs best on biosecurity and environmental protection because engineered strains stay in contained facilities and only enzymes reach people’s homes, which reduces the risk of uncontrolled spread or long‑term persistence of GMOs in indoor environments. Option 3 performs best on equity and autonomy by requiring clear information, consent, and local oversight before any experimental uses in residential housing, which is crucial given that damp and mould problems disproportionately affect low‑income and marginalised tenants.

6. Weekly reflection on ethics

Several ethical concerns from the Week 1 material feel directly relevant to this project. The gene‑drive example highlighted how powerful self‑propagating systems can escape local control and have irreversible ecological impacts, which is a useful warning for any design that might persist or spread beyond intended contexts. The “virus hunting” discussion raised issues around dual use, benefit‑sharing, and unequal distribution of risks and benefits when work is done in or on communities that already face health and economic disadvantages. Both examples emphasise the importance of thinking not just about whether we can build a technology, but whether, where, and under what conditions we should.

Week 2 HW: DNA Read, Write and Edit

Part 1: Benchling & In-silico Gel Art

• Benchling: 48,502 bp Lambda phage DNA.
  Open Project
• Enzymes: NdeI, PvuII, SacI only.
  

Part 3: Protein Design - Human Lysozyme (hLYZ) (no Part 2 because no lab attendance yet)

UniProt: P61626
Why: Lyses Gram+ bacteria (clearing zones on plates); ~130 aa antimicrobial for food/biofilms.

Codon-opt for E. coli (I used VectorBuilder: :
ATGAAAGCGCTGATTGTGCTGGGCCTGGTGCTGCTGAGCGTGACCGTGCAGGGCAAAGTGTTTGAACGCTGTGAACTGGCCCGTACCCTGAAACGTCTGGGCATGGATGGCTATCGCGGCATTAGCCTGGCGAACTGGATGTGCCTGGCGAAATGGGAAAGCGGATATAACACCCGCGCGACCAACTATAACGCAGGCGATCGTAGCACCGATTATGGCATTTTCCAGATTAACAGCCGTTATTGGTGCAATGATGGCAAAACCCCGGGCGCCGTGAACGCGTGCCATCTGAGCTGTAGCGCCCTGCTGCAGGATAACATTGCGGATGCCGTGGCCTGCGCGAAACGCGTGGTGCGCGATCCGCAGGGCATTCGCGCGTGGGTGGCGTGGCGCAACCGCTGCCAGAACCGCGATGTTCGCCAGTACGTGCAGGGCTGTGGCGTG

Why codon optimized?:

DNA’s genetic code is degenerate: 64 codons encode 20 amino acids, but cells prefer “optimal” codons with abundant tRNAs. Human lysozyme (P61626) uses eukaryotic codons (e.g., AGA/CGA/CGG/CGC for Arg), rare in bacteria → ribosomal pausing → low protein (~1-10% max yield).

Optimization process:

• Scan AA seq → Replace rare codons (e.g., AGA Arg → CGT/CGC).
• GC ~50% (E. coli sweet spot).
• Avoid RE sites/repeats.

Why E. coli?

K12/BL21 canonical:

• 20min doubles.
• T7/plasmids stocked.
• Lysis halos assay.
• BSL1/$1L.

Protein Production Technologies

To produce lysozyme from my codon-opt DNA (plasmid: T7 promoter + RBS + hLYZ + term + AmpR), two main paths: cell-dependent (in vivo) and cell-free (TXTL)—both HTGAA staples.

1. Cell-Dependent (E. coli BL21(DE3))

Tech: Chemical transformation → IPTG induction → lysis/purification.

Steps:

1. Transform plasmid into competent BL21(DE3) (T7 RNA pol strain).
2. LB+Amp plate (37°C overnight) → single colonies.
3. Liquid culture → mid-log (OD600~0.6) → add 1mM IPTG (T7 binds promoter).
4. Express 4h (A280 monitor) → centrifuge → lysozyme buffer lysis → Ni-NTA purify (His-tag).

Why: Scalable (L-scale), cheap, natural folding/chaperones. Yield: 50-200 mg/L. Assay: Lysis halos on Micrococcus luteus agar. Limit: Inclusion bodies if misfold.

2. Cell-Free (TXTL / PURExpress)

Tech: NEB PURExpress or NEB TXTL kit (plasmid-fed).

Steps:

1. Mix: DNA template (10-100 ng), lysate/lysate-free enzymes, NTPs/aa-tRNAs, energy (PEP), Mg/ATP.
2. 29-37°C incubate 4-16h (plate reader monitor if fluorescent tag).
3. Direct assay (no purification): Add bacteria → halo formation.

Why: Rapid prototyping (hours), no cloning/transformation, toxic proteins OK. Yield: 1-10 μM (~0.1-1 mg/mL). HTGAA fave for Week 2 demos. Limit: Costly, short-lived.

Transcription/Translation (Central Dogma): DNA → transcription (RNA pol binds promoter → mRNA w/ RBS/5’UTR; ~100 nt/s).
mRNA → translation (30S ribosome binds RBS → 70S → tRNAs decode codons → peptide chain ~20 aa/s → release factor → fold).

Lysozyme: ~130 aa → active in <1 min post-Tx. Cell-free skips replication/folding issues!

Part 4 - STILL EDITING

Part 5.1 DNA Read

(i) What DNA would I sequence & why?

Synthetic DNA data storage archive (e.g., Microsoft/ETH Zurich 1PB encoded in 13.3 zettabytes of oligo pools).
Why: DNA stores 1 exabyte/gram (10^18 density, 1000+ year stability at 4°C vs HDD 5yrs/90% capacity loss). Perfect for AI training data, climate models, genomic libraries—HTGAA synbio extension (store lysozyme designs, node protocols). Error-correcting codes (FEC) handle 1-5% synthesis/seq errors. Applications: Space (NASA), biobanks, “DNA cloud”.

(ii) Technology: Oxford Nanopore MinION + PromethION (3rd-generation)

Why MinION: Long reads (>10kb) assemble repetitive oligo pools; portable (USB, London node); real-time basecalling (Guppy 6.0, 98% raw→99.9% consensus). PromethION scales to 290Gb/run for PB archive.

Generation: 3rd—nanopore sequencing (single-molecule, no amplification bias, real-time vs SBS parallel/PCR).

Input/Preparation (100ng oligos):

1. End-repair/A-tail: Blunt→A-overhangs (NEB Ultra II).
2. Adapter ligation: SQK-LSK114 (motor protein + tether). No fragmentation (native length preserves codes).
3. Loading: 30-48 flowcells, R9.4.1 pore.

Essential Steps & Base Calling:

1. Threading: Helicase ratchets ssDNA through protein nanopore (α-hemolysin mutant).
2. Ionic current: 4 bases uniquely modulate ~100pA current (A=high, C=mid, G=low, T=mid-low). Homopolymers ~3-10nt compress signal.
3. Raw signal → Guppy neural net (RNN+Attention): Translates squiggles → FAST5 → FASTQ (Q10=99.9%). Dorado (GPU) live.
4. Polish: Medaka/RACON consensus from 30x coverage.

Output: FASTQ reads (10-100kb), PHRED scores. Assemble → ECC decode → binary data.

Limits: Homopolymers (5% err), speed (450bases/s/pore), but duplex mode (both strands) → Q20.

Part 5.2 DNA Write

(i) What DNA to synthesize & why?

Codon-optimized human lysozyme (hLYZ, P61626) + Gibson homology arms for pET28a integration.
Why: Antimicrobial for food preservation (Gram+ lysis), biofilms, therapeutics. Twist-free (~$40), test lysis Week 4. Applications: Self-sterilizing surfaces, phage replacement.

(ii) Technology: Twist Bioscience (arrayed phosphoramidite synthesis → assembly)

Why: $0.09/bp, 1-5kb genes error-free, 10-14 day turnaround, HTGAA partner (Benchling→Twist integration).

Essential Steps:

1. Oligo array synthesis: Inkjet print A/C/G/T phosphoramidites → 10^6 spots/plate → 200nt oligos (60mer tiles).
2. Error correction: 4x coverage unique tiles → NGS validate → discard errors.
3. Hierarchical assembly: Gibson (chemo-enzymatic overlap), NEBuilder HiFi.
4. Cloning/Verification: Electrocompetent DH5α → colony PCR/Sanger → ship 2-5μg plasmid.

Limitations:

• Speed: 2wks (vs enzymatic 2days).
• Accuracy: 99.9% post-assembly (raw oligos 1:100 err).
• Scalability: 5kb genes ($450), 100kb pathways $$$; toxic/repeats fail.

Part 5.3 DNA Edit

(i) What DNA to edit & why?

E. coli MG1655 genome—insert hLYZ cassette (constitutive pro, RBS, hLYZ, term) at neutral locus (betT/aga).
Why: Chromosomal antibiotics eliminate plasmid loss/stability issues. Industrial: Self-lysing probiotics for food safety. Conservation: Edit coral bacteria to fight bleaching pathogens.

Edit: 1.2kb insertion replacing pseudogene.

(ii) Technology: CRISPR-Cas12a recombineering (enAsCas12a)

Why: TTN PAM (AT-rich E.coli), dual crRNAs multiplex, 90% HR efficiency w/ RecET. Better than Cas9 (NG PAM-limited).

How it edits: Cas12a crRNA binds target → RuvC cleaves → RecET boosts HDR from ssDNA donor.

Preparation/Steps:

1. Design (Benchling): 22nt spacer + TTN PAM, 500-800nt ssDNA donor (40nt homology arms).
2. Inputs: enAsCas12a protein (20μM), crRNA/tracrRNA (1μM), ssDNA (1μM), BL21 RecET cells.
3. Electroporation: 25μF, mix → recover 37°C → plate Amp + lysis assay.
4. Screen: PCR + Sanger verify insertion.

Limitations:

• Efficiency: 10-50% (donor design critical).
• Precision: 0.1% off-target (Cas12a collateral nuclease quenched).
• Size: <3kb insertions optimal.

My spacer: GTTGATCTGGAAGCTGACCGC (betT locus).