Week 1 HW: Principles and Practices

Week 01 : Ethics and Principles

Ethics, safety, and security are key considerations throughout (and beyond!) this design process.
For Week 1, I use my project idea to think through how an engineered olfactory biosensor could be developed responsibly.

Class Assignment

1. Biological engineering application or tool

1. First, describe a biological engineering application or tool you want to develop and why. This could be inspired by an idea for your HTGAA class project and/or something for which you are already doing in your research, or something you are just curious about.

I am interested in engineering olfactory biosensors in E. coli that can detect specific odorant molecules and translate them into measurable outputs (fluorescence or an engineered scent). This connects directly to my background and upcoming internship in a sensory chemistry lab (ChemSenSim), where the focus is on understanding and simulating smell.

Motivation

• Scientific: Build sensors that “smell” defined odorants using genetic circuits (odorant receptor / odorant-binding module).
• Clinical: In the long term, such biosensors could contribute to standardized smell tests or detection of disease‑related compounds
• Ethical / practical: Bacterial sensors could reduce dependence on animals

Project trajectory (three aims)

• Aim 1 : Single‑odorant biosensor
  Design (and if feasible, assemble) a simple E. coli circuit that responds to one safe, well-characterized odorant with a reporter output (e.g. ATF1-based banana scent).

• Aim 2 : Combinatorial odor coding
  Extend the design to multiple odorants with distinct reporters, so combinations of odorants produce specific reporter patterns, ie the combinatorial coding of olfaction.

• Aim 3 : Disease‑linked sensing
  Conceptually explore biosensors that detect signatures associated with disease for potential clinical trials.

2. Governance / policy goals

2. Next, describe one or more governance/policy goals related to ensuring that this application or tool contributes to an “ethical” future, like ensuring non-malfeasance (preventing harm). Break big goals down into two or more specific sub-goals. Below is one example framework (developed in the context of synthetic genomics) you can choose to use or adapt, or you can develop your own. The example was developed to consider policy goals of ensuring safety and security, alongside other goals, like promoting constructive uses, but you could propose other goals for example, those relating to equity or autonomy.

For this project, I focus on three overarching governance goals, each split into sub‑goals:

2.1 Ensure biosafety of engineered olfactory E. coli

1. Prevent unintended environmental release of biosensor strains through appropriate strain choice, containment, and kill‑switches when needed.
2. Limit cross-contamination risks, specifcally repurposing sensors to monitor illicit substances or stigmatized health conditions without oversight.
3. Align with existing biosafety frameworks (NIH, WHO

2.2 Protect participants, lab workers, and the environment in odorant use

1. Use only non‑toxic, well‑characterized odorants with adapted toxicology knowledge.
2. Treat odorants and cultures with life‑cycle thinking: plan for safe waste treatment.
3. Educate researchers about odorant exposure risks.

2.3 Promote equitable, transparent, and clinically responsible use

1. Design population‑aware odor panels, recognizing that smell perception varies across populations
2. Protect olfactory privacy as HIPAA and RGPD Compliant sensitive health data requiring informed consent.
3. Communicate clearly with the public

3. Potential governance actions

3. Next, describe at least three different potential governance “actions” by considering the four aspects below (Purpose, Design, Assumptions, Risks of Failure & “Success”). Try to outline a mix of actions (e.g. a new requirement/rule, incentive, or technical strategy) pursued by different “actors” (e.g. academic researchers, companies, federal regulators, law enforcement, etc). Draw upon your existing knowledge and a little additional digging, and feel free to use analogies to other domains (e.g. 3D printing, drones, financial systems, etc.). Below, I describe three potential governance actions.

Option 1 : Odor‑Biosensor Biocontainment & Certification

Actors: Institutional Biosafety Committee, HTGAA teaching staff. 👌

• Purpose
  At present, many E. coli projects follow generic biosafety rules without specific guidance for whole‑cell biosensors that might interact with clinically relevant odorants or human samples.
  This action introduces a biosensor‑specific biocontainment and certification process, so that olfactory E. coli projects are reviewed with their particular risks in mind.

• Design

  • Define an approved strain and plasmid whitelist for teaching and exploratory biosensor work (non‑pathogenic E. coli K‑12 derivatives, minimized antibiotic resistance markers).
  • Require IBC registration for any constructs designed to respond to clinical VOCs or used with human-derived samples.
  • Couple certification with training modules specific to odor biosensing (handling volatile compounds, respiratory protection, odor exposure limits).

• Assumptions

  • Institutional committees have the time and expertise to evaluate biosensor projects.
  • Additional certification will not completely discourage student‑level innovation.
  • Genetic safeguards will be reliable enough for the scope of intended use.

• Risks of failure & “success”

  • Failure: The process becomes a bottleneck, discouraging exploratory student work or pushing experiments into unregulated DIY spaces.
  • Unintended success: Strong local certification may create an impression that biosensors are fully “safe” for clinical or commercial deployment before external regulators (e.g. FDA and ANSM) have weighed in.

Option 2 : In‑Silico Odorant Safety Screening & Standardized Panel

Actors: Sensory chemistry labs, toxicologists, researchers using odorants.

• Purpose
  Odorants for experiments are often chosen based on familiarity and convenience, with ad hoc toxicology checks. This action implements a computational pipeline and shared odorant panel to systematically select safe, representative stimuli for research and teaching.

• Design

  • Build a curated database of odorants with:
    • physico‑chemical properties,
    • perceptual descriptors,
    • toxicology and exposure limits, and
    • environmental fate where available.
  • Use machine learning and cheminformatics to propose panels of non‑toxic, diverse odorants suitable for lab use.
  • Publish standard operating procedures (SOPs) for preparation, serial dilutions, exposure protocols, and waste treatment.
  • Make the panel and SOPs open and versioned so that other labs can reuse and adapt them.

• Assumptions

  • Toxicology data are sufficiently complete for most common odorants.
  • A standardized panel will still leave room for exploratory work with new chemicals when justified.
  • Researchers will adopt shared tools when they clearly reduce preparation time and uncertainty.

• Risks of failure & “success”

  • Failure: Databases are incomplete or biased, leading either to false reassurance or unnecessary exclusion of promising odorants.
  • Unintended success: The standardized panel becomes a de facto global standard and unintentionally narrows the chemical search space, possibly overlooking disease‑relevant VOCs.

Option 3 : Clinical Framework for Olfactory Biosensor

Actors: Clinical collaborators, hospital ethics committees, patient organizations, PIs.

• Purpose
  Smell‑based biosensors can reveal sensitive health information (e.g., early signs of neurodegenerative disease) and might, in principle, detect behavioral or lifestyle markers. This action develops an ethical and clinical framework so that future translational work is guided.

• Design

  • Draft template informed‑consent language that explicitly addresses:
    • what information smell tests may reveal,
    • who will see the data, and
    • whether data may be reused or combined with other health records.
  • Require that any human‑facing pilot study with olfactory biosensors undergo review, regardless of perceived minimal risk.
  • Work with communications teams to create public‑facing explanations (webpages, leaflets) about how bacterial smell tests work, what they cannot infer, and which safeguards exist.

• Assumptions

  • Smell‑based biomarkers will be strong enough to raise privacy risks in at least some conditions.
  • Patients and the public will engage constructively if information is clear and honest.

• Risks of failure & “success”

  • Failure: An overly cautious framework slows down useful clinical validation or drives work to jurisdictions with weaker protections.
  • Unintended success: Heavy public emphasis on risks may stigmatize odor‑based diagnostics, making recruitment difficult even for low‑risk, high‑benefit applications.

4. Scoring governance actions

4. Next, score (from 1-3 with, 1 as the best, or n/a) each of your governance actions against your rubric of policy goals. The following is one framework but feel free to make your own:

Using the suggested rubric (1 = best / most effective, 3 = least effective, n/a = not applicable):

(1 = most effective / favorable; 3 = least; n/a = not applicable)

5. Prioritized governance options

5. Last, drawing upon this scoring, describe which governance option, or combination of options, you would prioritize, and why. Outline any trade-offs you considered as well as assumptions and uncertainties. For this, you can choose one or more relevant audiences for your recommendation, which could range from the very local (e.g. to MIT leadership or Cambridge Mayoral Office) to the national (e.g. to President Biden or the head of a Federal Agency) to the international (e.g. to the United Nations Office of the Secretary-General, or the leadership of a multinational firm or industry consortia). These could also be one of the “actor” groups in your matrix.

Audience: Potential clinical collaborators interested in olfactory diagnostics.

I would prioritize a combination of Option 1 and Option 2.

• Option 2 (Odorant Panel) is the fastest, lowest‑cost improvement. It directly supports lab safety and environmental protection.
• Option 1 (Certification) provides a clear institutional framework, ensures biosensor projects are back-controlled by safety committees.
• Option 3 (Ethics & Clinical Framework) becomes more important as the project moves from classroom experiments to clinical. Therefore, it might come useful on a longer time-frame.

Trade‑offs & uncertainties

• There is a tension between speed of innovation and depth of review: tighter certification and ethical review could slow creativity, especially for students, but may avoid larger problems later.
• Standardizing odorants improves safety and comparability but may restrict us from addressing some disease‑linked compounds.
• The predictive power of odor‑based biomarkers is still evolving despite governance.

Conclusion & Reflection

Working through this assignment made the project feel “real” to me. At first, teaching E. coli to “smell” was technically exciting to connect my interests in sensory chemistry and synthetic biology. Taking a step back and thinking about governance puts you into a hypothetical situation to see it as a potential diagnostic technology to interpret smell in people.

-> Several ethical concerns stood out this week:

• Olfactory privacy and identity. I had not considered directly that smell performance could become a new kind of biometric/health data. Once a bacterial sensor can distinguish disease states or behaviors, the line between a benign “smell test” and a sensitive diagnostic becomes very thin.

• Bias in what we choose to sense. Odor panels and diagnostic thresholds are not neutral. Choices about which odorants are “standard” could reflect particular cultures, and biosensors might reproduce those biases

• Replacing animals with engineered cells. I feel positively about reducing reliance on animals.

From these concerns, the governance actions that feel most realistic at this stage are: Treat smell-related data as health data by default, the same way data is handled throughout clinical trials in regard of HIPAA and RGPD Patient Protection.

References

might add those throughout the next courses !

Week 2 HW — DNA Read, Write, and Edit

Week 2 Homework: DNA Read, Write, and Edit

In preparation for Week 2’s lecture, I reviewed the lecture slides and papers.

Professor Jacobson’s Questions

1. Polymerase Error Rate and the Human Genome

What is the error rate of polymerase?

Nature’s error-correcting polymerase has an error rate of 1 in (10^6) (1 error per 1,000,000 bases).

How does this compare to the length of the human genome?

The human genome contains approximately 3*10^{9 base pairs (3 billion bp). If polymerase were the only fidelity mechanism, an error rate of 1:10}6 on a genome would result in approximately 3,000 errors every time a cell divides

How does biology deal with that discrepancy?

Biology employs Mismatch Repair systems to achieve additional error correction. Protein complex identifies and repairs mismatches that escape the polymerase’s initial proofreading, reducing the final error rate to approximately.

2. Coding Degeneracy for Human Proteins

How many different ways are there to code for an average human protein?

The average human protein is encoded by approximately 1,036 base pairs (≈345 amino acids) . Because the genetic code is redundant—with an average of roughly 3 synonymous codons per amino acid, the number of distinct DNA sequences capable of encoding the same protein sequence is approximately: 3^345

In practice, what are some reasons that all of these different codes don’t work?

While many DNA sequences encode the same amino acid sequence, they do not function equivalently in the cell due to properties of the mRNA itself

RNA Secondary Structure:
The mRNA sequence determines how the molecule folds. Certain sequences form tight secondary structures that physically block ribosome access and inhibit translation. The lecture slides showed us NUPACK simulations, showing alternative folding.

Thus, sequence matters beyond amino acid identity. The mRNA must be optimized

Dr. LeProust’s Questions

3. Current Oligo Synthesis Methods

What’s the most commonly used method for oligo synthesis currently?

The most commonly used method is solid-phase phosphoramidite chemistry. This cyclic process involves four steps repeated for each nucleotide addition:

1. Coupling — adding the next phosphoramidite-protected nucleotide
2. Capping — blocking unreacted 5’-OH groups to prevent truncation products
3. Oxidation — converting the phosphite triester to a stable phosphate triester
4. Deblocking — removing the 5’-protecting group (typically DMT) to prepare for the next cycle

4. Length Limitations in Direct Synthesis

Why is it difficult to make oligos longer than 200nt via direct synthesis?

The difficulty arises from the accumulation of truncation products and synthesis errors over many cycles . Each coupling step has a typical efficiency of 98-99.5%. For a 200-nucleotide oligo: 16%

As length increases, the proportion of full-length product drops exponentially, while incomplete sequences (n-1, n-2, etc.) accumulate. The slides illustrated this with chromatograms showing significant impurity peaks for 500-nucleotide synthesis attempts .

Additionally, depurination and other side reactions become more probable with longer sequences, further reducing fidelity.

5. Gene Assembly vs. Direct Synthesis

Why can’t you make a 2000bp gene via direct oligo synthesis?

Direct phosphoramidite synthesis is ~200-500 nucleotides due to error accumulation and yield collapse . To construct a 2000bp gene, we ca, use gene assembly methods:

1. Synthesize many shorter oligonucleotides (typically 40-200mers)
2. Design overlapping sequences between adjacent oligos
3. Assemble via PCR (polymerase chain reaction) or enzymatic assembly (e.g., Gibson assembly, Golden Gate)

This modular approach allows the cumulative error rate to be managed across multiple synthesis reactions, with final assembly and error correction performed enzymatically .

George Church’s Question (Option 1)

6. Essential Amino Acids and the “Lysine Contingency”

What are the 10 essential amino acids in all animals?

The 10 amino acids considered essential for animals are :

1. Arginine (R)
2. Histidine (H)
3. Isoleucine (I)
4. Leucine (L)
5. Lysine (K)
6. Methionine (M)
7. Phenylalanine (F)
8. Threonine (T)
9. Tryptophan (W)
10. Valine (V)

How does this affect your view of the “Lysine Contingency”?

The “Lysine Contingency” from Jurassic Park proposed engineering dinosaurs unable to produce lysine, ensuring they would die without dietary supplements provided by the park. Understanding that lysine is an essential amino acid reveals this contingency as scientifically flawed for two reasons:

By definition, an “essential” amino acid is one that vertebrates already cannot synthesize. Animals naturally lack the lysine biosynthesis pathway (found in bacteria, plants, and fungi). Therefore:

• The dinosaurs would already be natural lysine auxotrophs without any genetic engineering
• “Removing” the ability to produce lysine is impossible—the pathway was never present in vertebrate genomes
• The genetic modification described in the film would be unnecessary or meaningless