Week 1 HW: Principles and Practices

• Biological engineering application or tool

  I am interested in developing a low-cost wearable biomedical monitoring device designed for early detection of health deterioration in elderly or vulnerable populations, particularly in low-resource or rural settings. The device would integrate non-invasive biosensors (such as heart rate, oxygen saturation, and motion detection) with basic data processing to identify abnormal patterns that could indicate falls, cardiovascular events, or sudden health decline.

  This application is motivated by the growing aging population and the lack of continuous medical monitoring outside hospital environments. Many preventable health emergencies become severe simply because they are detected too late. By enabling early alerts and remote monitoring, this tool could reduce harm, improve response times, and support caregivers and healthcare systems without replacing human medical judgment.

• Governance/policy goals for an ethical future

  Main ethical goal:

  Ensure that the development and deployment of wearable biomedical monitoring technologies minimize harm while promoting equitable, safe, and responsible use.

  Sub-goals:

  Non-malfeasance and safety

  Prevent physical harm caused by device malfunction, inaccurate data, or false alarms.

  Reduce the risk of over-reliance on automated systems in medical decision-making.

  Data protection and autonomy

  Protect sensitive health data from misuse, breaches, or unauthorized surveillance.

  Ensure users understand what data is collected and retain meaningful consent.

  Equity and accessibility

  Avoid creating technologies that only benefit wealthy populations.

  Promote affordability and usability for non-technical users.

• Governance actions

  Option 1: Mandatory baseline safety and accuracy standards for wearable biomedical devices

  One possible governance action is the implementation of mandatory baseline safety and accuracy standards for wearable biomedical monitoring devices. Currently, many low-cost or early-stage health wearables are developed and deployed with limited validation, which can increase the risk of malfunction or misleading data. This option proposes that before such devices are widely used, they meet minimum performance and safety requirements. These standards would be designed and enforced by national health regulators or medical device authorities, requiring developers to conduct basic testing and clearly communicate the limitations of their systems. This approach assumes that regulatory bodies have the capacity to evaluate these technologies efficiently and that developers can comply without excessive burden. However, there is a risk that overly strict regulation could slow innovation or discourage small research teams. Even if successful, this approach may lead some developers to focus only on meeting minimum requirements rather than pursuing higher safety standards.

  Option 2: Privacy-by-design requirements in biomedical technologies

  A second governance action focuses on integrating privacy protections directly into the design of biomedical monitoring technologies. Rather than treating data privacy as an afterthought, this approach encourages researchers and developers to adopt privacy-by-design principles from the earliest stages of development. This could be promoted through university research guidelines, institutional review boards, and funding agency requirements, emphasizing practices such as data minimization, local data processing, and secure encryption. This option assumes that engineering teams have sufficient knowledge of data ethics and that privacy-preserving designs do not significantly compromise device functionality. Potential risks include increased development complexity and costs. Additionally, if implemented successfully, there is a risk that users may develop a false sense of complete security and underestimate residual privacy vulnerabilities.

  Option 3: Incentives for equitable and responsible deployment

  A third governance action involves creating incentives to promote equitable deployment and socially responsible use of biomedical monitoring devices. Instead of relying solely on regulation, this approach encourages positive behavior by offering grants, academic recognition, or public procurement preferences to projects that prioritize accessibility, affordability, and use in underserved communities. Universities, public health agencies, and non-governmental organizations would play key roles in implementing these incentives. This option assumes that incentives can meaningfully influence design and deployment decisions and that equity-related outcomes can be effectively measured. A potential risk is that such incentives may become symbolic rather than impactful. Even in cases of success, technologies may be deployed in vulnerable communities without sufficient local infrastructure or long-term support, limiting their real-world effectiveness.

5. Prioritized governance approach

Based on this analysis, I would prioritize a combination of Option 2 (privacy-by-design) and Option 3 (equity-focused incentives). Together, these approaches promote ethical responsibility without significantly slowing innovation or increasing regulatory burden.

Option 1 remains important for safety-critical applications but should be applied proportionally to avoid discouraging early-stage research. The main trade-off considered is between innovation speed and risk reduction, acknowledging uncertainty in how users and institutions adopt new technologies.

Intended audience:

This recommendation is directed toward university leadership, research institutions, and public funding agencies, as they are well-positioned to shape early design choices and norms.

6. Ethical reflection

One ethical concern that became particularly evident during this week’s class is the risk of normalizing continuous health surveillance through biomedical monitoring technologies. While these tools are designed to improve safety and enable early detection of medical issues, they may also unintentionally reduce personal autonomy or create pressure for individuals to share sensitive health data constantly. This raised new questions for me about where the boundary lies between beneficial monitoring and intrusive oversight. To address these concerns, appropriate governance actions should include clear limitations on data collection and retention, transparent and user-centered consent processes, and stronger ethical training for engineers developing biomedical technologies. This reflection reinforced the idea that ethical biological engineering is not only about preventing direct harm, but also about protecting dignity, autonomy, and long-term trust between patients, engineers, and healthcare systems.

Homework Questions – Professor Jacobson

1. What is the error rate of DNA polymerase? How does this compare to the length of the human genome, and how does biology deal with this discrepancy?

DNA polymerase has an intrinsic error rate of approximately 1 mistake per 10⁷ nucleotides during replication. With proofreading activity, this improves to about 1 error per 10⁹–10¹⁰ nucleotides. The human genome is roughly 3 × 10⁹ base pairs long, which means that without correction mechanisms, each round of DNA replication would introduce hundreds of errors. Biology addresses this discrepancy through multiple layers of error control, including polymerase proofreading and post-replication DNA repair systems such as mismatch repair. Together, these mechanisms reduce the effective mutation rate to a level compatible with stable inheritance and organismal survival.

2. How many different ways are there to code for an average human protein? Why don’t all of these codes work in practice?

An average human protein of about 300 amino acids could theoretically be encoded by an enormous number of different DNA sequences due to the degeneracy of the genetic code (multiple codons per amino acid). In theory, this number is astronomical (on the order of 10¹⁴⁰ possible sequences). In practice, however, many of these sequences do not function well because of factors such as codon bias, mRNA secondary structure, GC content, regulatory elements embedded in coding regions, and effects on transcription and translation efficiency. Additionally, some sequences may be unstable, toxic to the host, or prone to recombination or errors, making only a small subset of possible codes biologically viable.

Homework Questions – Dr. LeProust

1. What is the most commonly used method for oligo synthesis currently?

The most commonly used method for oligonucleotide synthesis today is solid-phase phosphoramidite chemistry. This method builds DNA sequences one nucleotide at a time on a solid support, allowing for automated, high-throughput synthesis with relatively high efficiency.

2. Why is it difficult to make oligos longer than 200 nucleotides via direct synthesis?

As oligo length increases, the cumulative error rate also increases because each nucleotide addition step is not 100% efficient. Even small inefficiencies compound over hundreds of cycles, leading to truncated sequences and synthesis errors. Beyond approximately 200 nucleotides, the yield of full-length, error-free oligos becomes very low, making direct synthesis impractical.

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

A 2000 bp gene cannot be synthesized directly because the error rate and yield would be unacceptable. The probability of producing a full-length, correct sequence drops exponentially with length. Instead, long genes are assembled from shorter, overlapping oligos using enzymatic methods such as PCR-based assembly or Gibson assembly, which allow error correction and selection of correct constructs.

Homework Question – George Church

(Chosen option: Essential amino acids & the Lysine Contingency)

What are the 10 essential amino acids in animals, and how does this affect the “Lysine Contingency”?

The ten essential amino acids in animals are histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, valine, and arginine (arginine is essential during growth). These amino acids cannot be synthesized de novo by animals and must be obtained from the diet. This has important implications for the “Lysine Contingency,” which highlights lysine as a critical limiting amino acid in many food systems. Since animals depend entirely on external sources for lysine, disruptions in lysine availability—whether due to ecological change, agricultural limitations, or engineered biological systems—can strongly constrain growth and survival. This underscores how deeply biological systems are shaped by metabolic dependencies and how small molecular constraints can scale into major evolutionary, nutritional, and ethical considerations.