Week 1 HW: Principles and Practices

Class Assignment Week 1 HW

1. First, describe a biological engineering application or tool you want to develop and why.

• Programmable LINE-1–based gene insertion for safe, locus-specific genome engineering

The biological engineering tool I am developing as part of my research is a programmable LINE-1 (L1)–based gene insertion system that enables targeted, large-payload integration into mammalian genomes without relying on double-strand breaks. This system builds on the natural target-primed reverse transcription (TPRT) mechanism of LINE-1 retrotransposase, combined with programmable Cas9 nickases to direct integration toward predefined genomic loci.

Current genome engineering methods face trade-offs between payload size, efficiency, and genomic safety. Viral vectors have limited cargo capacity and integration risks, while CRISPR-based homology-directed repair struggles with efficiency and cell-type specificity. A controlled, RNA-mediated L1 platform could enable safer gene insertion for cell therapy, synthetic circuits, and long-term recording applications, particularly in post-mitotic or hard-to-edit cells.

However, because L1 elements are naturally mobile, mutagenic, and evolutionarily active, engineering them raises legitimate concerns around genomic instability, misuse, and unintended propagation. This makes the technology a strong candidate for proactive governance design alongside technical development.

2. Governance and Policy Goals

Primary goal: Enable beneficial use while preventing harm

To ensure that engineered LINE-1–based genome writing technologies contribute to an ethical future, I focus on three overarching governance goals. Each goal is broken down into concrete, actionable sub-goals.

Goal 1: Prevent harmful or uncontrolled genomic integration

The technology should minimize biological risks associated with unintended genome modification.

• Sub-goal 1.1: Reduce off-target insertions and genomic instability
• Sub-goal 1.2: Prevent autonomous, self-sustaining, or uncontrolled retrotransposition activity

Goal 2: Reduce misuse or repurposing for harmful applications

Safeguards should limit the potential for misuse while maintaining accountability.

• Sub-goal 2.1: Limit the use of engineered LINE-1 systems to clearly defined research, therapeutic, or diagnostic contexts
• Sub-goal 2.2: Ensure traceability and responsibility for engineered constructs and their downstream use

Goal 3: Preserve accessibility for constructive applications

Governance should support innovation rather than restrict legitimate research.

• Sub-goal 3.1: Encourage use within regulated therapeutic, academic, and diagnostic settings
• Sub-goal 3.2: Avoid governance frameworks that unnecessarily block basic research, creativity, or responsible innovation

3. Next, describe at least three different potential governance “actions” by considering the four aspects below (Purpose, Design, Assumptions, Risks of Failure & “Success”).

Below, I outline three complementary governance actions that involve technical, institutional, and regulatory actors. Each action is described in terms of its Purpose, Design, Assumptions, and Risks of Failure or Success.

Option 1: Technical containment in the design

Purpose

Currently, much of the safety of engineered retroelements relies on best practices and trust. I propose embedding molecular containment features directly into L1-based systems to reduce the risk of uncontrolled genome modification.

Design

Technical safeguards could include:

• Splitting ORF2p into components that only function together
• Making retrotransposition activity dependent on exogenous factors
• Encoding sequence barcodes or self-limiting elements (e.g. kill switches)
• Inactivating autonomous retrotransposition capacity

Actors involved

• Academic researchers developing the tools
• Funding agencies and scientific journals
• Companies developing clinical or industrial applications

Assumptions

• Molecular safeguards remain stable over time
• Researchers implement safeguards faithfully

Risks of failure or “success”

• Failure: Safeguards could fail or be bypassed, particularly under loss of selective pressure
• Success risk: Effective containment may create false confidence and reduce oversight

Option 2: Tiered access and licensing for engineered retrotransposon systems

Purpose

At present, access to genome engineering tools is largely unrestricted once published. I propose a tiered access model, similar to approaches used in pathogen research or dual-use chemicals.

Design

• Open publication of concepts and safety data
• Controlled access to full constructs via material transfer agreements (MTAs)
• Licensing tied to institutional biosafety approval and defined project scope

Actors involved

• Universities and technology transfer offices
• Funding bodies
• Journals and data repositories

Assumptions

• Access restrictions meaningfully reduce misuse
• Institutions enforce licensing and access rules consistently

Risks of failure or “success”

• Failure: Informal or “black-market” sharing persists
• Success risk: May disadvantage low-resource laboratories or institutions

Option 3: Traceability and post-use monitoring

Purpose

Most governance efforts focus on prevention rather than what happens after deployment. I propose post-use monitoring to detect unintended spread or misuse of engineered L1 systems.

Design

• Standardized sequence barcodes embedded in engineered L1 constructs
• Genomic monitoring in clinical or industrial settings
• Shared reporting standards for insertion sites and activity

Actors involved

• Clinical developers
• Regulatory agencies
• Standards organizations

Assumptions

• Barcodes remain stable and detectable over time
• Monitoring data are shared transparently

Risks of failure or “success”

• Failure: Monitoring may be incomplete or inconsistently applied
• Success risk: Privacy, consent, or data governance concerns in clinical contexts

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.

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.

Based on this analysis, I would prioritize Option 1 (technical containment) as the basic governance strategy, complemented by Option 3 (traceability and post-use monitoring) in higher-risk or clinical contexts.

I prioritize technical containment because it integrates safety directly into the biological system, reducing reliance on user behaviour or downstream oversight. This approach fits well with how SynBio is actually practiced in academic labs, where most control exists at the design stage. It is also scalable and low-burden, allowing safety features to propagate naturally as tools are shared and reused.

As containment alone is not sufficient, option 3 addresses what happens after deployment by allowing detection, and accountability if unintended outcomes occur. This is particularly important in clinical or industrial contexts, where engineered systems may persist long-term. The main trade-off is added complexity and potential privacy concerns, but these can be managed with clear standards and limited monitoring scopes.

I would apply Option 2 (tiered access and licensing) more selectively. While appropriate for especially powerful systems, broad restrictions risk slowing basic research and disadvantaging smaller labs, and may not effectively prevent misuse.

Overall, this recommendation assumes that molecular safeguards remain stable over time. It is primarily aimed at academic labs, funding agencies, and the braod SynBio community, where early incentives and expectations can shape responsible development without impeding innovation.

Personal Reflection

One thing that stood out to me this week is how easy it is for governance to become either too abstract or too restrictive. As someone working directly on developing genome engineering tools, I feel the balance between wanting freedom to innovate and recognizing that some tools need extra regulation.

A governance action I’d add is encouraging or requiring short ethical impact statements alongside new genome writing tools, especially at the funding or publication stage, as a way to slow down to think.

Assignment (Week 2 Lecture Prep)

Professor Jacobson

1. Error rate of DNA polymerase, genome size, and how biology deals with it

The intrinsic error rate of DNA polymerase is approximately 10⁻⁶ errors per nucleotide incorporated. With proofreading activity (3’→5’ exonuclease), this improves to around 10⁻⁷, and after post-replicative mismatch repair, the final error rate is reduced to roughly 10⁻⁹–10⁻¹⁰ per base per replication.
(Source: https://www.nature.com/articles/cr20084)

The human genome is approximately 3 × 10⁹ base pairs, meaning that without error correction, thousands of mutations would occur every time a cell divides. Biology resolves this through multiple layers of quality control:

• Polymerase proofreading during replication
• Mismatch repair systems acting after replication
• Cell-cycle checkpoints and apoptosis to eliminate heavily damaged cells

2. How many different DNA codes can specify an average human protein, and why most don’t work in practice

An average human protein is approximately 350 amino acids long. Because many amino acids are encoded by multiple codons, there are many DNA sequences that could encode the same protein sequence.

In practice, however, most of these sequences do not work well. Codon choice affects translation efficiency, mRNA folding and stability, ribosome pausing, and co-translational protein folding. Some sequences can create splice sites, polyadenylation signals, or regulatory motifs that disrupt expression. Although the genetic code is redundant, biological constraints strongly limit which DNA sequences are functionally viable.

Dr. LeProust

3. Most commonly used method for oligo synthesis

The most widely used method for oligonucleotide synthesis is solid-phase phosphoramidite DNA synthesis, in which DNA is built one nucleotide at a time on a solid support through sequential chemical coupling steps.

4. Why oligos longer than ~200 nt are difficult to make

Each chemical coupling step has a small probability of failure. As oligos increase in length, these small errors accumulate. Beyond approximately 200 nucleotides, the fraction of molecules that are full-length and error-free drops sharply, making synthesis inefficient and error-prone.

5. Why a 2000 bp gene cannot be synthesized directly

Direct synthesis of a 2000 bp gene would require thousands of consecutive chemical reactions. The cumulative error rate would be so high that almost none of the molecules would be correct. Instead, long genes are typically assembled enzymatically from shorter oligos using methods such as PCR-based assembly or Gibson assembly.

George Church (chosen question)

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
• Arginine

Animals cannot synthesize these amino acids de novo and must obtain them through their diet.

In Jurassic Park, the lysine contingency is presented as a biological fail-safe: the engineered dinosaurs are unable to synthesize lysine and are therefore assumed to be unable to survive outside a controlled environment where lysine is supplemented. However, in my opinion this logic does not hold up biologically. All animals, including mammals and birds, are naturally lysine-auxotrophic and survive perfectly well by acquiring lysine through their diet. Herbivores obtain lysine from plants or plant-associated microbes, and carnivores obtain it by consuming herbivores.

Given this, lysine auxotrophy does not constitute a meaningful biocontainment strategy. Rather than creating a biological dependency, it simply makes the dinosaurs metabolically equivalent to normal animals. In this sense, the lysine contingency treats a normal dietary limitation as if it were a true safety mechanism, when in reality it offers no real containment.
(Source: https://jurassicpark.fandom.com/wiki/Lysine_contingency)