Week 1 HW: Final Project Ideas

Target: Paper-based or hydrogel-based trace gas sensors specifically tuned to detect hydrogen (H2), carbon dioxide (CO2), and hydrogen sulfide (H2S) in human breath.

Motivation: These gasses are used to diagnose issues with the gut microbiome. Trace levels of these specific gases in human breath can be expencive and challenging to measure. Currently, accurately measuring these gases at parts-per-million (ppm) or parts-per-billion (ppb) concentrations requires either bulky gas chromatography equipment. Theere is some sucess in usingsolid-state MEMS sensors.

For Hydrogen (H2) Palladium MEMS or Metal Oxide Semiconductor (MOx)

• MOx sensors are highly sensitive to humidity
• MOx sensors are will cross-react with other reducing gases

For Carbon Dioxide (CO2) NDIR (Non-Dispersive Infrared) or Thermal Conductivity MEMS

• NDIR sensors are almost immune to cross-sensitivity from other gases, but they are extreamly sensitive to condensation
• diagnosing some digestive issues (H. pylori) requires measuring the exact ratio of the Carbon-13 isotope to the Carbon-12 isotope (13CO2 / 12CO2). Standard NDIR/MEMS sensors cannot differentiate between isotopes.
• Thermal conductivity sensors are impacted by ambient temperature or the presence of other gases.

For Hydrogen Sulfide (H2S) Doped MOx MEMS or Micro-Amperometric (Electrochemical)

• H2S sensors easily cross-react with other volatile sulfur compounds (like methyl mercaptan) and broad Volatile Organic Compounds (VOCs) found in breath (from food, mouthwash, or endogenous metabolism).
• Condensation from breath will immediately blind or destroy an electrochemical sensor’s membrane and heavily disrupt a MOx sensor’s electrical baseline
• High concentrations of H2S can permanently bind to and “poison” metal oxide sensors, drastically shortening their lifespan. They are also easily poisoned by siloxanes (found in cosmetics and shampoos)

Biomanufacturing these sensors using engineered proteins may offer a distinct advantage: biological receptors operate natively at ambient temperatures and might achieve chemical specificity without cross-reacting with the hundreds of other volatile organic compounds (VOCs) present in human breath. The goal is to create a low-cost, disposable diagnostic strip that changes color or fluoresces when exposed to specific breath gases. This also avoids the risk of damagin and expensive sensor. With test strips if they are damaged they can be mroe easily replaced and or consumable anyways.

Regection reasons for livign cells:

• Living cells are possibly a regulatory and safety nightmare.
• Shelf life is poor and long term storage condiditions are not practical
• Form factor is impractical for end user
• Transportation is impractical
• live cells might die or not be sensitive fast anough to these gasses to be practical.

Escherichia coli (E. coli) Crude Lysate A common choice for prototyping cell-free diagnostics.

• Pros:

  • Highest Yield: Capable of producing the highest amount of protein (often milligrams per milliliter of reaction).
  • Lowest Cost: E. coli grows rapidly in cheap media, making industrial scale-up incredibly cost-effective.
  • Proven Lyophilization: Highly optimized protocols exist for freeze-drying E. coli extracts onto paper while retaining activity for months.
  • Rapid Prototyping: Transcription and translation are coupled, allowing you to go from a raw DNA plasmid to functional sensor proteins in 2 to 4 hours.

• Cons:

  • Endotoxin Hazard: As a Gram-negative bacterium, it is packed with toxic lipopolysaccharides (LPS) which can trigger severe immune responses if inhaled or exposed to human mucous membranes.
  • The “Black Box”: Crude extract contains endogenous proteases and nucleases that can chew up your synthetic DNA or degrade your sensor proteins over time.
  • No Post-Translational Modifications: It lacks the biological machinery (chaperones) to properly fold complex mammalian proteins or add sugar chains (glycosylation).

Mitigation methods:

• Use an “Endotoxin-Free” Mutant Strain: I propose using a genetically modified strain specifically designed for medical biomanufacturing, such as ClearColi™.
• Physical Encapsulation (Hardware Safety): the paper strip must be separated from the mouthpiece by a hydrophobic, gas-permeable membrane (like a thin sheet of PTFE/Teflon or Nafion).
• Switch to a Eukaryotic Cell-Free System

Eukaryotic Lysates (Wheat Germ, Rabbit Reticulocyte, Yeast)

• Pros:
  • Completely Endotoxin-Free: Eukaryotic cells do not produce LPS, immediately solving the primary biological safety hazard for a human breath analyzer.
  • Complex Protein Folding: They contain the proper chaperones to fold complex, multi-domain proteins and can perform post-translational modifications.
• Cons:
  • Lower Yields: They typically produce significantly less protein per milliliter than E. coli.
  • High Cost & Complex Scaling: Harvesting reticulocytes from rabbits or processing toasted wheat germ is vastly more expensive and difficult to scale than fermenting a vat of bacteria.
  • Slower Kinetics: Eukaryotic translation is often slower, which might delay the time-to-result for a rapid diagnostic.

Alternative Bacterial Chassis (Bacillus subtilis or Vibrio natriegens)

• Pro:
• Con:

The PURE System (Protein synthesis Using Recombinant Elements)

• Pro:
  • Purity & Control:
  • High Stability:
  • Great for Engineering:
• Con:
  • Exorbitant Cost:

Technology: An automated desktop liquid handler, such as the Opentrons OT-2, paired with an automated plate reader.

Workflow Integration: Automation is critical for the prototyping and optimization phase. Formulating a cell-free sensor requires balancing dozens of variables: lysate concentration, DNA template ratios, buffer pH, and the matrix material (hydrogel vs. paper). The Opentrons would be programmed to rapidly dispense varying micro-aliquots of these components into 384-well microplates. We would then expose these plates to calibrated gas mixtures using a manifold and read the resulting fluorescence. The automation allows us to map the entire parameter space to find the formulation with the fastest reaction kinetics and highest signal-to-noise ratio quicker.

Totally an AI Wild guess:

(a) Scale: Let’s define our scale as a pilot clinical trial requiring 10,000 sensor strips.

• Assume each paper test strip requires a 5 µL spot of the active cell-free sensor reaction.
• 10,000 strips * 5 µL/strip = 50,000 µL = 50 mL of total active cell-free reaction volume.

(b) Time: To manufacture this pilot batch, the process breaks down as follows:

• Biomass Production (2 days): Growing a standard 5L benchtop bioreactor of the source E. coli strain to a high optical density.
• Lysate Preparation (1 day): Harvesting, washing, lysing the cells, and centrifuging to extract the raw active lysate. A 5L culture easily yields hundreds of milliliters of lysate.
• Sensor Synthesis (6 hours): Combining the lysate with the DNA templates and energy buffers, and running the bulk cell-free reaction to synthesize the sensor proteins.
• Automated Dispensing (3 hours): Using an industrial micro-dispenser to spot the 50 mL solution across 10,000 paper strips (at ~1 strip per second).

Total manufacturing time: Approximately 3.5 to 4 days to produce the biological active material for 10,000 pilot tests.

The 3 biggest technical bottlenecks to test:

• Reaction Kinetics vs. Exposure Time: A single human exhalation lasts only a few seconds. We do not know if the biological gas-binding event occurs rapidly enough to trigger a visible signal in that short window, or if the breath needs to be captured in a bag and exposed to the sensor over several minutes.
• Humidity and Cross-Reactivity: Human breath is nearly 100% relative humidity. We must experimentally determine if heavy water vapor condenses and washes out the lyophilized proteins on the paper matrix, or if the sensor cross-reacts with other VOCs.
• Shelf-Life Stability: While cell-free systems can be freeze-dried, their stability at fluctuating room temperatures degrades over time. We need accelerated aging tests to see if the sensors remain viable after 6+ months in storage.

Sources:

AI Citation:

• Client: Gemini
• Prompts Used:
  1. “I have a project assignment not for marks on a free course. I want to explore the possability to make sensors for measuring samples of trace gasses. Possibly for doign trace gas analysis of human breath for hydrogen, CO2 and H2S as those are somewhat difficult to measure accuratly wiht sensors at low cots… [Included the full HTBAA Homework #1 prompt]”
  2. “how viable is this idea, has it been done”
  3. “I can imagine breathing into a bag or balloon of a fixed volume to take the sample from… Alternatively some type of flow rate limiter might work with a sensor to measure the total flowed volume over the sample paper.”
  4. “What type of low-power miniature valves could be controlled by a microcontroller to trap the breath sample over the sensor?”
  5. “what mems sensorts are used for each of these gasses and what are their limitations such as cross sensitivity and moisture sensitivity and risk of contamination. Also are they sensitive enought to actualy diagnose issues in human digestions”
  6. “why: What living system would you use? Chassis: A Cell-Free Protein Synthesis (CFPS) system derived from E. coli lysate.”
  7. “but there are safety issues with E. coli Lysate”
  8. “list a bullet form of pros and cons to E. coli Lysate and then also list for other options”
  9. “provide me a list of sources for this last answer… and for the MEMS sensor data… in scientific style [and in a markdown text box]”

Cell-Free Protein Synthesis (CFPS) & Biomanufacturing Sources

• Carlson, E. D., Gan, R., Hodgman, C. E., & Jewett, M. C. (2012). Cell-free protein synthesis: Applications come of age. Biotechnology Advances, 30(5), 1185–1194. https://doi.org/10.1016/j.biotechadv.2011.09.016
• Jung, J. K., Alam, K. K., Verosloff, M. S., Capdevila, D. A., Desmau, M., Clauer, P. R., Lee, J. W., Nguyen, P. Q., Pastén, P. A., Matias, S. J., Bingol, K., Giedroc, D. P., Collins, J. J., & Lucks, J. B. (2020). Cell-free biosensors for rapid detection of water quality. Science Advances, 6(50), eaaz2869. https://doi.org/10.1126/sciadv.aaz2869
• Kelwick, R., Webb, A. J., MacDonald, J. T., & Freemont, P. S. (2016). Development of a Bacillus subtilis cell-free transcription-translation system for prototyping regulatory elements. Metabolic Engineering, 38, 370–381. https://doi.org/10.1016/j.ymben.2016.09.008
• Pardee, K., Green, A. A., Ferrante, T., Cameron, D. E., DaleyKeyser, A., Yin, P., & Collins, J. J. (2014). Paper-based synthetic gene networks. Cell, 159(4), 940–954. https://doi.org/10.1016/j.cell.2014.10.004
• Wang, Y., Zhang, W., Wu, X., & Li, Z. (2022). Cell-free PURE system: Evolution and achievements. Frontiers in Bioengineering and Biotechnology, 10, 878345. https://doi.org/10.3389/fbioe.2022.878345
• Wiegand, D. J., Lee, H. H., Ostrov, N., & Church, G. M. (2018). Cell-free protein synthesis from fast-growing Vibrio natriegens. Frontiers in Microbiology, 9, 2942. https://doi.org/10.3389/fmicb.2018.02942

MEMS Gas Sensors & Breathomics Sources

• Gatty, H. K., Stemme, G., & Roxhed, N. (2018). A miniaturized amperometric hydrogen sulfide sensor applicable for bad breath monitoring. Micromachines, 9(12), 612. https://doi.org/10.3390/mi9120612
• Li, Y., Chen, Z., et al. (2025). A ZnO packaged MEMS hydrogen sensor for reliable SIBO breath analysis with anti-H2S interference. Chemical Engineering Journal, 522, 167331. https://doi.org/10.1016/j.cej.2025.167331
• Shin, H., Kim, D. H., Jung, W., Jang, J. S., Kim, Y. H., Lee, Y., Chang, K., Lee, J., Park, J., Namkoong, K., & Kim, I. D. (2021). Surface activity-tuned metal oxide chemiresistor: Toward direct and quantitative halitosis diagnosis. ACS Nano, 15(9), 14207–14217. https://doi.org/10.1021/acsnano.1c01350
• Sun, Y., Han, L., et al. (2018). Simultaneous detection of hydrogen and methane in breath for the diagnosis of small intestinal bacterial overgrowth by fast gas chromatography. Analytical Methods, 10(35), 4329-4336. https://doi.org/10.1039/C8AY01128C
• Tai, H., Wang, S., Duan, Z., & Jiang, Y. (2020). Metal oxide-based gas sensors for the detection of exhaled breath markers. Sensors, 20(11), 3096. https://doi.org/10.3390/s20113096
• Zhang, Y., Wang, X., et al. (2024). Development of a compact NDIR CO2 gas sensor for a portable device. Micromachines, 15(10), 1203. https://doi.org/10.3390/mi15101203

Final Project