Week 1 HW: Final Proyect Ideas
What you would like to biomanufacture and why?
My motivation stems from a deep personal and professional commitment to women’s health. I believe that gynecology and maternal care are historically underserved areas in science, and my long-term goal as a scientist is to develop accessible technologies for women. I have already taken steps toward this mission by conducting research in breast cancer using AI models, which sparked my passion for early diagnostic tools.
Additionally, I have solid foundational experience in materials science; my undergraduate thesis focused on synthesizing chitosan-based polymeric films loaded with propolis and copper oxide nanoparticles for surgical wound healing.Building upon that expertise, I now want to pivot towards maternal health by biomanufacturing biodegradable breast pads engineered with a biosensor for the early detection of mastitis, designed specifically for breastfeeding mothers . This pad will feature a soft sodium alginate matrix cross-linked with calcium chloride (CaCl₂), a natural and non-toxic polysaccharide framework capable of safely encapsulating a safe biosensor.
This innovation addresses both a critical clinical need and an environmental crisis. Lactational mastitis affects between10% to 33% of breastfeeding women, with 17% of them abandoning breastfeeding due to severe complications like breast abscesses (WHO, 2000). The global breast pads market relies on synthetic consumables that generate tons of plastic waste. By engineering these pads from biodegradable biomaterials, this project will drastically reduce single-use plastic waste while delivering a life-changing diagnostic tool for maternal care.
What living system would you use?
Approach A: Microbial Cell System The first approach utilizes a safe, Biosafety Level 1 (BSL-1) biological chassis, such as Escherichia coli or Bacillus subtilis safe strains. The bacterial genetic circuit will be engineered to trigger a visible chromatic signal in response to increased salinity and high osmolality. Since medical literature demonstrates that breast milk experiences a sharp rise in sodium concentration during early-stage mastitis, we will use this ionic shift as our target biomarker. Consequently, the encapsulated bacteria within the alginate matrix will act as a real-time visual indicator, undergoing a localized color shift upon direct contact with the affected milk.
Approach B: Cell-Free Protein Synthesis (CFPS) System This approach eliminates the regulatory challenges of using living organisms near postpartum skin. The incorporation of CFPS systems into sodium alginate is strongly supported by recent scientific literature; for instance, Green et al. (2025) noted that the porous alginate architecture actually enhances the performance of CFPS systems by inducing a molecular crowding effect that favors the molecular interactions necessary for transcription and translation. In this system, the osmosensitive machinery and an enzymatic reporter are extracted from the cells and embedded directly into the polymeric matrix alongside a colorimetric substrate.
How would you use automation?
Inspired by Jun et al. (2025) and Petzold & Mukhopadhyay (2025), this framework demonstrates how every stage of biosensor manufacturing and operation can be fully automated within a centralized cloud lab. This integrated infrastructure connects robotic systems, automated liquid handling, and centralized data managementto repeat the Design–Build–Test–Learn (DBTL) cycle with minimal manual intervention.
The workflow operates across three main stages:
Phase 0: Formulation Screening: To handle highly viscous biomaterials (such as alginate, collagen, or chitosan) and reduce experimental variability, a Formulatrix Mantis dispenser utilizes microfluidic chips to precise-screen multiple initial formulations. The DBTL Core (Build & Design): Once formulation variables are locked, a robotic liquid-handling station controls parameters like dispensing volume and flow rate into hemispherical molds.Concurrently, a separate automated station standardizes the CaCl₂ crosslinking step by strictly regulating spraying patterns, distance, and flow conditions.
Adaptable Architecture (Test & Learn): The automated pipeline dynamically shifts between two biological paths before undergoing real-time analytical monitoring:
*Microbial Path: Robotic arms execute genetic assembly and automated E. coli transformation. Cells are then placed in robotic incubators that manage nutrients, temperature, and agitation. Real-time kinetic monitoring is handled via integrated plate readers and on-line flow cytometers measuring optical density and single-cell fluorescence.
*Cell-Free (CFPS) Path: Bypassing cellular viability constraints, a high-throughput pipeline automates cell extraction via centrifugation and high-pressure lysis. After hydrogel deposition, the mixture enters an automated freeze-dryer for vacuum dehydration, yielding room-temperature stable biosensor powders.
Ultimately, active learning closes the loop. Instead of testing all variations, a machine learning model correlates fabrication inputs (alginate concentrations, crosslinking, thickness) with measured outputs (sensitivity and mechanical resistance) to intelligently select the next most informative experiments.
How much would you need to make, and how long would it take?
To provide an accurate cost estimate, a demographic model was developed using Google AI to project the breastfeeding population and their product consumption in Chihuahua City.
By cross-referencing INEGI census data with the state’s annual births (53,871), it was determined that the municipality accounts for 13,467 annual births (1,122 per month). Applying prevalence rates from the National Health and Nutrition Survey (ENSANUT) — 34.2% for exclusive breastfeeding (under 6 months) and 46% for continued breastfeeding (6 to 12 months) — the model estimates an active population of 5,398 breastfeeding mothers on any given day.
Based on this data, daily nursing pad consumption was projected across two stages: an early high-demand phase (7 units/day per mother) and a regulated stabilization phase (3 units/day per mother). Segmenting the population yields a theoretical need of 20,798 pieces daily. Considering that 80% of urban mothers in Mexico use disposable products, Chihuahua City consumes approximately 16,638 disposable nursing pads per day. Commercially, this translates to a monthly market demand of 500,000 units, or roughly 16,600 standard boxes (60-count presentation).
To create the manufacturing production cost assumptions, I used the following prompt in Claude AI:
The project consists of manufacturing 500,000 biodegradable breast pads for early detection of lactational mastitis. Each pad will use a soft sodium alginate matrix cross-linked with calcium chloride (CaCl₂), shaped in a hemispherical mold, and integrated with either a safe BSL-1 microbial biosensor or a cell-free protein synthesis (CFPS) biosensor capable of producing a visible colorimetric response to changes associated with mastitis. The manufacturing process includes automated preparation and mixing of alginate and biological components, high-precision liquid dispensing using systems such as the Formulatrix Mantis, automated deposition of a consistent volume into hemispherical molds, controlled CaCl₂ spraying for crosslinking, and, for the CFPS version, automated freeze-drying for long-term storage. The production line could integrate robotic liquid handling, automated molding and crosslinking, quality-control imaging, data collection, and active-learning algorithms to optimize formulation and manufacturing parameters while minimizing experimental iterations. The complete workflow could ultimately be integrated into a cloud-lab or automated manufacturing platform. The objective is to estimate the capital equipment, consumables, raw materials, labor, energy, automation, quality control, packaging, and operating costs required to manufacture 500,000 pads at production scale.
Based on this prompt, I analyzed the results obtained, verified the cost information sources, and used my own judgment to determine if the data seemed consistent and aligned with the project.
For the economic scenario considered, production is assumed to be divided equally between the two approaches: 50% of the pads using CFPS and 50% using the microbial system, corresponding to 250,000 units of each type. A 10% production loss/rejection rate is included.
Based on these assumptions, the model estimates a total production cost of approximately $415,900 USD, corresponding to an estimated cost of $0.83 per pad. The half-and-half split is only for calculation purposes — in the initial steps, I would be deciding which of the two systems is best for the project.
It is important to note that this is a preliminary cost estimate, as labor costs have not yet been included. In addition, several manufacturing parameters — such as the final mold design, production layout, quality-control requirements, packaging infrastructure, and other auxiliary equipment — still need to be defined. Therefore, a more detailed quotation based on fewer assumptions and more finalized manufacturing specifications would likely result in a higher total cost. The current estimate should consequently be considered a preliminary reference for evaluating the economic feasibility of producing 500,000 pads at scale.
Organization and timeline (4 stages):
The target is to manufacture 500,000 usable breast pads per month. However, I would not initially divide production equally between the microbial and CFPS systems — the 50/50 distribution used in the cost estimate above was only an economic scenario. In practice, I would first use automated experimentation to determine which biological system should be selected for scale-up; once selected, manufacturing would be based primarily on that system.
The process was divided into four automated stages, detailed below:
Estimated overall time: approximately 7–8 weeks from the beginning of automated screening to completion of the first 500,000-pad manufacturing run, assuming the final formulation has already been validated and the manufacturing line can operate with the assumed parallel throughput.
What don´t you know yet?
An important possible bottleneck is the stage 1 screening as a decision gate — the biological system must be selected before scaling up, so any delay in the 1–2 week screening phase pushes back the entire 7–8 week timeline.
Biological reliability over time — cell viability for the microbial system, and successful rehydration/reactivation after freeze-drying for the CFPS system — both assumed, not confirmed.
Cost figures were generated via an AI prompt (Claude AI) and reviewed/adjusted with judgment — not direct supplier quotes for every line item.
Bibliogaphy Li, J., Wu, Y., He, J., & Huang, Y. (2016). A new insight to the effect of calcium concentration on gelation process and physical properties of alginate films. Journal Of Materials Science, 51(12), 5791-5801. https://doi.org/10.1007/s10853-016-9880-0 Mastitis: Causas y manejo. (s. f.). https://www.who.int/es/publications/i/item/WHO_FCH_CAH_00.13 Green, T. P., Talley, J. P., & Bundy, B. C. (2025). Recent Advances in Developing Cell-Free Protein Synthesis Biosensors for Medical Diagnostics and Environmental Monitoring. Biosensors, 15(8), 499. https://doi.org/10.3390/bios15080499 Jun, J., Hong, S., Park, J., Shin, J., & Lee, D. (2025). Automated and Programmable Cell-Free Systems for Scalable Synthetic Biology with a Focus on Biofoundry Integration. Journal Of Microbiology And Biotechnology, 35, e2507019. https://doi.org/10.4014/jmb.2507.07019 Petzold, C. J., & Mukhopadhyay, A. (2025). From bench to biofactory: high-throughput technologies and automated workflows to accelerate biomanufacturing. Current Opinion In Biotechnology, 94, 103320. https://doi.org/10.1016/j.copbio.2025.103320