Engineering Overview
Each component has a phase ring on the left. Click the phase buttons to toggle their description panels and the matching entries in the component's list below.
Hardware
Hardware
Iteration 1 - Agarose Filter Medium
PHASES
Design
The production method and properties of the initial filter medium were selected based on literature review and anticipated user needs. Freeze casting was selected as a relatively safe, accessible, effective, and low cost method of producing consistent porous media that support microbial growth in comparison to alternative methods [1], [2]. Agarose was selected as the structural material for its desirable properties, compatibility with a variety of microorganisms, and availability in synthetic biology labs [3], [4]. Yeast extract, tryptone, and glucose (YTG) nutrients were incorporated at the same concentration as YPD agar to allow comparison between the novel medium and the standard agar medium for yeast growth. 5 wt% agarose was selected based on desirable performance observed in previously reported studies [3] .
Goals:
1. Medium Validation
Assess the growth of S. cerevisiae on 5 wt% agarose medium in comparison to the standard agar medium, measured in terms of CFU/mL.
2. Freeze Casting Validation
Validate the freeze casting method used by assessing if the filter medium produced satisfies user needs and is reproducible.
3. Growth Validation
Determine whether S. cerevisiae grows on and in the filter medium in aqueous environments.
The production method and properties of the initial filter medium were selected based on literature review and anticipated user needs. Freeze casting was selected as a relatively safe, accessible, effective, and low cost method of producing consistent porous media that support microbial growth in comparison to alternative methods [1], [2]. Agarose was selected as the structural material for its desirable properties, compatibility with a variety of microorganisms, and availability in synthetic biology labs [3], [4]. Yeast extract, tryptone, and glucose (YTG) nutrients were incorporated at the same concentration as YPD agar to allow comparison between the novel medium and the standard agar medium for yeast growth. 5 wt% agarose was selected based on desirable performance observed in previously reported studies [3] .
Goals:
1. Medium Validation
Assess the growth of S. cerevisiae on 5 wt% agarose medium in comparison to the standard agar medium, measured in terms of CFU/mL.
2. Freeze Casting Validation
Validate the freeze casting method used by assessing if the filter medium produced satisfies user needs and is reproducible.
3. Growth Validation
Determine whether S. cerevisiae grows on and in the filter medium in aqueous environments.
Build
1. Medium Validation
Inoculated plates with YTG nutrients and either 5 wt% agarose or 1.5 wt% agar then compared growth in terms of CFU/ml S. cerevisiae broth culture and colony size.
2. Freeze Casting Validation
Freeze dried 10 to 30 mL of media in tubes and flat dishes using the freeze casting procedure then compared the resulting filter media in terms of their porosity, homogeneity, and structural properties including strength and ductility.
3. Growth Validation
Inoculated and assessed the growth of S. cerevisiae on the interior and exterior of filter medium prepared by freeze casting 30 mL of 5 wt% YTG agarose medium in a flat aluminum dish then rehydrating it in water. Growth was assessed visually with the naked eye and using a microscope.
1. Medium Validation
Inoculated plates with YTG nutrients and either 5 wt% agarose or 1.5 wt% agar then compared growth in terms of CFU/ml S. cerevisiae broth culture and colony size.
2. Freeze Casting Validation
Freeze dried 10 to 30 mL of media in tubes and flat dishes using the freeze casting procedure then compared the resulting filter media in terms of their porosity, homogeneity, and structural properties including strength and ductility.
3. Growth Validation
Inoculated and assessed the growth of S. cerevisiae on the interior and exterior of filter medium prepared by freeze casting 30 mL of 5 wt% YTG agarose medium in a flat aluminum dish then rehydrating it in water. Growth was assessed visually with the naked eye and using a microscope.
Figure 1: YTG agarose (left) and agar (right) plates.
Test
1. Medium Validation
The same number of colony forming units were observed on agarose and agar media but the colonies on the agarose medium were significantly smaller than the colonies on the agar medium.
2. Freeze Casting Validation
Freeze casting in plastic tubes resulted in a spongy texture with very large pores, a non-homogenous structure, and a thick film surrounding the filter medium.
Freeze casting in aluminum plates resulted in a rigid homogenous structure with small pores and the formation of a thin film on top of the filter medium.
Filter media freeze casted in aluminum plates were consistent between replicates and when different volumes of media were used.
3. Growth Validation
Growth was visible on the filter medium surface but was washed away when water was added.
Growth on the interior of the filter was not visible.
Microscopy was not possible due to the thickness and porosity of the filter medium, which prevented sufficient light transmission through the sample.
1. Medium Validation
The same number of colony forming units were observed on agarose and agar media but the colonies on the agarose medium were significantly smaller than the colonies on the agar medium.
2. Freeze Casting Validation
Freeze casting in plastic tubes resulted in a spongy texture with very large pores, a non-homogenous structure, and a thick film surrounding the filter medium.
Freeze casting in aluminum plates resulted in a rigid homogenous structure with small pores and the formation of a thin film on top of the filter medium.
Filter media freeze casted in aluminum plates were consistent between replicates and when different volumes of media were used.
3. Growth Validation
Growth was visible on the filter medium surface but was washed away when water was added.
Growth on the interior of the filter was not visible.
Microscopy was not possible due to the thickness and porosity of the filter medium, which prevented sufficient light transmission through the sample.
Figure 2: Filter media freeze dried in aluminum dishes (top) and tubes (bottom).
Learn
1. Medium Validation
Yeast can grow on 5 wt% YTG agarose medium but the rate of growth is slower than on standard 1.5 wt% YTG agar medium. The growth rate may be improved by increasing the concentration of YTG nutrients to increase their availability.
2. Freeze Casting Validation
Freeze casting in flat aluminum dishes produced the most consistent and replicable results, indicating that this method results in the best heat transfer. Filter media produced using this method are more adaptable because the diameter and thickness can be changed and still produce consistent results. This adaptability and the high structural integrity observed make the produced filter medium well suited for implementation in a variety of filtration systems.
3. Growth Validation
S. cerevisiae is capable of growing on rehydrated freeze casted filter medium but does not strongly adhere to its surface. Adhesion should be improved to ensure that cells remain attached to the filter medium when used in real filtration applications where water will flow through the filter. Although yeast is not visible when water is added, a small amount of cells may remain adhered to the scaffold in quantities invisible to the naked eye. For this reason, it is unknown if cells are not present in the filter medium's interior or present but not visible. Since microscopy was determined to be ineffective at visualizing yeast on the filter medium, alternative methods of assessing yeast growth must be investigated.
1. Medium Validation
Yeast can grow on 5 wt% YTG agarose medium but the rate of growth is slower than on standard 1.5 wt% YTG agar medium. The growth rate may be improved by increasing the concentration of YTG nutrients to increase their availability.
2. Freeze Casting Validation
Freeze casting in flat aluminum dishes produced the most consistent and replicable results, indicating that this method results in the best heat transfer. Filter media produced using this method are more adaptable because the diameter and thickness can be changed and still produce consistent results. This adaptability and the high structural integrity observed make the produced filter medium well suited for implementation in a variety of filtration systems.
3. Growth Validation
S. cerevisiae is capable of growing on rehydrated freeze casted filter medium but does not strongly adhere to its surface. Adhesion should be improved to ensure that cells remain attached to the filter medium when used in real filtration applications where water will flow through the filter. Although yeast is not visible when water is added, a small amount of cells may remain adhered to the scaffold in quantities invisible to the naked eye. For this reason, it is unknown if cells are not present in the filter medium's interior or present but not visible. Since microscopy was determined to be ineffective at visualizing yeast on the filter medium, alternative methods of assessing yeast growth must be investigated.
Iteration 2 - Improved Agarose Filter Medium
PHASES
Design
The previous iteration indicated that the growth and adhesion of S. cerevisiae should be improved to ensure the effectiveness and safety of the filter. Higher growth rate results in more cells in the filter medium which can participate in lead accumulation, increasing the effectiveness of the filter. Higher adhesion prevents cells from contaminating the water being filtered, which is important for ensuring the biocontainment of lead-filled cells. Different nutrient compositions were tested based on literature indicating that using 1/5th or double the amount of nutrients may improve S. cerevisiae growth and adhesion by promoting biofilm formation [5], [6]. Different incubation times were also investigated based on Speranza et al.'s findings that biofilm formation occurs after 7 days [5]. Zeolite was incorporated to improve adhesion of S. cerevisiae cells and its metabolites, further supporting the filter's biosafety [7]. Different methods of visualizing growth in the filter medium were assessed to address the problems identified in the previous iteration.
Goals:
1. Improve Growth and Adhesion
Determine the optimal filter medium composition for growth and adhesion.
2. Growth Validation
Determine whether S. cerevisiae grows on and in the filter medium in aqueous environments.
The previous iteration indicated that the growth and adhesion of S. cerevisiae should be improved to ensure the effectiveness and safety of the filter. Higher growth rate results in more cells in the filter medium which can participate in lead accumulation, increasing the effectiveness of the filter. Higher adhesion prevents cells from contaminating the water being filtered, which is important for ensuring the biocontainment of lead-filled cells. Different nutrient compositions were tested based on literature indicating that using 1/5th or double the amount of nutrients may improve S. cerevisiae growth and adhesion by promoting biofilm formation [5], [6]. Different incubation times were also investigated based on Speranza et al.'s findings that biofilm formation occurs after 7 days [5]. Zeolite was incorporated to improve adhesion of S. cerevisiae cells and its metabolites, further supporting the filter's biosafety [7]. Different methods of visualizing growth in the filter medium were assessed to address the problems identified in the previous iteration.
Goals:
1. Improve Growth and Adhesion
Determine the optimal filter medium composition for growth and adhesion.
2. Growth Validation
Determine whether S. cerevisiae grows on and in the filter medium in aqueous environments.
Build
1. Improve Growth and Adhesion
Agarose plates were made with different concentrations of agarose, nutrients, and zeolite then inoculated with S. cerevisiae broth culture. After 2 and 9 days of incubation, the quantity of cell biomass on all plates was compared. Small representative sections of the plates were then swabbed before and after washing with water and the swabs were incubated overnight. The optical density of the broth the swabs were incubated in was then compared to assess the difference between the amount of cells on the plate before and after washing.
2. Growth Validation
a) Developed a protocol for dissolving the filter medium using beta-agarase, separating the cells using centrifugation, then quantifying using a hemocytometer.
b) Analyzed a cross section of the filter medium using phase contrast microscopy.
1. Improve Growth and Adhesion
Agarose plates were made with different concentrations of agarose, nutrients, and zeolite then inoculated with S. cerevisiae broth culture. After 2 and 9 days of incubation, the quantity of cell biomass on all plates was compared. Small representative sections of the plates were then swabbed before and after washing with water and the swabs were incubated overnight. The optical density of the broth the swabs were incubated in was then compared to assess the difference between the amount of cells on the plate before and after washing.
2. Growth Validation
a) Developed a protocol for dissolving the filter medium using beta-agarase, separating the cells using centrifugation, then quantifying using a hemocytometer.
b) Analyzed a cross section of the filter medium using phase contrast microscopy.
Figure 3: Agarose plates with different compositions.
Test
1. Improve Growth and Adhesion
S. cerevisiae grew on all plates and the CFU/mL were too numerous to count.
The optical density data was inconclusive due to high variability between replicates.
The beta-agarase was unable to dissolve the filter medium even when the quantity of beta-agarase recommended by the supplier was scaled to reflect the high agarose concentration in the filter medium. No degradation of the filter medium was observed.
Cells and pores were visible in the filter medium cross section when analyzed using phase contrast microscopy.
1. Improve Growth and Adhesion
S. cerevisiae grew on all plates and the CFU/mL were too numerous to count.
The optical density data was inconclusive due to high variability between replicates.
Figure 4: Inoculated agarose plates.
2. Growth ValidationThe beta-agarase was unable to dissolve the filter medium even when the quantity of beta-agarase recommended by the supplier was scaled to reflect the high agarose concentration in the filter medium. No degradation of the filter medium was observed.
Cells and pores were visible in the filter medium cross section when analyzed using phase contrast microscopy.
Figure 5: Phase contrast microscopy image of filter medium cross section showing S. cerevisiae cells in and around a micropore.
Learn
1. Improve Growth and Adhesion
High variability between replicates resulted in inconclusive data. This was likely due to human error which caused slight differences in the surface area swabbed and pressure used while swabbing. This experiment should be repeated with more replicates, and the quantity of cells should be measured using CFU/mL rather than optical density. The ideal composition results in the highest biomass and smallest difference between cell counts before and after washing, indicating the best growth and adhesion, respectively.
2. Growth Validation
The beta-agarase was likely unable to dissolve the filter medium because the filter medium could not be melted at temperatures within S. cerevisiae's tolerable range. This is likely due to the high concentration of agarose in the filter medium.
1. Improve Growth and Adhesion
High variability between replicates resulted in inconclusive data. This was likely due to human error which caused slight differences in the surface area swabbed and pressure used while swabbing. This experiment should be repeated with more replicates, and the quantity of cells should be measured using CFU/mL rather than optical density. The ideal composition results in the highest biomass and smallest difference between cell counts before and after washing, indicating the best growth and adhesion, respectively.
2. Growth Validation
The beta-agarase was likely unable to dissolve the filter medium because the filter medium could not be melted at temperatures within S. cerevisiae's tolerable range. This is likely due to the high concentration of agarose in the filter medium.
Iteration 3 - Filtration System
PHASES
Design
The permeability of the filter medium was assessed to ensure that it could be used for filtration applications. A filter apparatus that holds the filter media was designed to test the inoculated filter medium. Prototypes were developed based on input from students with engineering and biology backgrounds to ensure the final design accurately reflected real water filtration systems and met the needs of the genetically engineered yeast. A simple modular design that can be easily modified to meet the needs of other researchers was modelled in SolidWorks.
Goals:
1. Assess Filter Medium Permeability
Ensure that water can flow through the filter medium.
2. Filter Apparatus Design Validation
Ensure that the filter apparatus can house the filter medium and operate without leaks.
The permeability of the filter medium was assessed to ensure that it could be used for filtration applications. A filter apparatus that holds the filter media was designed to test the inoculated filter medium. Prototypes were developed based on input from students with engineering and biology backgrounds to ensure the final design accurately reflected real water filtration systems and met the needs of the genetically engineered yeast. A simple modular design that can be easily modified to meet the needs of other researchers was modelled in SolidWorks.
Goals:
1. Assess Filter Medium Permeability
Ensure that water can flow through the filter medium.
2. Filter Apparatus Design Validation
Ensure that the filter apparatus can house the filter medium and operate without leaks.
Build
1. Assess Filter Medium Permeability
Water dyed with food coloring was dispensed onto the filter medium and its permeation through the material was measured visually.
2. Filter Apparatus Design Validation
The filter apparatus was 3D printed and assembled with a rehydrated filter medium. Water was added and the system was assessed for leaks.
1. Assess Filter Medium Permeability
Water dyed with food coloring was dispensed onto the filter medium and its permeation through the material was measured visually.
2. Filter Apparatus Design Validation
The filter apparatus was 3D printed and assembled with a rehydrated filter medium. Water was added and the system was assessed for leaks.
Test
1. Assess Filter Medium Permeability
The surface of the rehydrated filter medium is much smoother and less porous than its interior. When colored water was dispensed on the surface, it slid off rather than permeating into the material.
2. Validate Filter Apparatus Design
The filter apparatus did not leak but water leaked around the edges of the filter medium where it meets the walls of the filter apparatus.
1. Assess Filter Medium Permeability
The surface of the rehydrated filter medium is much smoother and less porous than its interior. When colored water was dispensed on the surface, it slid off rather than permeating into the material.
2. Validate Filter Apparatus Design
The filter apparatus did not leak but water leaked around the edges of the filter medium where it meets the walls of the filter apparatus.
Learn
1. Assess Filter Medium Permeability
The filter medium is not suitable for use as a filter immediately after rehydration due to poor permeability on the surface which impedes the flow of water through the filter medium. Based on the differences in porosity of the surface and interior of the filter medium, the outer surfaces of the filter medium should be removed to expose the more porous interior and thereby improve permeability.
2. Validate Filter Apparatus Design
A tighter fit between the filter medium and the walls of the filter apparatus is required for effective operation of the system. This could be achieved by decreasing the inner diameter of the filter apparatus and rehydrating the filter medium inside the apparatus. During rehydration, the filter medium will expand into the walls of the filter apparatus, creating a tighter seal. The filter apparatus itself is watertight, indicating that the tolerances of the threads used at the connections are correct.
1. Assess Filter Medium Permeability
The filter medium is not suitable for use as a filter immediately after rehydration due to poor permeability on the surface which impedes the flow of water through the filter medium. Based on the differences in porosity of the surface and interior of the filter medium, the outer surfaces of the filter medium should be removed to expose the more porous interior and thereby improve permeability.
2. Validate Filter Apparatus Design
A tighter fit between the filter medium and the walls of the filter apparatus is required for effective operation of the system. This could be achieved by decreasing the inner diameter of the filter apparatus and rehydrating the filter medium inside the apparatus. During rehydration, the filter medium will expand into the walls of the filter apparatus, creating a tighter seal. The filter apparatus itself is watertight, indicating that the tolerances of the threads used at the connections are correct.
Iteration 4 - Improved Filtration System
PHASES
Design
The filtration system was improved based on the learnings of the previous iteration to improve its effectiveness in water filtration applications.
Goals:
1. Improve Filter Medium Permeability
Assess methods of controlling the flow rate of water through the filter medium.
2. Improve Filter Apparatus Design
Improve the connection between the filter medium and walls of the filter apparatus to make it watertight.
The filtration system was improved based on the learnings of the previous iteration to improve its effectiveness in water filtration applications.
Goals:
1. Improve Filter Medium Permeability
Assess methods of controlling the flow rate of water through the filter medium.
2. Improve Filter Apparatus Design
Improve the connection between the filter medium and walls of the filter apparatus to make it watertight.
Build
1. Improve Filter Medium
Water dyed with food coloring was dispensed onto the filter medium with the surface removed and its permeation through the material was measured visually. The water was both gently pipetted and more forcefully applied with a dropper to determine whether different inlet flow rates could be used to control the flow rate of water through the filter medium.
2. Improve Filter Apparatus
The inner diameter of the filter apparatus was reduced and reprinted. A filter medium was rehydrated in the filter apparatus then water was added to assess leakage.
1. Improve Filter Medium
Water dyed with food coloring was dispensed onto the filter medium with the surface removed and its permeation through the material was measured visually. The water was both gently pipetted and more forcefully applied with a dropper to determine whether different inlet flow rates could be used to control the flow rate of water through the filter medium.
2. Improve Filter Apparatus
The inner diameter of the filter apparatus was reduced and reprinted. A filter medium was rehydrated in the filter apparatus then water was added to assess leakage.
Figure 6: Dyed water permeating through filter medium
Test
1. Improve Filter Permeability
Dyed water slowly permeated through the filter medium when gravity-driven.
The water could be pushed through the filter medium faster by applying the water with a higher flow rate.
2. Improve Filter Apparatus Design
No leakage was observed when the new system was tested.
1. Improve Filter Permeability
Dyed water slowly permeated through the filter medium when gravity-driven.
The water could be pushed through the filter medium faster by applying the water with a higher flow rate.
2. Improve Filter Apparatus Design
No leakage was observed when the new system was tested.
Figure 7: 3D printed filter apparatus.
Learn
1. Improve Filter Permeability
Removing the outer layer of the filter medium significantly improved its permeability and therefore effectiveness in filtration applications. The rate of permeation can be controlled by the flow rate of the water at the inlet, allowing for easy control of the amount of time water is in contact with the S. cerevisiae cells. This is important for filtration applications because the S. cerevisiae cells will require time to uptake lead, so the rate of water flow can be modified to achieve optimal treatment efficiency.
2. Improve Filter Apparatus Design
Rehydrating the filter medium in the filter apparatus with an inner diameter smaller than the diameter of the rehydrated filter medium created a watertight seal, which is required for effective water treatment.
1. Improve Filter Permeability
Removing the outer layer of the filter medium significantly improved its permeability and therefore effectiveness in filtration applications. The rate of permeation can be controlled by the flow rate of the water at the inlet, allowing for easy control of the amount of time water is in contact with the S. cerevisiae cells. This is important for filtration applications because the S. cerevisiae cells will require time to uptake lead, so the rate of water flow can be modified to achieve optimal treatment efficiency.
2. Improve Filter Apparatus Design
Rehydrating the filter medium in the filter apparatus with an inner diameter smaller than the diameter of the rehydrated filter medium created a watertight seal, which is required for effective water treatment.
Figure 8: Water on top of a rehydrated filter medium in the filter apparatus.
Software Development
Software
MVP 1: The Core Simulation Engine
PHASES
Design
Sprint Planning: Our initial focus was a proof-of-concept for the core backend. The primary task was to create a single API endpoint that could accept a predefined model, execute it using the Tellurium/RoadRunner simulation engine, and return the resulting time-series data. This sprint was designed to validate our core technical foundation.
Sprint Planning: Our initial focus was a proof-of-concept for the core backend. The primary task was to create a single API endpoint that could accept a predefined model, execute it using the Tellurium/RoadRunner simulation engine, and return the resulting time-series data. This sprint was designed to validate our core technical foundation.
Build
Development: The sprint involved setting up the Python-based backend server and integrating the Tellurium library. We developed the /simulate endpoint with a hardcoded Antimony model, ensuring that the C++-based simulation engine could be called reliably from our web server.
Development: The sprint involved setting up the Python-based backend server and integrating the Tellurium library. We developed the /simulate endpoint with a hardcoded Antimony model, ensuring that the C++-based simulation engine could be called reliably from our web server.
Test
Sprint Review: We demonstrated the working endpoint using an API client. We sent a request and showed the successful return of a JSON payload containing the simulation data, confirming that our technical architecture was sound.
Sprint Review: We demonstrated the working endpoint using an API client. We sent a request and showed the successful return of a JSON payload containing the simulation data, confirming that our technical architecture was sound.
Learn
Sprint Retrospective: We learned that managing Tellurium's dependencies could be complex, reinforcing the need for robust containerization and clear setup documentation. We concluded that the next priority was to replace the hardcoded model with dynamic, user-generated input.
Sprint Retrospective: We learned that managing Tellurium's dependencies could be complex, reinforcing the need for robust containerization and clear setup documentation. We concluded that the next priority was to replace the hardcoded model with dynamic, user-generated input.
MVP 2: GPT-Powered Input
PHASES
Design
Sprint Planning: This sprint's goal was to create an intuitive user interface that could translate natural language into a simulatable model. We planned to integrate OpenAI's GPT-4o to parse plain English descriptions into a structured JSON format. The main backend task was to develop a converter that could transform this JSON object into a valid SBML (Systems Biology Markup Language) file.
Sprint Planning: This sprint's goal was to create an intuitive user interface that could translate natural language into a simulatable model. We planned to integrate OpenAI's GPT-4o to parse plain English descriptions into a structured JSON format. The main backend task was to develop a converter that could transform this JSON object into a valid SBML (Systems Biology Markup Language) file.
Build
Development: We implemented a frontend chat interface that sent user text to the GPT-4o API, which was specifically prompted to return the structured JSON. The core of the sprint was building a backend function that would parse this JSON and programmatically construct a valid SBML XML document from it, correctly defining species, reactions, and parameters.
Development: We implemented a frontend chat interface that sent user text to the GPT-4o API, which was specifically prompted to return the structured JSON. The core of the sprint was building a backend function that would parse this JSON and programmatically construct a valid SBML XML document from it, correctly defining species, reactions, and parameters.
Test
Sprint Review: We demonstrated the full pipeline by typing "A protein X degrades naturally." We then showed the valid SBML file that was generated by the backend, confirming that the natural language input was successfully converted into a structured, standard biological model format.
Sprint Review: We demonstrated the full pipeline by typing "A protein X degrades naturally." We then showed the valid SBML file that was generated by the backend, confirming that the natural language input was successfully converted into a structured, standard biological model format.
Learn
Sprint Retrospective: We learned that prompt engineering was critical for getting consistent and accurate JSON output from the GPT model. We also found that rigidly defining our internal JSON schema was essential for making the conversion to the complex SBML format reliable and robust.
Sprint Retrospective: We learned that prompt engineering was critical for getting consistent and accurate JSON output from the GPT model. We also found that rigidly defining our internal JSON schema was essential for making the conversion to the complex SBML format reliable and robust.
MVP 3: Model Hardening & Automation
PHASES
Design
Sprint Planning: With the input mechanism working, we focused on making model generation more robust. The plan was to implement logic that automatically added default compartments and generated mass-action rate laws for reactions when users did not specify kinetics, thereby reducing the level of detail they needed to provide.
Sprint Planning: With the input mechanism working, we focused on making model generation more robust. The plan was to implement logic that automatically added default compartments and generated mass-action rate laws for reactions when users did not specify kinetics, thereby reducing the level of detail they needed to provide.
Build
Development: This work was concentrated in the backend's model-generation function. We implemented the logic to insert default values and created a consistent naming convention (e.g., Reaction R1 -> parameter k_r1) that allowed the backend to automatically generate complete, valid rate laws from minimal user input.
Development: This work was concentrated in the backend's model-generation function. We implemented the logic to insert default values and created a consistent naming convention (e.g., Reaction R1 -> parameter k_r1) that allowed the backend to automatically generate complete, valid rate laws from minimal user input.
Test
Sprint Review: We demonstrated the increased efficiency by providing a simple input, such as "A degrades into B." We then showed the fully formed Antimony model string generated by the backend, which now included the default compartment and the auto-generated mass-action rate law (k1 * A), proving the automation was successful.
Sprint Review: We demonstrated the increased efficiency by providing a simple input, such as "A degrades into B." We then showed the fully formed Antimony model string generated by the backend, which now included the default compartment and the auto-generated mass-action rate law (k1 * A), proving the automation was successful.
Learn
Sprint Retrospective: We realized that while auto-generation is powerful for new users, expert users would need the ability to override it with their own custom, complex kinetic laws. This became a key requirement for a future sprint.
Sprint Retrospective: We realized that while auto-generation is powerful for new users, expert users would need the ability to override it with their own custom, complex kinetic laws. This became a key requirement for a future sprint.
MVP 4: Dynamic Real-Time Simulation
PHASES
Design
Sprint Planning: The goal of this sprint was to make the simulation results interactive. We designed a UI with sliders for parameters and initial species concentrations. The technical plan focused on creating a high-speed feedback loop: a slider change would trigger a new simulation, and the returned data would update the plot in real-time.
Sprint Planning: The goal of this sprint was to make the simulation results interactive. We designed a UI with sliders for parameters and initial species concentrations. The technical plan focused on creating a high-speed feedback loop: a slider change would trigger a new simulation, and the returned data would update the plot in real-time.
Build
Development: We implemented the slider components on the frontend and optimized the backend /simulate endpoint for faster, repeated calls. We used a frontend plotting library to handle efficient, real-time updates of the graph without requiring a full page reload, creating a seamless user experience.
Development: We implemented the slider components on the frontend and optimized the backend /simulate endpoint for faster, repeated calls. We used a frontend plotting library to handle efficient, real-time updates of the graph without requiring a full page reload, creating a seamless user experience.
Test
Sprint Review: We presented the interactive plot. We dragged a slider for a transcription rate parameter and watched the corresponding protein concentration curve on the graph change its shape instantly. This provided a powerful and intuitive way to explore the model's sensitivity.
Sprint Review: We presented the interactive plot. We dragged a slider for a transcription rate parameter and watched the corresponding protein concentration curve on the graph change its shape instantly. This provided a powerful and intuitive way to explore the model's sensitivity.
Learn
Sprint Retrospective: We learned that running a full backend simulation on every slider movement could introduce lag for very complex models. We created a backlog item to investigate solutions, such as debouncing API calls or exploring future WebAssembly-based simulations that could run entirely in the browser.
Sprint Retrospective: We learned that running a full backend simulation on every slider movement could introduce lag for very complex models. We created a backlog item to investigate solutions, such as debouncing API calls or exploring future WebAssembly-based simulations that could run entirely in the browser.
MVP 5: Refinement and Deployment
PHASES
Design
Sprint Planning: For the final sprint before our official release, we focused on polish and stability. The key goals were to refine the codebase, improve error handling, style the user interface for a clean and professional aesthetic, and prepare the entire application for cloud deployment.
Sprint Planning: For the final sprint before our official release, we focused on polish and stability. The key goals were to refine the codebase, improve error handling, style the user interface for a clean and professional aesthetic, and prepare the entire application for cloud deployment.
Build
Development: This sprint involved comprehensive code reviews, refactoring complex functions for clarity, and writing unit tests for critical backend logic. The frontend team finalized the CSS and component library. Lastly, we created Dockerfiles and deployment scripts for our target cloud platform.
Development: This sprint involved comprehensive code reviews, refactoring complex functions for clarity, and writing unit tests for critical backend logic. The frontend team finalized the CSS and component library. Lastly, we created Dockerfiles and deployment scripts for our target cloud platform.
Test
Sprint Review: We demonstrated the final, polished application. We showcased the consistent design language, responsive layout, and helpful error messages. We also walked through the successful deployment of the application to a staging environment, proving its readiness for production.
Sprint Review: We demonstrated the final, polished application. We showcased the consistent design language, responsive layout, and helpful error messages. We also walked through the successful deployment of the application to a staging environment, proving its readiness for production.
Learn
Sprint Retrospective: We concluded that allocating a dedicated sprint for refinement and deployment was crucial for delivering a high-quality, reliable tool. This final pass caught several minor bugs and significantly improved the user experience, ensuring a successful launch.
Sprint Retrospective: We concluded that allocating a dedicated sprint for refinement and deployment was crucial for delivering a high-quality, reliable tool. This final pass caught several minor bugs and significantly improved the user experience, ensuring a successful launch.
Wet Lab
Wet Lab
Cycle 1 - Entry Vector Assembly & GGA optimization
PHASES
Design
In order to assemble our entry vectors, transcriptional units and multigene plasmids iGEM Guelph has decided to use Golden Gate Assembly (GGA) as the cloning method. The one-pot, scarless and pre-determined ligation order, fits well with the overarching goal of our project; the assembly of a multi-component genetic circuit. More specifically, we have decided to use the Molecular Cloning Yeast Tool Kit (MoClo-YTK), to enable the modularity of designing, building and testing the various components of our circuit (Lee et al. 2015). For this cycle we aimed to optimize the protocol for golden gate assembly using BsaI, to increase the reliability of entry vector, and transcriptional unit construction. Starting with the optimization of golden gate assembly ensures that downstream workflows involving the construction of plasmids using the MoClo-YTK, is efficient and resource efficient.
In order to assemble our entry vectors, transcriptional units and multigene plasmids iGEM Guelph has decided to use Golden Gate Assembly (GGA) as the cloning method. The one-pot, scarless and pre-determined ligation order, fits well with the overarching goal of our project; the assembly of a multi-component genetic circuit. More specifically, we have decided to use the Molecular Cloning Yeast Tool Kit (MoClo-YTK), to enable the modularity of designing, building and testing the various components of our circuit (Lee et al. 2015). For this cycle we aimed to optimize the protocol for golden gate assembly using BsaI, to increase the reliability of entry vector, and transcriptional unit construction. Starting with the optimization of golden gate assembly ensures that downstream workflows involving the construction of plasmids using the MoClo-YTK, is efficient and resource efficient.
Build
Optimization of GGA protocol was completed by attempting to assemble entry vectors 1 -- 5 which consist of 6 level 0 parts from the MoClo-YTK: type 1 -- left connector, type 234r -- GFP dropout, type 5 -- right connector, type 6 -- URA3, type 7 -- CEN6/ARS4, and type 8 -- KanR/ColE1. The following reaction conditions were attempted:
After each reaction was completed, the products were transformed into E.coli DH5α, they were then grown on kanamycin-LB plates and grown at 37 °C
Optimization of GGA protocol was completed by attempting to assemble entry vectors 1 -- 5 which consist of 6 level 0 parts from the MoClo-YTK: type 1 -- left connector, type 234r -- GFP dropout, type 5 -- right connector, type 6 -- URA3, type 7 -- CEN6/ARS4, and type 8 -- KanR/ColE1. The following reaction conditions were attempted:
| Component | GGA.v1 | GGA.v2 | GGA.v3 | GGA.v4 | GGA.v5 |
|---|---|---|---|---|---|
| Nuclease Free H₂O | 15.0 μL | 10.0 μL | 5.0 μL | 0.0 μL | 10.0 μL |
| T4 DNA Ligase Buffer | 1.0 μL | 2.0 μL | 2.0 μL | 1.7 μL | 2.0 μL |
| HI-T4 DNA Ligase | 0.5 μL | 1.0 μL | 1.0 μL | 1.0 μL | 1.0 μL |
| BsaI-HFv2 | 0.5 μL | 1.0 μL | 1.0 μL | 1.0 μL | 1.0 μL |
| Total DNA (6 Parts) | 3.0 μL | 6.0 μL | 6.0 μL | 16.3 μL | 6.0 μL |
| Reaction Volume | 20.0 μL | 20.0 μL | 15.0 μL | 20.0 μL | 20.0 μL |
| GGA Reaction Version | Digestion (°C) | Digestion Time (min) | Ligation (°C) | Ligation Time (min) | Cycles |
|---|---|---|---|---|---|
| GGA.v1 | 37 | 2:00 | 25°C | 5:00 | 23X |
| GGA.v2 | 37 | 2:00 | 25°C | 5:00 | 23X |
| GGA.v3 | 42 | 2:00 | 16 | 5:00 | 70X |
| GGA.v4 | 42 | 2:00 | 16 | 5:00 | 35X |
| GGA.v5 | 42 | 2:00 | 16 | 5:00 | 35X |
Test
In order to evaluate the success of the golden gate reactions as mentioned we aimed to observe green colonies, this suggest that the GFP dropout part was properly incorporated as part of the entry vector. Additionally, selected colonies would be subject to miniprep, restriction enzyme digest and 1.0% gel electrophoresis to determine successful assembly through molecular weight observations. The transformant E.coli of GGA.v1, GGA.v2 and GGA.v4 exhibited minimal and/or a failure to grow any colonies when grown on kanamycin-LB plates, suggesting low transformation or assembly efficiency. Interestingly, GGA.v1 exhibited the growth of multiple pink colonies on a few transformant plates, this is an indicator that the KanR-ColEI (pYTK084) was transformed. This further provides evidence that inefficient assembly was at play, since a transformation of a level 0 part points towards low efficiency digestion or high activity of re-ligation. The only GGA reactions to grow green colonies were GGA.v3 and GGA.v5 which only differed from each other by their total rection volumes: 15.0 μL and 20.0 μL respectively. The subsequent overnight culture, miniprep, digestion and 1.0% gel electrophoresis revealed that, indeed the entry vectors were assembled successfully.
In order to evaluate the success of the golden gate reactions as mentioned we aimed to observe green colonies, this suggest that the GFP dropout part was properly incorporated as part of the entry vector. Additionally, selected colonies would be subject to miniprep, restriction enzyme digest and 1.0% gel electrophoresis to determine successful assembly through molecular weight observations. The transformant E.coli of GGA.v1, GGA.v2 and GGA.v4 exhibited minimal and/or a failure to grow any colonies when grown on kanamycin-LB plates, suggesting low transformation or assembly efficiency. Interestingly, GGA.v1 exhibited the growth of multiple pink colonies on a few transformant plates, this is an indicator that the KanR-ColEI (pYTK084) was transformed. This further provides evidence that inefficient assembly was at play, since a transformation of a level 0 part points towards low efficiency digestion or high activity of re-ligation. The only GGA reactions to grow green colonies were GGA.v3 and GGA.v5 which only differed from each other by their total rection volumes: 15.0 μL and 20.0 μL respectively. The subsequent overnight culture, miniprep, digestion and 1.0% gel electrophoresis revealed that, indeed the entry vectors were assembled successfully.
Learn
From the various iterations of GGA reactions which were attempted, it is clear that for downstream applications GGA.v5 should be used as a starting point for plasmid assembly. By comparing and contrasting the various GGA reactions and conditions used to assemble the entry vectors, iGEM Guelph has learned the following regarding the optimization of GGA: 1. Temperature play a key role in the efficiency of the reaction 2. If the concentration of DNA is too low the reaction is more inefficient 3. The addition of water plays a critical role in the function of restriction enzymes and ligase Although the optimization of GGA using BsaI for the construction of entry vectors and transcription units has been optimized, we still plan to complete a secondary round of GGA which uses BsmBI to assemble multigene plasmids from transcriptional units and a pre made entry vector. If and/or when optimization and troubleshooting is required, we will first asses the temperature, DNA concentration, and amount of water (in the reaction)to streamline workflow.
From the various iterations of GGA reactions which were attempted, it is clear that for downstream applications GGA.v5 should be used as a starting point for plasmid assembly. By comparing and contrasting the various GGA reactions and conditions used to assemble the entry vectors, iGEM Guelph has learned the following regarding the optimization of GGA: 1. Temperature play a key role in the efficiency of the reaction 2. If the concentration of DNA is too low the reaction is more inefficient 3. The addition of water plays a critical role in the function of restriction enzymes and ligase Although the optimization of GGA using BsaI for the construction of entry vectors and transcription units has been optimized, we still plan to complete a secondary round of GGA which uses BsmBI to assemble multigene plasmids from transcriptional units and a pre made entry vector. If and/or when optimization and troubleshooting is required, we will first asses the temperature, DNA concentration, and amount of water (in the reaction)to streamline workflow.
Cycle 2 - Colony PCR of Plasmid for Detection Mechanism Testing
PHASES
Design
To test the riboswitch mechanism used in out genetic regulatory circuit to detect lead, three transcriptional units were required to be assembled: [1] to test the riboswitch mechanism using the Pb7S aptamer (pPb7STest), [2] to test the riboswitch mechanism using the Pb14S aptamer (pPb14STest), and (3) a positive control to ensure that our reporter is expressed in Saccharomyces cerevisiae (pGFPControl). The parts were designed on SnapGene and contained BsaI overhangs sites which followed that of the MoClo-YTK part type system (Lee et al. 2015). Fragments/inserts were synthesized from IDT and Twistbioscience, which arrived in pUC19 plasmids. For more information on the usage of these transcriptional units please refer to Results. Additionally, primers were designed to target the type 1 and type 5 connectors parts to allow for the confirmation of correct assemblies. Since the connectors parts flanked the region of interest, primers which are targeted to these sites enable the PCR amplification of the cloned sequence.
To test the riboswitch mechanism used in out genetic regulatory circuit to detect lead, three transcriptional units were required to be assembled: [1] to test the riboswitch mechanism using the Pb7S aptamer (pPb7STest), [2] to test the riboswitch mechanism using the Pb14S aptamer (pPb14STest), and (3) a positive control to ensure that our reporter is expressed in Saccharomyces cerevisiae (pGFPControl). The parts were designed on SnapGene and contained BsaI overhangs sites which followed that of the MoClo-YTK part type system (Lee et al. 2015). Fragments/inserts were synthesized from IDT and Twistbioscience, which arrived in pUC19 plasmids. For more information on the usage of these transcriptional units please refer to Results. Additionally, primers were designed to target the type 1 and type 5 connectors parts to allow for the confirmation of correct assemblies. Since the connectors parts flanked the region of interest, primers which are targeted to these sites enable the PCR amplification of the cloned sequence.
| Primer | Sequence (5' → 3') |
|---|---|
| ConLS.Fwd | acaagcaacgatctccaggaccatctgaatcatgcgc |
| ConR1.Rev | ggtcatcatgcagtcatccgagcgtgtattgca |
Build
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors were used as the plasmid back bone which our fragments of insert were cloned into. In order to confirm the correct assembly of the desired transcriptional unit's colony PCR and 1.0% gel electrophoresis was used to amplify and visualize the insert of interest. Positive samples will be sent for third generation sequencing to further validate findings.
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors were used as the plasmid back bone which our fragments of insert were cloned into. In order to confirm the correct assembly of the desired transcriptional unit's colony PCR and 1.0% gel electrophoresis was used to amplify and visualize the insert of interest. Positive samples will be sent for third generation sequencing to further validate findings.
Test
In order to determine the successful assembly of the transcriptional units, transformant E.coli DH5α of the GGA reaction were initially green/white colony screened. After the selection of four green colonies per transcriptional unit, colony PCR using DreamTaq polymerase was completed (for more information regarding the DreamTaq Polymerase ThermoScientific protocol please visit experiments). The subsequent 1.0 gel electrophoresis (as shown below) used to confirm the correct PCR amplicons suggested that out insert was indeed successfully cloned into our plasmid backbone, creating the correct transcriptional unit for a majority of colonies selected. Two colonies per transcriptional unit were selected for third generation sequencing to completely ensure correct assembly, which aligned with the ideal constructs on SnapGene.
In order to determine the successful assembly of the transcriptional units, transformant E.coli DH5α of the GGA reaction were initially green/white colony screened. After the selection of four green colonies per transcriptional unit, colony PCR using DreamTaq polymerase was completed (for more information regarding the DreamTaq Polymerase ThermoScientific protocol please visit experiments). The subsequent 1.0 gel electrophoresis (as shown below) used to confirm the correct PCR amplicons suggested that out insert was indeed successfully cloned into our plasmid backbone, creating the correct transcriptional unit for a majority of colonies selected. Two colonies per transcriptional unit were selected for third generation sequencing to completely ensure correct assembly, which aligned with the ideal constructs on SnapGene.
Figure 1: 1.0% agarose gel electrophoresis of colony PCR results
Learn
In order to streamline our workflow and increase the efficiency of our screening transformants E.coli DH5α of potentially successful golden gate assembly, of desired transcription units or entry vectors. Colony PCR should be used to enable the high output screening prior to further confirmation by third generation sequencing. Furthermore, since primers target MoClo-YTK connector parts, primers are versatile and can be used to amplify any desired insert, as long as the appropriate entry vector is used as the plasmid backbone.
In order to streamline our workflow and increase the efficiency of our screening transformants E.coli DH5α of potentially successful golden gate assembly, of desired transcription units or entry vectors. Colony PCR should be used to enable the high output screening prior to further confirmation by third generation sequencing. Furthermore, since primers target MoClo-YTK connector parts, primers are versatile and can be used to amplify any desired insert, as long as the appropriate entry vector is used as the plasmid backbone.
Cycle 3 - Site Directed Mutagenesis
PHASES
Design
Upon transforming our positive control for testing the detection mechanism, GFP was not produced as intended. Altering growth conditions and timing did not restore expression, so we examined the insert sequences. We found: - Positive control: An unintended ATG in the spacer shifted the reading frame, likely producing an incorrect protein. - pPb7STest and pPb14STest: - (1) The riboswitch was out of frame with downstream GFP, producing an incorrect protein. - (2) An upstream start codon initiated translation independently of lead (Pb²⁺). Due to time constraints, a new positive control insert was designed and ordered from TwistBioscience. To expedite results, site-directed mutagenesis (SDM) was attempted on pPb7STest and pPb14STest. Two primer sets were designed to: - (1) Insert bases downstream of the non-stop codon to restore the reading frame. - (2) Change the upstream ATG to ACG to remove premature translation.
Upon transforming our positive control for testing the detection mechanism, GFP was not produced as intended. Altering growth conditions and timing did not restore expression, so we examined the insert sequences. We found: - Positive control: An unintended ATG in the spacer shifted the reading frame, likely producing an incorrect protein. - pPb7STest and pPb14STest: - (1) The riboswitch was out of frame with downstream GFP, producing an incorrect protein. - (2) An upstream start codon initiated translation independently of lead (Pb²⁺). Due to time constraints, a new positive control insert was designed and ordered from TwistBioscience. To expedite results, site-directed mutagenesis (SDM) was attempted on pPb7STest and pPb14STest. Two primer sets were designed to: - (1) Insert bases downstream of the non-stop codon to restore the reading frame. - (2) Change the upstream ATG to ACG to remove premature translation.
| Primer | Sequence (5 → 3') |
|---|---|
| Pb7S. A3293C.Fwd | ttagacagcctagatctgtttagttaattatagttcgttgaccgtatattctaa |
| Pb7S. A3293C_Ins3271A.Rev | acagatctaggctgtctaacacctctaacactctctaacggga |
| Pb14S.A3291.Fwd | ttagacagcctagatctgtttagttaattatagttcgttgaccgtatattct |
| Pb14S.A3291C_Ins3269A.Rev | acagatctaggctgtctaacacctctaacacttctctaacgacg |
Build
SDM was completed using Q5 polymerase (for more information on protocol please visit Experiments). Products from SDM were digested with varying DpnI to remove any wanted methylated template DNA (as shown in the table). A subsequent 1.0% gel electrophoresis was used to confirm SDM products prior to transformation, miniprep and third generation sequencing.
SDM was completed using Q5 polymerase (for more information on protocol please visit Experiments). Products from SDM were digested with varying DpnI to remove any wanted methylated template DNA (as shown in the table). A subsequent 1.0% gel electrophoresis was used to confirm SDM products prior to transformation, miniprep and third generation sequencing.
| Digestion Reagent | pPb7STest | pPb14STest | ||
|---|---|---|---|---|
| 1 uL of Dpn1 | 2 uL of Dpn1 | 1 ul of Dpn1 | 2 ul of Dpn1 | |
| Water | 14 uL | 13 uL | 14 uL | 13 uL |
| 10X FD Buffer | 5 uL | 5 uL | 5 uL | 5 uL |
| PCR Product | 10 uL | 10 uL | 10 uL | 10 uL |
| DpnI | 1 uL | 2 uL | 1 uL | 2 uL |
Test
Once the SDM PCR reaction was completed, the DpnI digest and 1.0% gel electrophoresis revealed that SDM was indeed successful, as a band corresponding to the size of the linearized plasmid product was observed. The products were then transformed into E. coli DH5α and allowed to grow. The transformed DpnI-digested SDM products, pPb7STest and pPb14STest, yielded only a few colonies under both digest conditions (1 µL and 2 µL). Two to three colonies were selected for overnight culture, miniprep, and third-generation sequencing. Analysis of the sequencing results, by alignment with the non-mutated ideal pPb7STest and pPb14STest, revealed that the desired changes were present only in samples where 2 µL of DpnI was used for digestion.
Once the SDM PCR reaction was completed, the DpnI digest and 1.0% gel electrophoresis revealed that SDM was indeed successful, as a band corresponding to the size of the linearized plasmid product was observed. The products were then transformed into E. coli DH5α and allowed to grow. The transformed DpnI-digested SDM products, pPb7STest and pPb14STest, yielded only a few colonies under both digest conditions (1 µL and 2 µL). Two to three colonies were selected for overnight culture, miniprep, and third-generation sequencing. Analysis of the sequencing results, by alignment with the non-mutated ideal pPb7STest and pPb14STest, revealed that the desired changes were present only in samples where 2 µL of DpnI was used for digestion.
Learn
The site-directed mutagenesis was successful in correcting the reading frame issues and removing premature translation start sites. The results demonstrated that higher DpnI concentration (2 µL) was necessary for complete digestion of the methylated template DNA, ensuring that only the mutated plasmids were transformed. This approach allowed us to quickly correct the sequence errors without having to re-synthesize the entire constructs, saving both time and resources.
The site-directed mutagenesis was successful in correcting the reading frame issues and removing premature translation start sites. The results demonstrated that higher DpnI concentration (2 µL) was necessary for complete digestion of the methylated template DNA, ensuring that only the mutated plasmids were transformed. This approach allowed us to quickly correct the sequence errors without having to re-synthesize the entire constructs, saving both time and resources.
Cycle 4 - PCR and Colony PCR Confirmation of Plasmid for Memory System Testing
PHASES
Design
To test the function of the serine integrase used in the memory system of our genetic regulatory circuit, three transcriptional units were required to be assembled: [1] a module that allowed us to determine integrase function and efficiency (pIntegraseTest), [2] a negative control to ensure that β-galactosidase activity is dependent on φBT1 serine integrase expression, as opposed to background expression or spontaneous recombination events (pLacZNegative), and [3] a positive control to ensure that β-galactosidase is functional in Saccharomyces cerevisiae (pLacZPositive). The parts were designed in SnapGene and contained BsaI overhang sites following the MoClo-YTK part type system (Lee et al., 2015). Fragments/inserts were synthesized by IDT and Twist Bioscience, arriving in pUC19 plasmids. For more information on the usage of these transcriptional units, please refer to Results. Additionally, primers were designed to target the type 1 and type 5 connectors parts to allow for the confirmation of correct assemblies. Since the connectors parts flanked the region of interest, primers which are targeted to these sites enable the PCR amplification of the cloned sequence.
To test the function of the serine integrase used in the memory system of our genetic regulatory circuit, three transcriptional units were required to be assembled: [1] a module that allowed us to determine integrase function and efficiency (pIntegraseTest), [2] a negative control to ensure that β-galactosidase activity is dependent on φBT1 serine integrase expression, as opposed to background expression or spontaneous recombination events (pLacZNegative), and [3] a positive control to ensure that β-galactosidase is functional in Saccharomyces cerevisiae (pLacZPositive). The parts were designed in SnapGene and contained BsaI overhang sites following the MoClo-YTK part type system (Lee et al., 2015). Fragments/inserts were synthesized by IDT and Twist Bioscience, arriving in pUC19 plasmids. For more information on the usage of these transcriptional units, please refer to Results. Additionally, primers were designed to target the type 1 and type 5 connectors parts to allow for the confirmation of correct assemblies. Since the connectors parts flanked the region of interest, primers which are targeted to these sites enable the PCR amplification of the cloned sequence.
| Primer | Sequence (5' → 3') |
|---|---|
| ConLS.Fwd | acaagcaacgatctccaggaccatctgaatcatgcgc |
| ConR1.Rev | ggtcatcatgcagtcatccgagcgtgtattgca |
Build
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors served as the plasmid backbone into which our fragments of interest were cloned. To confirm the correct assembly of the desired transcriptional units, PCR and 1.0% gel electrophoresis were used to amplify and visualize the insert of interest. Positive samples will be sent for third-generation sequencing to further validate the sequences.
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors served as the plasmid backbone into which our fragments of interest were cloned. To confirm the correct assembly of the desired transcriptional units, PCR and 1.0% gel electrophoresis were used to amplify and visualize the insert of interest. Positive samples will be sent for third-generation sequencing to further validate the sequences.
Test
From the numerous failed PCR and colony PCR attempts, the iGEM Guelph Wet Lab team learned a valuable lesson: the cost of attempting to troubleshoot the PCR and colony PCR reactions far exceeded the cost of sending the samples for third-generation sequencing. Therefore, the decision was made to send the samples for sequencing without prior PCR confirmation. The sequencing results revealed that the transcriptional units were assembled correctly, and the plasmids were ready for downstream applications.
From the numerous failed PCR and colony PCR attempts, the iGEM Guelph Wet Lab team learned a valuable lesson: the cost of attempting to troubleshoot the PCR and colony PCR reactions far exceeded the cost of sending the samples for third-generation sequencing. Therefore, the decision was made to send the samples for sequencing without prior PCR confirmation. The sequencing results revealed that the transcriptional units were assembled correctly, and the plasmids were ready for downstream applications.
Learn
From this cycle, we learned that troubleshooting PCR and colony PCR can be time-consuming and resource-intensive. In some cases, it may be more efficient to proceed directly to third-generation sequencing for confirmation of plasmid assembly, especially when time is a limiting factor. This approach can save valuable time and resources while still ensuring the accuracy of the constructs.
From this cycle, we learned that troubleshooting PCR and colony PCR can be time-consuming and resource-intensive. In some cases, it may be more efficient to proceed directly to third-generation sequencing for confirmation of plasmid assembly, especially when time is a limiting factor. This approach can save valuable time and resources while still ensuring the accuracy of the constructs.
Cycle 5 - Restriction Enzyme Digest Confirmation of Plasmid for Memory System Testing
PHASES
Design
To test the function of the serine integrase used in the memory system of our genetic regulatory circuit, three transcriptional units were required to be assembled: [1] a module that allowed us to determine integrase function and efficiency (pIntegraseTest), [2] a negative control to ensure that β-galactosidase activity is dependent on φBT1 serine integrase expression, as opposed to background expression or spontaneous recombination events (pLacZNegative), and [3] a positive control to ensure that β-galactosidase is functional in Saccharomyces cerevisiae (pLacZPositive). The parts were designed in SnapGene and contained BsaI overhang sites following the MoClo-YTK part type system (Lee et al., 2015). Fragments/inserts were synthesized by IDT and Twist Bioscience, arriving in pUC19 plasmids. For more information on the usage of these transcriptional units, please refer to Results.
To test the function of the serine integrase used in the memory system of our genetic regulatory circuit, three transcriptional units were required to be assembled: [1] a module that allowed us to determine integrase function and efficiency (pIntegraseTest), [2] a negative control to ensure that β-galactosidase activity is dependent on φBT1 serine integrase expression, as opposed to background expression or spontaneous recombination events (pLacZNegative), and [3] a positive control to ensure that β-galactosidase is functional in Saccharomyces cerevisiae (pLacZPositive). The parts were designed in SnapGene and contained BsaI overhang sites following the MoClo-YTK part type system (Lee et al., 2015). Fragments/inserts were synthesized by IDT and Twist Bioscience, arriving in pUC19 plasmids. For more information on the usage of these transcriptional units, please refer to Results.
| Primer | Sequence (5' → 3') |
|---|---|
| ConLS.Fwd | acaagcaacgatctccaggaccatctgaatcatgcgc |
| ConR1.Rev | ggtcatcatgcagtcatccgagcgtgtattgca |
Build
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors were used as the plasmid back bone which our fragments of insert were cloned into. To confirm the correct assembly of the desired transcriptional unit’s restriction enzyme digestion and 1.0% gel electrophoresis was used to amplify and visualize the assembled plasmids in a linearized form.
The construction of transcriptional units was completed using a one-pot Golden Gate Assembly (GGA) reaction with BsaI. The entry vectors were used as the plasmid back bone which our fragments of insert were cloned into. To confirm the correct assembly of the desired transcriptional unit’s restriction enzyme digestion and 1.0% gel electrophoresis was used to amplify and visualize the assembled plasmids in a linearized form.
Test
To determine the successful assembly of the transcriptional units, transformed E. coli DH5α from the GGA reaction were initially green/white colony screened. Three to four green colonies were selected for overnight culture, miniprep, PstI digestion, and 1.0% gel electrophoresis. The results from the gel strongly indicate that the transcriptional units were assembled correctly, as the observed band sizes align with the expected sizes (as shown below). Two colonies per transcriptional unit were selected for third-generation sequencing to fully confirm correct assembly, which aligned with the ideal constructs in SnapGene.
To determine the successful assembly of the transcriptional units, transformed E. coli DH5α from the GGA reaction were initially green/white colony screened. Three to four green colonies were selected for overnight culture, miniprep, PstI digestion, and 1.0% gel electrophoresis. The results from the gel strongly indicate that the transcriptional units were assembled correctly, as the observed band sizes align with the expected sizes (as shown below). Two colonies per transcriptional unit were selected for third-generation sequencing to fully confirm correct assembly, which aligned with the ideal constructs in SnapGene.
Figure 1: (A) 1.0% agarose gel electrophoresis of pIntegraseTest. (B) 1.0% agarose gel electrophoresis of pLacZNegative. (C) 1.0% agarose gel electrophoresis of pLacZPositive.
Learn
To further streamline our workflows and provide alternatives to using PCR-based methods of successful plasmid assembly, the usage of single-cutter restriction enzyme digest proves to be a fast and reliable method.
To further streamline our workflows and provide alternatives to using PCR-based methods of successful plasmid assembly, the usage of single-cutter restriction enzyme digest proves to be a fast and reliable method.
Cycle 6 - ortho-Nitrophenyl-β-galactoside (ONPG) Assay
PHASES
Design
The ONPG assay was selected as the primary method to assess the functionality of the integrase system. To test the system, three transcriptional units were created: a positive control, a test, and a negative control. The positive control constitutively expresses LacZ, the negative control does not express LacZ, and the test conditionally expresses LacZ in response to promoter inversion mediated by integrase-catalyzed DNA recombination. A second negative control using wild-type yeast was also included. The ONPG assay was chosen for its simplicity, low cost, and ability to provide quantitative data on reporter gene expression. The goal was to confirm integrase functionality and quantify its efficiency.
The ONPG assay was selected as the primary method to assess the functionality of the integrase system. To test the system, three transcriptional units were created: a positive control, a test, and a negative control. The positive control constitutively expresses LacZ, the negative control does not express LacZ, and the test conditionally expresses LacZ in response to promoter inversion mediated by integrase-catalyzed DNA recombination. A second negative control using wild-type yeast was also included. The ONPG assay was chosen for its simplicity, low cost, and ability to provide quantitative data on reporter gene expression. The goal was to confirm integrase functionality and quantify its efficiency.
Build
5 mL overnight cultures were prepared for all four test groups (positive control, negative control, test, and wild-type negative control), with triplicates made for subsequent statistical analysis. The optical density (OD₆₀₀) of the overnight cultures was measured, and the ONPG assay was performed following the iGEM Guelph protocol. Samples were then analyzed using a spectrophotometer to measure the β-galactosidase activity (OD₄₂₀).
5 mL overnight cultures were prepared for all four test groups (positive control, negative control, test, and wild-type negative control), with triplicates made for subsequent statistical analysis. The optical density (OD₆₀₀) of the overnight cultures was measured, and the ONPG assay was performed following the iGEM Guelph protocol. Samples were then analyzed using a spectrophotometer to measure the β-galactosidase activity (OD₄₂₀).
Test
The OD₄₂₀ readings were normalized to the initial OD₆₀₀ values to account for differences in cell density. The results were analyzed using a two-tailed Student’s t-test (α = 0.05) to determine the significance of differences in normalized β-galactosidase activity expression between groups.
The OD₄₂₀ readings were normalized to the initial OD₆₀₀ values to account for differences in cell density. The results were analyzed using a two-tailed Student’s t-test (α = 0.05) to determine the significance of differences in normalized β-galactosidase activity expression between groups.
Figure 1: Results of ONPG assay reported in normalized β-galactosidase activity Significant differences in β-galactosidase relative to yeast lacking an entry vector are designated by asterisk (student's t-test, two-tails; * = p ≤ 0.05, ** = p ≤ 0.01, *** = p ≤ 0.001).
Learn
The results from the ONP assay indicate that the serine integrase is functional in Saccharomyces cerevisiae. However, the relative activity compared to the positive control, is quite low (positive control: LacZ constitutively expressed by a strong promoter). Further testing such as a time assay is required to determine the possible factors limiting the overall activity.
The results from the ONP assay indicate that the serine integrase is functional in Saccharomyces cerevisiae. However, the relative activity compared to the positive control, is quite low (positive control: LacZ constitutively expressed by a strong promoter). Further testing such as a time assay is required to determine the possible factors limiting the overall activity.
Cycle 7 - Time-Course ONPG Assay
PHASES
Design
To measure LacZ reporter gene expression over time, a series of five ONPG assays were performed across five days. The ONPG assay was chosen for its simplicity, low cost, and ability to provide quantitative data. As in the initial ONPG assay, four test groups were included: a positive control, a test, and two negative controls.
To measure LacZ reporter gene expression over time, a series of five ONPG assays were performed across five days. The ONPG assay was chosen for its simplicity, low cost, and ability to provide quantitative data. As in the initial ONPG assay, four test groups were included: a positive control, a test, and two negative controls.
Build
5 mL of overnight cultures were prepared for all four test groups, with triplicates made for subsequent statistical analysis. Each day, cell density (OD₆₀₀) readings were taken, and cell harvests were performed according to the iGEM Guelph Protocol. Cultures were refreshed with new broth throughout the five-day period to maintain cell health. On the final day, ONPG assays were performed on cells from all five days, and the β-galactosidase activity (OD₄₂₀) of all samples was measured.
5 mL of overnight cultures were prepared for all four test groups, with triplicates made for subsequent statistical analysis. Each day, cell density (OD₆₀₀) readings were taken, and cell harvests were performed according to the iGEM Guelph Protocol. Cultures were refreshed with new broth throughout the five-day period to maintain cell health. On the final day, ONPG assays were performed on cells from all five days, and the β-galactosidase activity (OD₄₂₀) of all samples was measured.
Test
The OD₄₂₀ readings were normalized to the initial OD₆₀₀ values to account for differences in cell density. Data were analyzed using a one-way ANOVA followed by Tukey’s test and graphed using PRISM. The analysis revealed no statistically significant difference between the test and negative control groups. Additionally, the positive control displayed considerable inconsistencies over the five-day period.
The OD₄₂₀ readings were normalized to the initial OD₆₀₀ values to account for differences in cell density. Data were analyzed using a one-way ANOVA followed by Tukey’s test and graphed using PRISM. The analysis revealed no statistically significant difference between the test and negative control groups. Additionally, the positive control displayed considerable inconsistencies over the five-day period.
Learn
The results of the ONPG assay performed on the positive control group indicate human error. The positive control consistently expresses the reporter gene; therefore, a graph of LacZ levels from this group should appear uniform. However, our results show that the level of LacZ expression in the positive control fluctuated sporadically over the course of five days. The disparity between the expected and experimental results suggests an issue with the protocol. Moreover, some replicates were contaminated, meaning not all test groups had triplicates for every day. The results from the positive control indicate a problem with the experimental design or execution, and the loss of data points due to contamination means that the data collected from this assay cannot be used to support or refute the functionality of the integrase system. In the future, careful attention to sterile technique must be maintained during experiments. Additionally, increasing the number of replicates would help mitigate the effects of contamination and allow for better identification and removal of outliers during data analysis.
The results of the ONPG assay performed on the positive control group indicate human error. The positive control consistently expresses the reporter gene; therefore, a graph of LacZ levels from this group should appear uniform. However, our results show that the level of LacZ expression in the positive control fluctuated sporadically over the course of five days. The disparity between the expected and experimental results suggests an issue with the protocol. Moreover, some replicates were contaminated, meaning not all test groups had triplicates for every day. The results from the positive control indicate a problem with the experimental design or execution, and the loss of data points due to contamination means that the data collected from this assay cannot be used to support or refute the functionality of the integrase system. In the future, careful attention to sterile technique must be maintained during experiments. Additionally, increasing the number of replicates would help mitigate the effects of contamination and allow for better identification and removal of outliers during data analysis.
References
[1] Krishani, M., Shin, W. Y., Suhaimi, H., & Sambudi, N. S. (2023). Development of Scaffolds from Bio-Based Natural Materials for Tissue Regeneration Applications: A Review. Gels, 9(2), 100. https://doi.org/10.3390/gels9020100
[2] Liu, C., Xia, Z., & Czernuszka, J. T. (2007). Design and Development of Three-Dimensional Scaffolds for Tissue Engineering. Chemical Engineering Research and Design, 85(7), 1051–1064. https://doi.org/10.1205/cherd06196
[3] Guastaferro, M., Baldino, L., Reverchon, E., & Cardea, S. (2021). Production of Porous Agarose-Based Structures: Freeze-Drying vs. Supercritical CO2 Drying. Gels, 7(4), 198. https://doi.org/10.3390/gels7040198
[4] Millipore Sigma. (n.d.). Agarose: Properties and Research Applications. Millipore Sigma. Retrieved May 5, 2025, from https://www.sigmaaldrich.com/CA/en/products/chemistry-and-biochemicals/biochemicals/agarose?srsltid=AfmBOor0-clvHaNQABFQvAFMHQvZe6ia9ek_M-hBtjlM1gj3PMx43-jv
[5] Speranza, B., Corbo, M. R., Campaniello, D., Altieri, C., Sinigaglia, M., & Bevilacqua, A. (2020). Biofilm formation by potentially probiotic Saccharomyces cerevisiae strains. Food Microbiology, 87, 103393. https://doi.org/10.1016/j.fm.2019.103393
[6] Nozaka, S., Furukawa, S., Sasaki, M., Hirayama, S., Ogihara, H., & Morinaga, Y. (2014). Manganese Ion Increases LAB-yeast Mixed-species Biofilm Formation. Bioscience of Microbiota, Food and Health, 33(2), 79–84. https://doi.org/10.12938/bmfh.33.79
[7] Emami Moghaddam, S. A., Harun, R., Mokhtar, M. N., & Zakaria, R. (2018). Potential of Zeolite and Algae in Biomass Immobilization. BioMed Research International, 2018, 6563196. https://doi.org/10.1155/2018/6563196