Modelling & Prototype Journal

Week 1: May 5th - May 9th

Modelling team formed. The big idea discussed during the week was to digitalize the diagnostic test using Arduino microcontroller or Raspberry Pi. The idea was based on glucose meters with the potential for disposable strips with the device to reduce cost per test. We looked into the advantages and disadvantages of Arduino vs Raspberry Pi and how they operate. At this point of time, Arduino seems to be a more suitable choice for our purposes. We looked into how to program Arduino and started writing simple programs to familiarize ourselves with microcontroller. As part of the digital reporter system, we are looking into using a sensor to detect a change from the biological reaction. As part of the research we did this week, we found that there are Arduino sensors that for light, current, methane, ethanol, CO2, propane, butane, and H2. We also researched the basics of microfluidics and the potential application this could have with our designs for the project. Additionally, everyone on the team participated in workshops on DNA and cell biology, PCR (Polymerase Chain Reaction) and PCR optimization, DNA extraction, cloning and transformation, bioinformatics, theory of proteins, protein purification, and protein optimization to get everyone familiar with basic lab techniques and biology theory that we potentially require during the project.

Week 2: May 12th - May 16th

This week we focused more on getting familiar with Arduino. In particular, we were working with Arduino Uno. One of our team members wrote a few programs for Arduino to make the light blink in different patterns. In order to program the Arduino, a programming language very similar to C++ is used; all of the members of the modelling team have C++ knowledge. We also found out that if we have to measure the input current, we wouldn’t need a sensor as analog in on the Arduino can measure the input voltage so we would be able to calculate the current. We also looked at the prices for different sensors and where we can buy them, most of the sensors can be purchased for under 5-7 dollars. (https://solarbotics.com/product/cds/) The Arduino Uno itself costs 35 dollars (https://solarbotics.com/product/50450/). We also checked resources left by last year’s engineering/modelling team and found resistors, capacitors, op-amps, instruments and additional resourceful items that we can use this year as well. Part of our mission is to produce an economically sound low cost device for developing countries and therefore we plan on researching how to adjust our design to reduce cost next week. It is important to consider the economics of the device if it is to be considered for practical use in these areas of the world.

Week 3: May 19th -May 23rd

This week we started looking into what kind of software we can use for modelling purposes. We decided to use Autodesk Maya to visually model the processes that are occurring within the device. We have access to Autodesk software and connections with knowledgeable advisors who would be able to help us we have any issues that arise. There are also plugins for Maya that can be used for modelling biological parts such as proteins or enzymes. This week we also continued exploring and researching Arduino’s functions as well as continuing to look into different designs of a device and what we would use to produce a low cost but effective device. Potential ideas have been to have a device that uses disposable strips for individual tests to reduce costs, while other models suggest effective technology such as microfluidics. In coming weeks we plan to continue research and talking to professors about the benefits and drawbacks of these potential prototypes. Using quantitative modelling in the future we will be able to effectively design such a device so that it operates at its greatest potential.

Week 4: May 26th - May 30th

This week was primarily focussing on the use of Autodesk Maya and learning the functions of the program. Initially Maya 2015 was downloaded, however due to compatibility issues 2013 might be downloaded in the coming weeks to work effectively with the various plugins. Tutorials were performed to learn the basic functions of modelling and animation of the program. Using this knowledge we would be able to eventually use this programming to model the proteins of the project. Additionally we continued to look into potential sensors that could be compatible with the Arduino boards and what other designs could be used for a device. After all team members are familiar with using Maya for modelling, we are planning on using the animation features to visually illustrate the biological processes that are occurring within the device when in use. This will help to better explain the project through visual aspects and can cause a greater curiosity and interest in the device. Learning all of the skills of Maya will be an ongoing process throughout the summer, however a focus was placed on developing basic skills this week.

Week 5: June 2nd - June 6th

This week we continued using Maya for modelling and animation purposes. We also began to explore and refresh our memories using Autodesk AutoCad and Inventor for design purposes, as well as downloading matlab for quantitative modelling purposes. Although we currently don’t have all of the necessary data for some of the proposed models, we are currently examining the software to see how we would create these models when the data becomes available. We are continuing to gain more knowledge of Maya, Autocad and Matlab we will be prepared to create quantitative and visual models when the data becomes available. This week we also started looking into mathematical modelling for biological systems and gaining background knowledge in the field through literature. In particular, we looking into basic definitions, process of modelling, relation between state variables and description methods, time constants for cellular processes, fundamental of enzyme kinetics, cell division and growth. By studying and learning proper methods of quantitative analysis we will be able to directly perform this aspect of modelling as soon as data becomes available. We began brainstorming what important aspects of the project we can quantitatively model and what would be useful to help us in the initial design phase. As of next week, a proposed sensor will be chosen and then we will resume our process with the Arduino microcontroller and writing programs to control the device. Additionally we seeked resources regarding the potential of using a voltage or current sensor with the device and are continuing our research for this option. Literature References: A. Kremling, Systems biology : mathematical modelling and model analysis, CRC Press, 2014

Week 6: June 9th - June 13th

This week we contacted two microfluidics experts to get insight on microfluidics and to answer some of the questions we have. Dr. Karan Kaler replied to us saying that he is away at the moment, but will be back in July and will be willing to meet up with our team to assist with the project. He said that their research lab is currently utilizing droplet based microfluidics for the detection of human pathogens using chip based real time qPCR technology developed in their laboratory. He thinks it may be relevant to us and suitable for us to explore. This week we also continued working with Maya and Matlab software. Dave submitted a request so we can get SimBiology toolbox for Matlab that will be very useful for quantitative modelling. We also got the 2013 version of Maya instead of 2015 so that we can get ePMV plug-in that can be used for modelling and animation purposes. After installing the software, we watched tutorials to get familiar with how the plug in works. Unfortunately, so far the tutorials we found were not very useful, so we might have to look for more resources. Also, we reviewed the process of transformation and watched existing animations so we can create our own transformation animation in the near future for practicing purposes. We also found an online book-tutorial by Autodesk for Maya which is so far the best teaching resource for Maya. It explores the functions through examples. It also includes the tutorial on Dynamics and a variety of plug-ins such as nParticle which might be relevant for our project. It is a relatively long book so it would take at least a week to go through it, but it is extremely useful for getting advanced with Maya functions. This week we also arranged regular meeting with Dr. Nygren which will take place every Thursday morning so we can discuss the progress of the engineering/modelling team. We also signed up for Matlab workshops which will take place on Wednesday, June 18th. Next week we are also planning on contacting Cesar from Autodesk to further discuss Maya applications and software, and potential skype meetings in the future.

Week 7: June 16th - June 20th

This week consisted of further looking into Matlab applications and data analysis as well as looking further into what materials we will need to purchase to begin modelling with the Arduino board. On the 18th we attended a matlab workshop and information session that was focused on proper data analysis and representation. Additionally we have begun weekly engineering meetings focussed on quantitative modelling and the state of our prototyping. We have begun to look into how we can use quantitative analysis for modelling purposes, in particular, to idealize the device, optimize its materials, dimensions and components as well as use this data and analysis to reevaluate and make changes to our prototype device. At our meeting with Dr. Nygren we further discussed the potential add on to the device in which the results could be easily read or documented on a smartphone or computer. Although this aspect of the design would not be designed for rural areas in the developing world, there is potential of the modular device to have a worldwide impact in not only hospitals in the developing world with more resources, but the developed world as well. The proposed chosen reporter is a fluorescent protein such as Red Fluorescent Protein (RFP) or Green Fluorescent Protein (GFP). Another option for the reporter is lacZ. Using a light sensor, and potentially a filter to ensure we are not detecting ambient light we will detect the presence of light and convert it to a voltage that the Arduino can detect and display a result. This will most likely be designed as a closed environment in order to eliminate other sources of light that could potentially trigger the sensor. The concept of light dependent resistors was also discussed, however after further research it was discovered that the level of sensitivity of these components may not be at the level necessary for the device. Although we do not yet have specific numbers on how luminescence the reporter is, we are not currently pursuing this proposed option. We are planning on purchasing an Arduino and several light sensors to begin testing and quantifying the effectiveness of the different components. In the next week we are hoping that the proposed reporter will be able to test that part of the system and determine which sensor and process will be the most effective to determine if the signal is present. Additionally the modelling team prepared a presentation on light emission and potential it has in our project and how we are planning on designing the higher level device with a light sensor and a digitized device. In parallel to the research regarding the reporters and the light sensors we have been continuing to explore Maya and Matlab ensuring our skills will be ready in time for application of lab data.

Week 8: June 23rd - June 27th

This week we continued an extensive search into filters and light sensors and any other necessary components we need to purchase such as the Arduino microcontroller. We have also began designing experimental conditions for testing the brightness of the chosen reporter and the sensitivity of the light sensors itself. After this testing occurs of the idealized brightness of the reporter and the sensitivity is determined, we will be able to idealize the device and the amount of reporter that is necessary to trigger the light sensor and consequential voltage. We continued to do research into different options of sensors, and what we would additionally need for our device circuit including resistors and the microcontroller itself. As it comes to light filters, the three main types (used in robots) are coloured gel filter, interference filter, and dichroic filter. Coloured gel filter is made by mixing dyes into a plastic base. Depending on what dye is used, only a certain band of wavelengths can pass by. It is the cheapest out of the two, but the least accurate one. Interference filter is made of several chemical layers. Each layer blocks a certain range of wavelengths so only a very small range of wavelengths can pass by. It is more accurate than gel filter (because the range of wavelengths allowed through can be made really small), but expensive. Dichroic filter is made of organic dyes and other chemicals. It absorbs light at certain wavelength. Again, it is better than coloured gel filter, but expensive. We looked more into interference filters and rough calculations show that if we want a wavelength of 609 nm to pass, the material in the middle of the filter will need to have an index of refraction (n) = 1.35 and the thickness would be 225.55555 nm. Possible materials include 10 percent glucose solution in water ( n = 1.3477), 20 percent glucose solution in water (n = 1.3635), Teflon ( n = 1.35 – 1.38), and ethanol (n = 1.361). We wanted to commence testing as soon as possible and therefore have put in several purchase orders for products. After more research into light sensors and potential filters we were able to locate affordable and applicable materials that we will be able to use to build our device. Firstly we bought an Arduino Uno microcontroller for the team as the one we have currently been programming personally belongs to a member of our sub-team. We also ordered an educational kit for Arduino Uno that includes educational materials as well as useful parts for the construction of our future circuits such as resistors, capacitors, sensors, LED lights, and other. Additionally we have purchased a Parallax ColorPAL-Color and Light Sensor instead of a simple light sensor as we came to the realization that it would be difficult with only one sensor to differentiate between multiple diseases if all of them use RFP as the reporter. Initially, our idea was to excite RFP with yellow light (or UV light is another option) and detect the emitted light (609 nm wavelength) using a light sensor and a filter. However, if the reporters indicating the presence of different diseases were that of different colours, the a colour sensor would be able to directly differentiate. We also ordered two temperature sensors so we can keep our device at constant temperature and an 3W LED light if we need to excite the RFP. Our team is planning on formally drafting an initial design of the device after the meeting with our microfluidics expert to determine the features that this technology could impact the appearance and logistics of our project. We found an bandpass filter that allows light of wavelengths between 604 and 616 nm to pass through (the emission wavelength of RFP is 609 nm). The filter could be used in conjunction with the colour sensor, or with a basic light sensor and LED’s. It is going to take various tests and trials to determine which option is the most logical and successful and will provide accurate results for this important test. In order to construct our circuits, we purchased a soldering station which will potentially be useful for teams that come after us. In regards to the beginning of quantitative modelling we are breaking the system into smaller parts and starting to model each small part with the potential of combining all of these parts to model the entire system and the parameters. We are going to being with the reporter RFP which has already been plated and is ready to undergo testing, we are currently waiting on our purchase orders with our materials to see if the colour sensor is able to accurately detect this colour, as well as determine the true sensitivity of the device so that this is not a limiting factor. Additionally we are going to look at every reaction and process taking place biologically in the system and start dissecting the system and working backwards from the reporter. After this has been completed we are planning on finding the most sensitive components that will have a positive impact on time and are the “bottleneck” of the device. Using these discovered factors we are going to apply them to the system and measure to see if any change has taken place and if our model is an accurate representation of the mathematical processes occurring. Additionally we are are planning on looking more literature of modelling biological systems and into some of the build in features of the Simbiology toolbox. Our plans from next week consist of contacting another lab that has a plate reader that will be able to automatically measure absorbance and RFP cell count in the bacteria culture transformed with RFP. This will give us initial data from the reporter to begin working with.

Week 9: June 30th - July 4th

We started this week by exploring how our color sensor operates, so when we receive the component, we can begin testing immediately. Additionally, we discovered a tutorial on using ColorPal sensor with Arduino Uno which included the sample code. (http://learn.parallax.com/colorpal-arduino-demo). In order to easily convert the output of the sensor to colors, we downloaded ColorPal Color Matching Program. The program additionally has the capability to assist with calibration of the sensor when detecting black and white surfaces. The sensor has its own LED light and it shines red, blue, and green light onto to the surface. Relative and respective amounts of red, blue, and green reflected back indicate the color of the surface, although the reflected light is mixed with the LED light. This effect can be eliminated by sampling first without any LEDs turned on, and then subtracting this reading from each of the red, green, and blue components. We also started designing a formal experiment for testing the color sensor. We began brainstorming aspects of the sensor/device that it was very important to both quantify and optimize the operation of the sensor itself: 1. What is an optimal distance between the surface and the light sensor? 2. How does color intensity effects measurements? (example: bright red vs faint red) 3. What is the minimum amount of transformed bacteria colonies needed to produce enough color to be detected? 4. How does ambient light and the environment affect the calibration and readings? We started writing up a more formal procedure for the experiment to begin to determine answers to these questions, after we are able to get the sensor functionally working. Next week we are hoping to perform a test using an accurate plate reader and the RFP to take a fluorescence reading every 30 minutes overnight of different concentrations of cells. After this data is obtained, we are planning on creating a graph and fitting a mathematical model to describe the results. This data will allow us to determine how to achieve an ideal signal and what we need to aim for when constructing other portions of the device. Once we get the sensor, we can perform similar measurements with an Arduino and colour sensor to ensure the results that we achieve using our device will match the results collected using a plate reader. Using the “gold standard” of the plate reader and the experimental sensor results will allow us to jumpstart our quantitative modelling. Additionally we broke down all of the processes and reactions that are taking place within our biological system in our device and are hoping to model each reaction individually - with the optimistic hope that we will eventually be able to model our entire system. In particular, we determined some of the parameters that we would have to determine with our models through this experiment. To collect a more vast background knowledge on quantitative modelling of biological systems we read literature recommended by Dr. Nygren. This week we also started getting familiar with Arduino microcontroller. We built simple circuits to make an LED light blink in different patterns and how to change the brightness of the light in response to being stimulated (LED will dim in a brightly lit room, and will increase brightness in a poorly lit room). We were additionally able to contact an electrical engineering lab technician that has had experience working with Arduino and our colour sensor to ensure our circuits and code for the programs we are planning on writing next week are the most efficient and effective they can be. Additionally, we continued with Maya tutorials to ensure our skill levels are ready and Simbiology toolbox for Matlab. Literature References: C. J. Myers, Engineering Genetic Circuits, CRC Press, 2009 U. Alon, An Introduction To Systems Biology: Design Principle of Biological Circuits, CRC Press, 2006

Week 10: July 7th - July 11th

Insert Text Here

Week 11: July 14th - July 18th

This week started with looking more into our Arduino - sensor set up. The colour sensor works for the most part, but every once in a while it would not turn on or would turn off when the serial window is opened in Arduino Programming Environment. Unplugging the Arduino and plugging it back in helps to get the sensor working again. The fact that the sensor is not working every single time suggests that there might be impurities in our circuit or code. We will be contacting lab technicians at Schulich School of Engineering that work with Arduino to arrange the meeting to discuss the problems with our set up. Our C++ program that outputs the colour of the sample is working fine, but we will need to integrate it into the code for Arduino, so that our system can do all of the necessary calculations. So we have been looking into Arduino code to find out what changes we need to make to our current C++ code to make it run on Arduino. We will also able to connect the LCD display to Arduino and make the display show “Hello World!”. As of now, all of the components of our future circuit are working (the components include the colour sensor, temperature sensor, and the LCD sensor with connections to Arduino), but certain improvements are required. The next step includes exploring the LCD code and figuring out how to make it output something more complex. Then we will start on putting all components together in one circuit. This week we also met with a microfluidics expert to get his insight on our idea to use microfluidics in device’s design. He suggested we use microtubules in our design rather than considering complicated microfluidics set up. Syringe can be used to pump the liquid across the device. We have come up with a basic outline of how we see our device’s design. The best material suggested for the tubules is Teflon. We are looking into getting necessary supplies for building the prototype for testing. Mechanical Shop at Schulich School of Engineering will be helping us to build a prototype. There will be a simplified design for the testing purposes - the syringe will be taken off to get all of the necessary components into the device and put back together. Since the tests are taking place in the lab, there are less concerns about contamination.The results received from a plate reader experiment do not meet the expectations - it seems like no or little cells grew over the night. After some discussion, it was decided to repeat the experiment next week. This week we also started preparing a 20 minute presentation for the rest of the team on what the engineering subgroup has achieved so far. This presentation will also be a good practice for the presentation at Giant Jamboree.

Week 12: July 21st - July 25th

Insert Text Here

Week 13: July 28th - August 1st

Insert Text Here

Week 14: August 5th - August 8th

The Colour Sensor Testing Plate was left to grow over the long weekend, so it was ready for testing on Tuesday this week. However, the colourrs look rather faint. When measurements were taken, there was hardly any difference between the blank square and all the other squares. The measurements seem to be better, but are still inconclusive as it is hard to tell the difference between just blue colour present and blue and red colours present. More plates need to be grown to see if the measurements would be consistent. The plate was also contaminated with fungus. This week we set up a meeting with Cesar Rodriguez, Senior Research Scientist with Autodesk Research, to discuss our project and to consult him about Maya modelling. Our existing device animation is at a presentable level and we can polish it later on. At this time, we have to start on the animation for the biological process taking place in our device. We contacted a member of last year’s team to get get help with getting started with ePMV plug in for Maya. We will also set up a meeting with the team members working on the biology part of the project to go over every detail in the biological processes in the system. From now on, we will have meetings with Cesar every week. Next time one of the topics to discuss is wetware schematics.

Week 15: August 11th - August 15th

Insert Text Here

Week 16: August 18st - August 22nd

Insert Text here - including the graphs or not?

Week 17: August 25th - August 29th

This week we presented on our work to the rest of the team. Overall, the presentation went well. We received lots of feedback on the 3D animation of our device from the team and Cesar. Little mistakes were pointed out and it was suggested to change the part with DNA processes to show the interaction between the repressor gene and the promoter/operator/reporter genes. In particular, it might be better to show that the repressor gene codes for the repressor proteins and transcription cannot happen while repressor proteins are bound. So this week we have been working on improving our Maya animation. We also talked to Cesar about creating a schematic to describe our system. He suggested a couple of software options we can use for the schematic. We also started working on writing up content for the wiki and made plans for future weeks.

Week 18: September 1st - September 5th

This week we have been working on improving our Maya animation. In particular, we simplified and improved the transformation animation. We also made the DNA pieces appear more clear and organized. The scene of the animation with DNA was also changed to show the interaction between the repressor gene/repressor proteins and the reporter circuit. This week we also continued writing up content for the wiki and editing the journal so we can start uploading content to the website. We also downloaded Microsoft Visio software so we can start working on the schematic for our biological system.

Week 19: September 8th - September 12th

This week we started uploading our content to the wiki so we can design and prepare for how we want to display our information. We collaborated with the various biology sub-teams to create a rough outline of our schematic. This week we were also preparing engineering content for the aGEM presentation, and determining which information was important to be communicated in our project at aGEM. Additionally we also practiced answering potential questions we might receive. We received valuable feedback regarding the alignment of the detection chambers in our prototype. If we place these chambers radially and symmetrically to the PCR chamber we automatically reduce the amount of Teflon tubing required to transport the sample from the PCR chamber to the detection chambers. This allows us to make a device that is both more economically feasible and more portable through the smaller design. Additionally we received feedback regarding the heavy base of our prototype. We wanted to ensure that during the development phase that we would have a device to showcase aesthetics and the fact that it was operational for several months through many tests. When creating a new design we plan on altering this and planning for a lighter, less expensive base, as tests will have to undergo on test, instead of multiple over several months.

Week 20: September 15th - September 19th

Initial draft of the schematic was prepared and sent to Cesar and the rest of the team for review. It was concluded that the schematic is currently not effective at showing the repression/expression of the reporter protein and does not show the interaction between the repressor gene and the reporter. Through exploration of different designs, arrangements and symbols we hope to be able to show our complicated system in a manner that can be understood by many different perspectives. Additionally we have begun cost analysis of our device. It is important to recognize that development costs will be much larger than the final cost of a device we are planning on producing. We are hoping to receive a cost breakdown of the specific costs of each material within our device, however for the time being we have calculated the dimensions of the various parts. Bulk prices of materials such as Teflon were searched with the goal of

Week 21: September 22nd - September 26th

This week we brainstormed a way to show the interaction between the repressor gene and the reporter gene. A new version of the schematic was prepared and discussed with Cesar and the rest of the team. Overall, the schematic is looking better, but there still are details that need to be changed to make the schematic as clear as possible. Additionally we have been working to render new images of our device with three chambers (aligned at 120 degrees) radially and symmetrically. This consists of both a positive and negative control and a disease test for proof of concept. This design can be altered or changed for as many detection wells needed depending on the task of the device and the resources available to build it. The amount of chambers can be adjusted as along as they are symmetrical with the PCR chamber as demonstrated by fluid flow analysis on Solid Works.

Week 22: September 29th - October 3rd

This week we continued uploading our work to the wiki. We also set up a vote to choose a team member to do a voice over for our animation so that this visual process can be narrated with the necessary background information and description of the process occurring. This week we also continued working on the design of a schematic to show the basic process of our system and how it operates. Additionally with our new design, we have began a cost analysis comparing the original device manufactured with our rendered picture of our new design and the changes we would make to the new design. Although collectively as a team we chose not to produce another prototype, we are still planning on doing a full analysis and justification of the chances were made and how this is more economically feasible and positively impacts the end-user. We began by looking up the bulk cost of many of the materials that were used, such as Teflon and calculating the dimensions of the new device. Although we have not received a full cost-breakdown of the device, we are trying to calculate the impact of mass production versus development cost would have on the device. We plan on analyzing all materials that were used in the creation of this device and making a plan with new dimensions to compliment the new design

Week 23: October 6th - October 10th

This week we worked on the schematic. After receiving feedback from the rest of the team, we decided to make the schematic flow horizontally from left to right rather than vertically. We also worked on writing a script for the narration for our 3D animation of the device. A sample narration was recorded to show the team. Minor changes to the script were suggested to ensure that the description accurately and concisely describes the illustrated process. Next week we will be making a final recording in collaboration with the narration and upload this final version to the wiki.

Week 24: October 13th - October 17th

This week we worked on adding our final information to the wiki and finalizing the website content and design. We also recorded the narration for 3D Maya animation to explain how our system works with our visual modelling. The animation was finalized and added to the wiki.

Works Cited: