Development of the Luminescence Detection Device
Introduction
Based on the project, we knew that we had to use some sort of hardware to detect the bioluminescence from bacteria in the presence of the mycotoxin produced by F. Oxysporum. As the procedure of the research study(Reference) we used as a reference follows, they utilized a phone camera. If we were to implement this, it would require constructing a software that can detect very small amounts of light emitted from a soil sample in a dark container using a camera equipped with an NREA algorithm.
We had an extensive brainstorming process about developing an app, and we determined it to be too complicated, and we decided to use a raspberry pi with a camera as it was easier to work with. We believed the raspberry pi light detector to not be sensitive enough. We understood that this model of camera was not the best for our project, as using a phone camera would’ve likely provided more sensitivity and potentially easier deployment, however using others were beyond the scope of our project due to the difficulty of working with them, considering that an NREA algorithm had to be used to process the very low amounts of light that needed to be detected.
Our final hardware solution is much easier to deploy than using a cell phone application, and as such is a better fit considering the scope of our project.
Considerations
Since our project targets mainly rural areas, and potentially people of low income, we wanted to design a hardware that is robust as well as simple to use.
We initially planned on using a cell phone camera to detect the bioluminescence from the bacteria. This would’ve been easier to eventually deploy in the agricultural space as well as potentially providing more sensitivity. However, we eventually opted to use a Raspberry Pi camera because we believed developing an app to be impractical at the scale we were operating on.
The Raspberry Pi camera ended up working really well because of its practicality. The camera is relatively inexpensive, which means it could still be accessible to banana farms, although potentially less practical than a cell phone would have been. The largest benefit to the Pi camera was its ease to work with and program, allowing bioluminescence to be detected with a single simple script, without the need to develop a fully fledged app with a UI.
Design
Our first step was to determine the light sensitivity of the camera(the amount of light detectable by the raspberry pi), for which we developed a small test using a mix between solutions of the substrate luciferin and the enzyme luciferase. These would be the same chemicals produced by the genetic circuit, however, due to time constraints, the actual bacteria were not available, so we used these enzymes and substrates to simulate the genetic circuit. We would’ve liked to complete it using the actual bacteria and an actual soil sample if time allowed.
For the test, we made a box out of cardboard which was wrapped with aluminum foil and sealed with black electrical tape. We made an opening on the side to fit the Raspberry Pi camera as well as a hole on the top of the box to place the test tube to carry the sample. The area of the tube exposed to natural light is wrapped with black electrical tape and the tube itself is taped onto the box. The goal was to create a pitch black environment which allowed the light emitted from the luciferase-luciferin reaction to be detected and limiting the amount of outside light and therefore the degree of inaccuracy in the lab.
Materials
- Cardboard
- Tape
- Aluminium foil folds
- Test tube
- Luciferin
- Luciferase
Conclusion and Future Improvements
The box’s design is functional, but could be improved to make it more robust, which could work to improve the accuracy of the experiment. In the future it would be better to use a more robust laser cut wood design instead of cardboard. This could also improve accuracy by further limiting the light let into the box.
Additionally, a reflective material could have been pasted onto the inside walls of the box in order to enhance the amount of light collected. This might have increased noise but could’ve been helpful to increase sensitivity.
Finally, the camera and box design is not easily deployable to rural banana farmers, although the modern, corporate nature of banana farming might make that less of a challenge.
Overall, our hardware was successful in detecting luciferase bioluminescence in a very controlled environment. However, time constraints made it so that we were unable to prove the accuracy of our hardware at detecting bioluminescence from actual bacteria in an environment akin to that in actual banana plantations. Further testing is needed to prove the effectiveness of our hardware at its intended purpose, however preliminary testing looks promising.