Considering that our engineered bacteria employ a lysis-release strategy, the bacterial population may significantly decrease in a short period when there are drastic changes in the external environment or large-scale nematode infestations. Therefore, we aim to develop a device to detect the number of engineered bacteria in the soil. The goal is to assist farmers in non-laboratory settings in supplementing the application of biopesticides, ensuring that farmland maintains long-term resistance to nematode invasions.

We plan to express a Flag-tag fusion protein on the surface of the engineered bacteria and prepare colloidal gold test strips to detect the presence of the Flag tag. To achieve our goal, we have designed a set of devices that includes a dilution module, a peristaltic pump, and a detection module. The dilution module enables quantitative dilution of soil samples, the peristaltic pump delivers the sample liquid in a precise manner to the colloidal gold test strip, and the detection module assists with image capture using a smartphone.

Dilution module

The main function of this module is to dilute the collected soil samples with water for subsequent colloidal gold detection experiments to assess the concentration of engineered bacteria in the soil.

Figure 2. Water inlet end cap.

Figure 3. Intermediate cylinder
containing discharge pipe and window.

Figure 4. Connector with thread.

Figure 5. Rotary paddle mixer with holes.

Figure 6. Driving base.

This dilution module consists of five key parts.

1. The first part is water inlet end cap,this component is used to introduce water into the system, ensuring precise addition of the required liquid during the sample preparation process.

2. The second part is intermediate cylinder containing discharge pipe and window. The discharge pipe allows the diluted liquid to flow out for subsequent detection or analysis. The window, marked with reference lines, enables real-time monitoring of the liquid level, allowing users to accurately assess the degree of dilution

3. The third part is connector with thread,this threaded connector links to the peristaltic pump hose and includes a built-in filter. The filter is designed to remove solid particles from the liquid, such as soil or other impurities, ensuring the purity and accuracy of the sample for subsequent analysis.

4. The fourth part is rotary paddle mixer with holes,this stirrer features a propeller-like design and perforations to enhance fluid flow. Its rotational motion ensures even mixing of soil and water, avoiding separation and preventing dead zones during the process. This makes it ideal for handling liquids of varying viscosities and concentrations, ensuring thorough mixing for different soil types.

5. The fifth part is driving base,this component powers the rotary paddle mixer, driving the stirring motion to ensure uniform dispersion of particles and a well-mixed solution.

Peristaltic pumps

The peristaltic pump is a key component for liquid transfer and precise control. The difference between our designed peristaltic pump and traditional ones is that we have increased the size of the rollers, allowing it to simultaneously squeeze five silicone tubes. This design is suitable for repeated experiments, enhancing the scientific reliability and accuracy of the results.

Figure 8. Pump casing.

Figure 9. Driving device: press roller, rotor and motor.

Figure 10. Hose.

Our designed peristaltic pump can be divided into three parts: the pump casing, the drive mechanism, and the tubing. The pump's casing and rollers tightly compress the tubing, and as the rollers rotate, they alternately compress and release the tubing to pump fluid. With the rotation of the rollers, the tubing returns to its original shape, creating negative pressure at the pump inlet, which drives the liquid flow and fills the tubing, thus achieving fluid transfer. By controlling the parameters of the stepper motor in the drive mechanism, we can adjust the pump's flow rate for precise liquid measurement.

Detection module

This module is designed to assist smartphones in capturing image data from colloidal gold test strips. After placing the prepared colloidal gold test strip into the "Sliding groove", the user quantifies and drops the diluted liquid into the "Sliding groove" through the peristaltic pump and pushes it into the "Closed test box". The cover is then closed to prevent external light interference or contamination. After a 15-minute reaction time, a color reaction will appear on the colloidal gold test strip, indicating the test result. At this point, the user only needs to align the smartphone camera with the specially designed camera hole on the cover and take a photo through the aperture.

Figure 12. Camera hole
optical window cover.

Figure 13. Closed test box.

Figure 14. Sliding groove.

The detection module consists of three components:

1. The first component is the optical window cover of the camera hole, whose main function is to provide an accurate shooting channel for the smartphone camera. The optical window cover is equipped with a circular aperture that matches the size of the phone's camera, ensuring that the camera can accurately align with the test area during photography. Below the cover, a high-efficiency LED lighting system is integrated, providing uniform illumination to ensure that the colloidal gold test strip is fully and evenly lit during the photo shoot. This reduces the impact of insufficient light or excessive reflection on image quality, thereby improving the precision and clarity of image capture.

2. The second component is the enclosed detection box. The detection box adopts a closed design to block external light interference, ensuring that the test strip is in optimal lighting conditions during photography. To further enhance the ability to resist light interference, the outer surface of the detection box is covered with a layer of black cardboard. This material effectively prevents external light reflections from entering the shooting area, ensuring the clarity and accuracy of the captured images.

3. The third component is the sliding groove, which features five independent slots designed to hold up to five colloidal gold test strips simultaneously. This setup enhances the accuracy of the results by repeating. Each slot is independently designed to prevent interference between test strips during the detection process.

We have learned that within a certain range, the color intensity of the bands in colloidal gold test strips is related to the concentration of engineered bacteria—the higher the concentration, the darker the bands[1]. Based on this phenomenon, we developed a local application to help users objectively analyze band intensity and estimate the concentration of engineered bacteria in soil samples.

Figure 15. Image analysis workflow.

Libraries used:

1. PyQt5: Used to create the graphical user interface (GUI).

2. OpenCV: Used for image reading, grayscale conversion, image enhancement, and thresholding.

3. NumPy: Used for matrix and array operations, particularly for summing rows of image data and analyzing band positions.

4. SciPy: Used to detect peaks in the image and calculate band widths.

5. Matplotlib: Used to visualize the results, including marking the band regions and plotting intensity profiles.

Program Logic:

The code reads the uploaded colloidal gold test strip image, converts it to grayscale, and enhances the image. It then preprocesses the image by applying blurring and contrast-limited adaptive histogram equalization (CLAHE) to improve the contrast of the band regions. The program uses peak detection algorithms to identify the C-line and T-line on the test strip and calculates intensity profiles, cumulative intensity, peak intensity, and valley intensity for each band. Finally, the results are visualized using Matplotlib and displayed through a PyQt5 graphical interface.

Application process:

Here is the interface and results of our program. After the user clicks the "Upload Image" button, they can select the image to be analyzed, and the program will perform calculations and display the results. Result Presentation 1 shows the calculated outcomes, including the cumulative intensity, peak intensity, and valley intensity of the bands, as well as the intensity ratio of the TC band. Result Presentation 2 provides more comprehensive information, displaying the boundaries of the C-line and T-line identified by the peak detection algorithm, along with the intensity vs. horizontal coordinate curve, accompanied by the calculated results.

Figure 16. Program start interface.

Figure 17. Result presentation 1.

Notes:

1. Image format: Common image formats (such as PNG, JPG, BMP) are supported, but the uploaded image must be of the front side of the test strip to ensure proper identification of the C-line and T-line.

2. Error handling: The code includes basic error handling, such as raising an exception if not enough bands are detected.

3. We recommend that users estimate the concentration of engineered bacteria in soil samples based on the T/C ratio, which helps eliminate spatial and temporal differences. The immunological reaction on the immunochromatographic test strip is an antigen-antibody kinetic process, where the color intensity of the T band and C band typically increases over time. Variations in sample concentration and testing conditions can also affect this process, leading to potential differences in T band values between batches and resulting inaccuracies.

1. First, collect a certain amount of soil samples from the specified depth around the plant root system and place them into the dilution device. Next, inject water into the device through the inlet pipe of the inlet cap until the water level reaches the first marking line in the middle cylinder window. Then, drive the propeller to fully stir the mixed liquid. After letting it stand for a period of time, observe the changes in the liquid and activate the peristaltic pump to begin operation.

2. When the water level drops to the second marking line (to be confirmed after testing the prototype) in the window, the liquid that needs to be tested has already filled the peristaltic pump. At this point, aim the hose at the other end of the peristaltic pump into the sliding groove of the detection module, and add 3 drops of test liquid onto the colloidal gold test strip inside the slot.

3. Then, push the sliding groove into the sealed detection box and close the lid. After a 15-minute reaction time, the colloidal gold test strip will have fully developed its color. At this point, the user can take a photo using a smartphone through the camera hole on the detection box lid. The camera hole has been optimized in design to ensure that the smartphone camera can focus accurately, preventing light reflection or distortion, thereby capturing high-quality, clear image data.

4. After users collect image data on their mobile phones, they can upload it to a computer and use our provided program for analysis to obtain the intensity ratio of the T band to the C band. Users can further assess the concentration of engineered bacteria in the soil sample based on this intensity ratio.

The real-time monitoring device developed in this project aims to address the urgent need for detecting the presence of engineered bacteria in soil, providing farmers with a convenient tool that can be operated without laboratory conditions. By efficiently monitoring the quantity of engineered bacteria in the soil, farmers can keep track of the soil's microbial dynamics in real-time, enabling them to supplement and apply biopesticides at the optimal time, further optimizing the timing and dosage of pesticide application. This not only enhances the effectiveness of biological control but also reduces excessive pesticide use, effectively protecting the environment.

The device, with its simple operating procedure and excellent detection performance, significantly lowers the technical barriers for farmers. More importantly, the technology offers great scalability, with the potential for deep integration with cutting-edge technologies such as artificial intelligence and automation in the future. This will drive technological innovation in agriculture, leading to the full digitalization and intelligent management of agricultural production. The widespread application of this innovative device will provide strong technical support for sustainable agricultural development, bringing tangible economic and ecological benefits to farmers.