Introduction

RGBradford is a software tool that analyzes microplate images and automatically generates a calibration curve and calculates protein concentrations.

Background

The Bradford protein assay is a widely used biochemical technique for measuring protein concentration in biological samples. This method is based on the binding of Coomassie Brilliant Blue dye to proteins, resulting in a color change from brown to blue that is proportional to the protein concentration [1]. The lab team frequently utilized the Bradford assay during the development of the "IXPOSE" project.

Motivation

This project was inspired by a practical challenge our team faced when the laboratory's spectrophotometer malfunctioned, leaving us unable to use it during a critical week of experiments. The software is primarily based on a JoVE (Journal of Visualized Experiments) paper by D. C. Moreira [2], discovered while searching for a workaround for the aforementioned issue. The paper introduces the RGBradford method – a technique that uses a smartphone camera to quantify protein concentration through RGB image analysis. It demonstrates that standard Bradford assays can be performed without a spectrophotometer by analyzing the color intensity of wells using commonly available camera technology.

Existing solutions (such as the “ReadPlate 3.0” macro for “ImageJ” [3]) posed considerable usability and accessibility challenges: confusing user interface, no web option, steep learning curve and the need to import additional plugins in order for the RGBradford method to work. This prompted us to develop a more user-friendly solution. Researchers and students, especially those in resource-limited settings, educational institutions, and small laboratories, require an affordable method to perform Bradford protein assays without investing in expensive spectrophotometric equipment, ideally using an everyday device.

Purpose

The purpose of this project is to create a user-friendly, easily accessible tool for conducting the Bradford assay experiment, eliminating the need for expensive lab equipment in areas where it may not be affordable.

Usage

Web Client

Having user-friendliness and accessibility as our main goals, the team’s software tool has been made easily available via a web client at https://rgbradford-frontend.onrender.com/.

A demonstration of the web client is available below. A video tutorial can be found in the application itself.

Local Hosting

The application is also available for download and easy local hosting in our project’s GitLab repository.

No additional software is needed, as the application was containerized with Docker for quick deployment, which automatically installs the database and needed dependencies. The only prerequisites needed:

• Docker is downloaded and running.
• Git (to clone the repository).

1. Clone the repository:

git clone https://gitlab.igem.org/2025/software-tools/vilnius-lithuania.git
cd vilnius-lithuania

2. Build and run the entire application (frontend, backend, and database) with a single command:

docker-compose up --build

3. Access the application:

• Frontend: http://localhost:3000

• Backend API: http://localhost:8080

• API Documentation: http://localhost:8080/swagger-ui.html

Planning

For the RGBradford project, we established both functional requirements (what the system does) and non-functional requirements (how the system performs) to guide development and ensure the final product meets scientific and usability standards. The requirements specification serves as the foundation for successful project execution. Defining a clear scope prevents feature creep and keeps development focused on delivering core functionality rather than attempting to build an overly complex system. Additionally, comprehensive requirements documentation enables future iGEM teams to extend functionality systematically while maintaining consistency with the original design vision.

Functional Requirements

Non-functional Requirements

Architecture

Based on the system analysis, a modular system comprising client, server, and database layers was constructed and developed, as illustrated in Fig. 1. The design focused on several key aspects:

• Expandability: The system utilizes Docker containerization for easy installation and deployment. This approach allows future iGEM teams to download and expand on it, making integration with existing synthetic biology software easier. Users who prefer not to use the web application can download the system locally with one easy command and no major prior configurations or dependencies.
• Ease of Use: Designed for users with minimal IT background, the web client makes the application accessible on both mobile and desktop devices. The application also provides a video tutorial and includes the Bradford assay protocol.

Fig. 1 RGBradford architecture diagram

Modeling

The RGBradford system implements a relational database model with a hierarchical structure mirroring the natural organization and workflow of Bradford assay experiments. The physical data model (see Fig. 2) was designed with a focus on data integrity, performance, and scientific workflow alignment.

• Lab Workflow: After consulting with the wet lab team, we broke down the Bradford assay workflow into key elements: organizing work by projects, where each project consists of a single experimental plate, and each plate contains multiple wells. The information stored in the wells’ table can be used for further analysis and calculations.
• Data Isolation: Each user's data is completely isolated, preventing unauthorized access to other researchers' experiments.
• Cascade Delete Logic: When a project is deleted, all associated plates, wells, and analyses are automatically removed to maintain referential integrity.

Fig. 2 RGBradford physical level ER diagram (database model)

Experimental Validation

The RGBradford application underwent experimental validation to ensure its accuracy and reliability as an alternative to traditional spectrophotometric methods. The system was validated by performing parallel measurements: protein samples were analyzed using both the RGBradford image-based method and a conventional microplate reader (spectrophotometer) as the gold standard reference.

Early Testing Stage

During initial validation experiments, we observed discrepancies between RGBradford and plate reader measurements, particularly for wells located at the plate's periphery.

Upon troubleshooting, we identified optical distortion as the primary source of measurement error, as can be seen in Fig. 3. Perspective distortion, which is caused by the camera’s the angle and distance of the camera relative to the plate, was the main culprit in introducing these problems:

1. Edge Effects: Wells at the plate edges experienced more severe distortion than central wells due to being further away from the camera.
2. Region of Interest (ROI) Contamination: The initial algorithm used 80% of the well diameter for measurements, inadvertently including distorted edge regions in the analysis.

Fig. 3 RGBradford physical level ER diagram (database model)

Algorithm Optimisation and Takeaways

Based on our findings, a critical modification was added to the image analysis algorithm to exclude edge defects from selection by reducing ROI (Region of Interest) diameter. Early testing also helped to identify the best practices for taking photos of the Bradford assays, which we outlined in our project’s GitLab repository’s README.md file [4].

Additionally, experimental validation helped to understand the software's limitations and outline improvements, which could be implemented in future versions of the project.

Final Testing Stage

After optimizing the software tool to version 1.0.0, it was once again compared to the results of a spectrophotometer.

This time, the results were significantly better. The assay was conducted on three different samples, which can be seen in the table below. We identified these key improvements:

• Strong correlation between RGBradford and plate reader measurements, a notable improvement over the previous version's limited results.
• Significantly improved accuracy, with edge discrepancies reduced to acceptable ranges for most wells.
• Comparable standard curve R² values between both methods.

API Documentation

Limitations and Suggested Future Improvements

References

1. W. B. I. L. U. Clara L Kielkopf, "Bradford Assay for Determining Protein Concentration," Cold Spring Harb Protocols, vol. 2020, no. 4, 2020.
2. D. C. Moreira, "RGBradford: Protein Quantitation with a Smartphone Camera," Journal of Visualized Experiments, vol. 199, 2023.
3. J. M. Delfino, "ImageJ ReadPlate 3.0," 2020. https://imagej.net/ij/plugins/readplate/index.html.
4. A. R. Augustė Tumaitė, "RGBraford - application for protein quanititation," 2025. https://gitlab.igem.org/2025/software-tools/vilnius-lithuania.