VacSol-ML(ESKAPE)

VacSol-ML(ESKAPE) is a machine learning–driven framework for vaccine target discovery, specifically designed to analyze and prioritize candidate proteins from ESKAPE pathogens. The tool integrates bioinformatics feature extraction with trained ML models to assist researchers in identifying promising vaccine candidates in a reproducible and scalable manner.

---

🔬 Background

ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.) are responsible for a large proportion of antimicrobial-resistant infections worldwide. Identifying effective vaccine targets against these pathogens is a critical challenge.

VacSol-ML(ESKAPE) addresses this by combining:

• Sequence-derived biological and physicochemical features
• Machine learning–based classification
• Automated prediction and confidence scoring

---

✨ Key Features

• 🧬 Protein-level vaccine target prediction
• 🤖 Machine learning–based classification pipeline
• 📊 Confidence scoring for predictions
• 🖥️ Immunological and Physiochemical descriptors extraction of the input sequences.
• 📦 Distributed as a pip-installable package

---

📦 Installation

Requirements

• Python >=3.10, <3.13

Install from PyPI

pip install vacsol-ml

This will install VacSol-ML along with its required dependencies.

⚠️ TensorFlow is a required dependency and will be installed automatically during setup. Depending on your system and internet speed, this step may take some time.

---

🚀 Usage

VacSol-ML is designed to be used after installation via pip through a simple command-line workflow that launches a local web interface.

Step 1: Run database migrations

After installing the package, initialize the required database tables by running:

vacsol-ml migrate

This command prepares the internal database required for analysis.

---

Step 2: Launch the web interface

Start the local web server using:

vacsol-ml web

This will launch a local server at:

http://127.0.0.1:8000/

Open this URL in your browser to access the VacSol-ML web interface and perform analyses interactively.

---

Step 3: Run analysis via the web UI

Once the server is running:

• Upload or select protein sequence data
• Configure analysis parameters
• Execute the ML-based vaccine target prediction pipeline
• View and download results directly from the interface

⚠️ Note: The web interface is intended for local or server-based use. It is not recommended for Google Colab or notebook-based environments.

---

📁 Input Format

• Input sequences must be provided in FASTA format
• Each entry should represent a protein sequence

Example:

>Protein_1
MKKLLPTAAAGLLLLAAQPAMA...

---

📤 Output

VacSol-ML generates tabular outputs containing:

• Protein ID
• Predicted class (e.g., Vaccine Candidate / Non-candidate)
• Model confidence score
• Downloadable datasets containing Immunological and Physiochemical features of the input sequences.

---

📚 Citation

If you use VacSol-ML in your research, please cite the associated publication:

VacSol-ML: An ML-driven framework for vaccine target discovery against ESKAPE pathogens Vaccine (2024). https://doi.org/10.1016/j.vaccine.2024.126204

---

👩‍🔬 Author

Samavi Nasir MSc Industrial Biotechnology (ASAB, NUST) 📧 Email: samavi.nasir@gmail.com 🔗 Google Scholar: https://scholar.google.com/citations?hl=en&user=K_okdSMAAAAJ

---

📄 License

This project is released under the MIT License. See the LICENSE file for details.

---

⚠️ Disclaimer

VacSol-ML(ESKAPE) is intended for research purposes only. VacSol-ML(ESKAPE) provides in silico predictions to support vaccine target identification. These predictions necessitate downstream experimental validation before clinical translation.

---

🤝 Contributing

Contributions, bug reports, and feature requests are welcome. Please open an issue or submit a pull request via the project repository.