Gut-Brain Explorer

An interactive dashboard exploring the connection between gut bacteria and mental health conditions.

Overview

Gut-Brain Explorer is a data-driven web application that visualizes research findings from the gut-brain axis literature. It demonstrates how gut microbiome composition correlates with various mental health conditions, and how diet influences these microbial communities.

Features

• Bacteria-Condition Explorer: Interactive chart showing which gut bacteria are associated with depression, anxiety, stress, cognitive function, and other conditions
• Key Species Profiles: Detailed cards on the most researched psychobiotics
• Diet & Microbiome: Visualizations of how dietary patterns affect gut bacteria

Architecture

Python Scripts (Data Ops)     →     JSON Files     →     Static Frontend
    ↓                                    ↓                     ↓
- Fetch PubMed data              - bacteria_conditions.json  - HTML/CSS/JS
- Process BugSigDB               - gut_brain_species.json    - Chart.js visualizations
- Curate diet associations       - diet_microbiome.json      - Interactive filters

Quick Start

1. Generate Data

python scripts/data_pipeline.py

This fetches and processes data from public sources into clean JSON files.

2. View Dashboard

Open index.html in a web browser, or start a local server:

# Python 3
python -m http.server 8000

# Then open http://localhost:8000

3. Deploy to GitHub Pages

1. Push to GitHub
2. Go to Settings → Pages
3. Select source branch (main) and root folder
4. Your dashboard will be live at https://username.github.io/repository-name

Data Sources

All data is manually curated from peer-reviewed sources:

1. BugSigDB-Inspired Curation

   • Associations manually curated based on entries in BugSigDB (bugsigdb.org)
   • Each association includes PubMed IDs for verification
   • Note: This is not a live API integration

2. PubMed E-utilities API

   • Fetches recent gut-brain axis research papers
   • Used for the "Recent Papers" feature

3. Review Literature

   • Diet-microbiome relationships from meta-analyses
   • All references cited with PMIDs

Data Methodology

Bacteria-Condition Associations

• Associations are manually curated from peer-reviewed studies
• "Study count" represents an estimated number of supporting studies, not an exhaustive count
• Effect direction (depleted/enriched/varied) based on consensus across cited studies
• All claims are linked to PubMed IDs for verification

Limitations

• This is an educational demonstration, not a comprehensive meta-analysis
• Study counts are estimates and may not reflect all published research
• Associations may vary by bacterial strain, study population, and methodology

Project Structure

gut-brain-explorer/
├── index.html              # Main dashboard page
├── css/
│   └── style.css           # Styling
├── js/
│   └── app.js              # Chart rendering & interactivity
├── data/
│   ├── bacteria_conditions.json    # Bacteria ↔ condition mappings
│   ├── gut_brain_species.json      # Key species profiles
│   └── diet_microbiome.json        # Diet ↔ bacteria relationships
├── scripts/
│   └── data_pipeline.py    # Data fetching & processing
└── README.md

Technical Details

Data Pipeline

The Python script (scripts/data_pipeline.py) demonstrates ETL skills:

• Extract: Fetches data from PubMed API and processes curated datasets
• Transform: Normalizes bacteria names, aggregates study counts, categorizes effects
• Load: Outputs clean JSON ready for visualization

Frontend

Built with minimal dependencies for easy deployment:

• Pure HTML/CSS/JavaScript
• Chart.js for interactive visualizations
• Responsive design for mobile/desktop

Key Talking Points

1. Data Engineering: Pipeline aggregates and normalizes microbiome research data from multiple sources
2. Domain Knowledge: Focus on GABA-producing bacteria (Lactobacillus), butyrate producers (Faecalibacterium), and their mental health connections
3. Accessibility: Making complex research accessible to non-experts aligns with the goal of creating tools for everyone

Disclaimer

This dashboard is for educational purposes only. The associations shown represent research findings and should not be interpreted as medical advice. Always consult healthcare professionals for health decisions.

License

MIT License - feel free to use, modify, and share.