This is a project to showcase results obtained for the Longevity hackathon

A Streamlit-based web application for biomedical drug research and discovery, integrating multiple pharmaceutical and biomedical databases for comprehensive drug analysis and exploration.

• Multi-Database Integration: Combines data from DrKG, DrugBank, MeSH, SIDER, DOID, and HGNC
• Interactive Web Interface: Built with Streamlit for intuitive data exploration
• Containerized Deployment: Docker-based setup for consistent environments
• Knowledge Graph Analysis: Leverages drug repurposing knowledge graphs
• Disease Ontology Integration: Incorporates structured disease classifications

• Docker and Docker Compose
• Make (optional, for convenience commands)
• Minimum 4GB RAM recommended
• At least 2GB free disk space for data and containers

The application integrates the following biomedical databases:

data/
├── doid/                    # Disease Ontology
│   └── doid.obo            # Disease classifications and relationships
├── drkg/                    # Drug Repurposing Knowledge Graph
│   ├── drkg.tsv            # Main knowledge graph data
│   ├── graph.gml           # Graph structure file
│   ├── relation_glossary.tsv # Relationship definitions
│   └── relation_glossary.xlsx
├── drugbank/                # DrugBank Database
│   └── drugbank_vocabulary.csv # Drug nomenclature and metadata
├── hgnc/                    # Human Gene Nomenclature Committee
│   └── HGNC_complete_set.tsv # Official gene symbols and names
├── mesh/                    # Medical Subject Headings
│   └── desc2025.xml        # Medical terminology hierarchy
├── sider/                   # Side Effect Resource
│   └── meddra_all_indications.tsv # Drug side effects and indications
├── drugbank_vocabulary.csv  # Additional drug vocabulary
└── entity_name_mapping.json # Entity name mappings across databases

• DOID: Disease Ontology - Standardized disease classifications
• DrKG: Drug Repurposing Knowledge Graph - Comprehensive biomedical knowledge graph
• DrugBank: DrugBank Database - Pharmaceutical knowledge base
• HGNC: HUGO Gene Nomenclature Committee - Official gene symbols
• MeSH: Medical Subject Headings - NLM's controlled vocabulary
• SIDER: Side Effect Resource - Drug side effects database

Place the following files in the access/ directory to enable Google Drive access: access/service_account.json. Instructions on creating a Google Drive client and connecting it to the code can be found here.

# Clone the repository
git clone <repository-url>
cd <project-directory>

# Build and start all services
make

Don't forget to place Google Drive client JSON file.

# Set environment variables
export UID=$(id -u)
export GID=$(id -g)

# Build and start
./run.sh

Once the containers are running, access the Streamlit application at:

http://localhost:8501

make help     # Show all available commands
make build    # Build all containers
make up       # Start containers in detached mode
make down     # Stop all containers
make logs     # View container logs
make ps       # Show container status
make restart  # Restart services
make clean    # Clean volumes (⚠️ removes data)

# Access container shell
make shell SERVICE=streamlit_app

# View logs in real-time
make logs

# Restart specific service
make restart SERVICE=streamlit_app

• Streamlit App: Port 8501 (configurable in docker-compose.yml)

• ./streamlit_app → /app/streamlit_app (Application code)
• ./data → /app/data (Database files)
• ./access → /app/access (Access control files)

.
├── docker-compose.yml      # Docker services configuration
├── Dockerfile             # Container build instructions
├── Makefile               # Development convenience commands
├── requirements.txt       # Python dependencies
├── project_config.py      # Project configuration
├── streamlit_app/         # Streamlit application code
│   └── root_page.py       # Main application entry point
├── app/                   # Core application modules
├── research_scripts/      # Research and analysis scripts
├── data/                  # Database files (see structure above)
└── access/                # Access control and authentication

1. Fork the repository
2. Create a feature branch: git checkout -b feature-name
3. Make your changes
4. Test with: make build && make up
5. Submit a pull request

This project is licensed under the MIT License.

Maintainers: mikhail.solovyanov@gmail.com