Dashboard Architecture

AMRnet is built using a modern, scalable MERN stack architecture designed to handle large-scale genomic surveillance data. This page provides both high-level overview and detailed component architecture diagrams.

Architecture Overview

AMRnet employs a four-layer architecture optimized for performance, scalability, and user experience:

  1. Data Sources & Ingestion - Automated harvesting from public genomic databases

  2. Data Lake & Processing - Harmonization and quality control in MongoDB Atlas

  3. Application Layer - MERN stack with optimized APIs and interactive React frontend

  4. User Access - Multi-modal access via web interface and RESTful API

Interactive Architecture Diagrams

We provide two complementary views of the AMRnet architecture:

Macro Architecture View

The macro view provides a high-level overview of the four main architectural layers:

📊 Macro Architecture View

High-level overview of the four main architectural layers with technology stack details.

🔍 View Macro Architecture

Detailed Component View

The expanded view shows detailed data flow and individual components:

🔬 Detailed Component View

Comprehensive data pipeline with all components, data sources, and processing stages.

🔍 View Detailed Architecture

Platform Capabilities

Technical Performance

AMRnet delivers enterprise-grade performance with significant optimizations:

  • 70-87% faster load times through advanced optimization techniques

  • 90% payload compression reducing bandwidth requirements

  • Real-time data synchronization with automated updates

  • 10,000+ API requests per hour capacity

Data Coverage

Our platform provides comprehensive AMR surveillance data:

  • 8 pathogen species from major public databases

  • 500,000+ genomic records with AMR predictions

  • 75+ countries represented in the global dataset

  • 4 languages supported (English, Spanish, French, Portuguese)

Architecture Components

Frontend Layer

React 18 Application

Modern React application with hooks, state management, and optimized rendering

Interactive Visualizations

Charts, maps, and trend analysis using D3.js integration

Multi-language Support

Complete internationalization with professional medical translation

Responsive Design

Mobile-first approach with progressive web app capabilities

Backend Layer

Node.js v18.20.4 Server

Express.js server with optimized endpoints and comprehensive middleware

RESTful API

Full-featured API with rate limiting, authentication, and detailed documentation

Performance Engine

Advanced optimization including pagination, compression, and intelligent caching

Security Framework

Helmet security, CORS protection, API authentication, and OAuth2 support

Database Layer

MongoDB Atlas

Cloud-hosted MongoDB with automated backups and horizontal scaling

Organism Collections

Specialized collections for 8 major AMR organisms with optimized schemas

Performance Indexes

Compound indexes optimized for geographic, temporal, and genotype queries

Aggregation Pipeline

Server-side processing for efficient data filtering and analysis

Infrastructure Layer

Heroku Platform

Cloud deployment with auto-scaling and continuous integration

Fixie Proxy Service

Secure database connections through SOCKS5 proxy

AWS S3 Storage

Data export storage and backup with public access capabilities

CI/CD Pipeline

Automated testing, building, and deployment through GitHub Actions

Data Sources Integration

Public Genomic Databases

Pathogenwatch

  • Salmonella Typhi with AMR predictions

  • Klebsiella pneumoniae surveillance data

  • Neisseria gonorrhoeae resistance profiles

Enterobase

  • Escherichia coli with hierarchical clustering

  • Shigella species surveillance

  • Salmonella enterica global dataset

Automated Data Harvesting

Spyder Robot

Custom web scraping tool for automated genomic data extraction

Data Ingestion APIs

RESTful endpoints for programmatic data retrieval and processing

Quality Control Pipeline

Automated validation, harmonization, and metadata standardization

Performance Optimizations

The AMRnet platform incorporates numerous performance enhancements:

Data Optimization

  • Optimized API endpoints (/api/optimized/*) with parallel processing

  • Field projection to minimize data transfer

  • Aggregation pipelines for server-side computation

  • Connection pooling for database efficiency

Frontend Optimization

  • Code splitting and lazy loading for faster initial loads

  • Service workers for offline functionality

  • Compression algorithms reducing payload sizes

  • Progressive web app features for mobile experience

Deployment Features

  • Auto-scaling based on traffic demands

  • Health monitoring with automated alerts

  • Performance tracking and analytics

  • Error logging with Sentry integration

Quality Assurance

Our architecture ensures data quality and system reliability:

Data Quality

  • Source validation from trusted genomic databases

  • Harmonization of country names, regions, and date formats

  • Deduplication algorithms preventing data redundancy

  • Version control maintaining data provenance

System Reliability

  • Automated testing with comprehensive test suites

  • Continuous monitoring of system performance

  • Backup strategies ensuring data protection

  • Security auditing with regular vulnerability assessments

Future Architecture Plans

Planned enhancements to the AMRnet architecture include:

  • Machine learning integration for predictive AMR modeling

  • Real-time alerts for emerging resistance patterns

  • Enhanced visualization with 3D geographic modeling

  • Blockchain integration for data provenance and sharing

  • Federated learning capabilities for distributed AMR analysis

For technical implementation details, please see our GitHub repository and API documentation.