🧬 Biomerkin - Multi-Agent Genomics Analysis Platform

AI-Powered Genomics Analysis using AWS Bedrock Multi-Model Architecture

---

🎯 Overview

Biomerkin is an advanced genomics analysis platform that leverages multiple AI models through AWS Bedrock to provide comprehensive, validated genomic insights. The system uses a multi-model consensus approach with Amazon Nova Pro, OpenAI GPT-OSS 120B, and OpenAI GPT-OSS 20B to ensure high-confidence results.

Key Features

• 🤖 Multi-Model AI Analysis - 3 AI models validate each other for 95%+ confidence
• ⚡ Real-Time Processing - Live progress updates during analysis
• 🔬 Comprehensive Genomics - Gene identification, variant detection, clinical significance
• 🧪 Proteomics Integration - Protein structure prediction and function analysis
• 📚 Literature Research - Automated scientific literature review
• 💊 Drug Discovery - Treatment recommendations and clinical trial matching
• 🏥 Medical Reports - AI-generated comprehensive medical reports
• ☁️ Cloud-Native - Fully deployed on AWS infrastructure

---

🚀 Live Demo

Frontend

• React App: http://biomerkin-frontend-20251018-013734.s3-website.ap-south-1.amazonaws.com
• Test Page: http://biomerkin-frontend-20251018-013734.s3-website.ap-south-1.amazonaws.com/test.html

API Endpoint

https://zb9j38oxx5.execute-api.us-east-1.amazonaws.com/prod/analyze

Quick Test

curl -X POST https://zb9j38oxx5.execute-api.us-east-1.amazonaws.com/prod/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "sequence": "ATGGATTTATCTGCTCTTCGCGTTGAAGAAGTACAAAATGTCATTAATGCTATGCAGAAAATCTTAGAGTGTCCCATCTGTCTGGAGTTGATCAAGGAACCTGTCTCCACAAAGTGTGACCACATATTTTGCAAAT",
    "analysis_type": "genomics",
    "use_multi_model": true,
    "real_time": true
  }'

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                    User Browser                              │
└──────────────────────┬──────────────────────────────────────┘
                       │
                       ↓
┌─────────────────────────────────────────────────────────────┐
│              Frontend (React + S3)                           │
│  - Modern UI with real-time updates                          │
│  - Sample data loader                                        │
│  - Progress visualization                                    │
└──────────────────────┬──────────────────────────────────────┘
                       │
                       ↓ HTTPS
┌─────────────────────────────────────────────────────────────┐
│          API Gateway (us-east-1)                             │
│  - RESTful API                                               │
│  - CORS enabled                                              │
│  - /analyze endpoint                                         │
└──────────────────────┬──────────────────────────────────────┘
                       │
                       ↓ AWS Lambda Proxy
┌─────────────────────────────────────────────────────────────┐
│     Lambda Function (Python 3.11)                            │
│  - Multi-model orchestration                                 │
│  - Real-time progress tracking                               │
│  - Error handling & recovery                                 │
└──────────────────────┬──────────────────────────────────────┘
                       │
                       ↓ Bedrock API
┌─────────────────────────────────────────────────────────────┐
│           AWS Bedrock (us-east-1)                            │
│  ┌─────────────────────────────────────────────────────┐    │
│  │ 1. Amazon Nova Pro (Primary Analysis)              │    │
│  │    - Deep genomic sequence analysis                 │    │
│  │    - Variant detection                              │    │
│  │    - Clinical significance assessment               │    │
│  └─────────────────────────────────────────────────────┘    │
│  ┌─────────────────────────────────────────────────────┐    │
│  │ 2. OpenAI GPT-OSS 120B (Validation)                │    │
│  │    - Secondary validation                           │    │
│  │    - Literature context                             │    │
│  │    - Alternative interpretations                    │    │
│  └─────────────────────────────────────────────────────┘    │
│  ┌─────────────────────────────────────────────────────┐    │
│  │ 3. OpenAI GPT-OSS 20B (Synthesis)                  │    │
│  │    - Multi-model consensus                          │    │
│  │    - Executive summary                              │    │
│  │    - Actionable recommendations                     │    │
│  └─────────────────────────────────────────────────────┘    │
└─────────────────────────────────────────────────────────────┘

---

� Project Structure

biomerkin/
├── frontend/                    # React frontend application
│   ├── src/
│   │   ├── components/         # React components
│   │   ├── pages/              # Page components
│   │   ├── services/           # API services
│   │   └── App.js              # Main app component
│   ├── public/                 # Static assets
│   ├── .env                    # Environment variables
│   └── package.json            # Dependencies
│
├── lambda_functions/           # AWS Lambda functions
│   ├── multi_model_orchestrator.py      # Main multi-model handler
│   ├── enhanced_bedrock_orchestrator.py # Advanced orchestration
│   ├── genomics_handler.py              # Genomics agent
│   ├── proteomics_handler.py            # Proteomics agent
│   ├── decision_handler.py              # Decision agent
│   ├── bedrock_literature_action.py     # Literature research
│   └── bedrock_drug_action.py           # Drug discovery
│
├── biomerkin/                  # Python package
│   ├── agents/                 # Agent implementations
│   │   ├── genomics_agent.py
│   │   ├── proteomics_agent.py
│   │   ├── literature_agent.py
│   │   ├── drug_agent.py
│   │   └── decision_agent.py
│   └── services/               # Service modules
│       ├── bedrock_agent_service.py
│       ├── bedrock_orchestration_service.py
│       └── bedrock_optimization_service.py
│
├── scripts/                    # Deployment & utility scripts
│   ├── deploy_everything_us_east_1.py   # Main deployment
│   ├── deploy_cors_fix.py               # CORS configuration
│   └── fix_api_gateway_cors.py          # API Gateway CORS
│
├── demo/                       # Demo scripts
│   ├── autonomous_bedrock_demo.py
│   ├── judge_demo_runner.py
│   └── DEMO_VIDEO_SCRIPT.md
│
├── .kiro/specs/                # Feature specifications
│   └── biomerkin-multi-agent-system/
│       ├── requirements.md
│       ├── design.md
│       └── tasks.md
│
├── README.md                   # This file
├── TECHNICAL_DOCUMENTATION.md  # Technical details
├── HACKATHON_PRESENTATION.md   # Presentation guide
├── BEDROCK_AGENTS_EXPLANATION.md        # Bedrock agents guide
├── BEDROCK_AGENTS_QUICK_REFERENCE.md    # Quick reference
└── AWS_SETUP_GUIDE_FOR_BEGINNERS.md     # AWS setup guide

---

🛠️ Technology Stack

Frontend

• React 18 - Modern UI framework
• Axios - HTTP client
• Framer Motion - Animations
• Lucide React - Icons
• Tailwind CSS - Styling

Backend

• AWS Lambda - Serverless compute
• AWS API Gateway - RESTful API
• AWS Bedrock - AI model access
• Python 3.11 - Runtime

AI Models

• Amazon Nova Pro - Primary analysis
• OpenAI GPT-OSS 120B - Validation
• OpenAI GPT-OSS 20B - Synthesis

Infrastructure

• AWS S3 - Frontend hosting
• AWS IAM - Access management
• AWS CloudWatch - Monitoring

---

🚀 Getting Started

Prerequisites

• AWS Account with Bedrock access
• Node.js 16+ and npm
• Python 3.11+
• AWS CLI configured

Local Development

1. Clone the repository

git clone https://github.com/yourusername/biomerkin.git
cd biomerkin

2. Install frontend dependencies

cd frontend
npm install

3. Configure environment

# frontend/.env
REACT_APP_API_URL=https://zb9j38oxx5.execute-api.us-east-1.amazonaws.com/prod

4. Run frontend locally

npm start

5. Install Python dependencies

cd ..
pip install -r requirements.txt

Deployment

1. Deploy Lambda function

python scripts/deploy_everything_us_east_1.py

2. Build and deploy frontend

cd frontend
npm run build
aws s3 sync build/ s3://your-bucket-name/ --delete

---

📖 Usage

Web Interface

1. Open the React app in your browser
2. Choose input method:
   • Upload a FASTA file
   • Enter DNA sequence text
   • Use sample data
3. Click "Start Analysis"
4. Watch real-time progress as 3 AI models analyze your data
5. View comprehensive results with multi-model consensus

API Usage

import requests

url = "https://zb9j38oxx5.execute-api.us-east-1.amazonaws.com/prod/analyze"

payload = {
    "sequence": "ATGGATTTATCTGCTCTTCGCGTTGAAGAAGTACAAAATGTCATTAATGCTATGCAGAAAATCTTAGAGTGTCCCATCTGTCTGGAGTTGATCAAGGAACCTGTCTCCACAAAGTGTGACCACATATTTTGCAAAT",
    "analysis_type": "genomics",
    "use_multi_model": True,
    "real_time": True
}

response = requests.post(url, json=payload)
results = response.json()

print(f"Models used: {results['models_used']}")
print(f"Confidence: {results['confidence']}")
print(f"Summary: {results['executive_summary']}")

---

🎯 Multi-Model Workflow

Step 1: Primary Analysis (Amazon Nova Pro)

• Deep genomic sequence analysis
• Gene identification
• Variant detection
• Clinical significance assessment

Step 2: Validation (OpenAI GPT-OSS 120B)

• Validates primary findings
• Adds scientific literature context
• Provides alternative interpretations
• Identifies potential concerns

Step 3: Synthesis (OpenAI GPT-OSS 20B)

• Creates unified summary
• Highlights consensus findings
• Notes any discrepancies
• Provides actionable recommendations

Result: High-Confidence Analysis

• 95%+ confidence through multi-model consensus
• Comprehensive insights from multiple perspectives
• Validated results with cross-model verification

---

📊 Features

Genomics Analysis

• ✅ DNA sequence analysis
• ✅ Gene identification
• ✅ Variant detection (SNPs, indels)
• ✅ Clinical significance assessment
• ✅ Pathogenicity prediction

Proteomics Analysis

• ✅ Protein structure prediction
• ✅ Functional annotation
• ✅ Binding site identification
• ✅ Post-translational modifications

Literature Research

• ✅ Automated PubMed search
• ✅ Relevant paper identification
• ✅ Key findings extraction
• ✅ Research recommendations

Drug Discovery

• ✅ Treatment recommendations
• ✅ Drug-gene interactions
• ✅ Clinical trial matching
• ✅ Pharmacogenomics insights

Medical Reporting

• ✅ Comprehensive medical reports
• ✅ Risk assessment
• ✅ Treatment options
• ✅ Follow-up recommendations

---

🔒 Security & Privacy

• ✅ CORS configured for secure cross-origin requests
• ✅ IAM roles with least-privilege access
• ✅ No data storage - analysis is real-time only
• ✅ HTTPS only - encrypted communication
• ✅ AWS security - leverages AWS security best practices

---

📚 Documentation

• Technical Documentation - Detailed technical guide
• Hackathon Presentation - Presentation guide
• Bedrock Agents Guide - AWS Bedrock agents
• Quick Reference - Quick reference guide
• AWS Setup Guide - AWS setup for beginners
• Demo Script - Demo video script

---

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

🙏 Acknowledgments

• AWS Bedrock for providing access to multiple AI models
• Amazon Nova Pro for powerful genomics analysis
• OpenAI GPT-OSS models for validation and synthesis
• React community for excellent frontend tools
• AWS community for serverless best practices

---

📧 Contact

For questions, issues, or collaboration opportunities:

• GitHub Issues: Create an issue
• Email: your.email@example.com

---

🎉 Hackathon Success

This project was built for [Hackathon Name] and demonstrates:

• ✅ Multi-model AI - Innovative use of multiple AI models
• ✅ Real-world application - Practical genomics analysis
• ✅ Production-ready - Fully deployed on AWS
• ✅ Scalable architecture - Serverless and cloud-native
• ✅ User-friendly - Intuitive interface with real-time feedback

---

Built with ❤️ using AWS Bedrock, React, and Python

Last Updated: January 2025