How tarenpaper/valoreax works
ValoreaX is an account-scoped biotech research application. A user signs in, selects a public healthcare company, and the frontend requests backend research APIs; SEC data is ingested, normalized and persisted before valuation, catalyst tracking and explainable signals are presented. Separate product branches cover news, analyst coverage, watchlists, backtesting and experimental clinical-trial ML. The graph reflects README capabilities and the supplied source index/excerpts; several API and service internals were not sampled, so undocumented connections are omitted.
Main entry points: v1 Versioned API · client.ts Typed API client · ingestion_service.py SEC ingestion · models Research data models
Components 26
User experience
- Account access
frontend/src/auth/AuthGate.tsx - Application navigation
frontend/src/App.tsx - Company dashboard
frontend/src/pages/Dashboard.tsx - News dashboard
frontend/src/pages/News.tsx - Watchlists
frontend/src/pages/Watchlist.tsx - Backtest view
frontend/src/pages/Backtest.tsx - Typed API client
frontend/src/api/client.ts
Data ingestion
- External data adapters
backend/app/providers/base.py - SEC ingestion
backend/app/services/ingestion_service.py - Financial normalization
backend/app/services/normalization.py
Research intelligence
- Drug valuation
backend/app/services/rnpv_valuation.py - Explainable signals
backend/app/services/signals.py - Catalyst tracking
backend/app/services/catalyst_ingestion.py - Analyst coverage
backend/app/services/analyst_ingestion.py - News intelligence
backend/app/services/news_analysis.py - Watchlist service
backend/app/services/watchlist.py - Signal evaluation
backend/app/services/evaluation.py - Evidence research
backend/app/api/v1/research.py - Gemini adapter
backend/app/providers/gemini.py
Backend platform
- Versioned API
backend/app/api/v1 - API authorization
backend/app/auth.py - Research data models
backend/app/models
Clinical ML
- Clinical ML panel
frontend/src/components/ClinicalMLPanel.tsx - Clinical ML workflow
backend/app/services/clinical_ml.py - Model training
backend/app/ml/training.py
Other components
- Research user
Connections 30
- Research user → Account access: signs in
frontend/src/auth/AuthGate.tsx - Account access → Application navigation: opens workspace
frontend/src/main.tsx - Application navigation → Company dashboard: renders
frontend/src/App.tsx - Application navigation → News dashboard: renders
frontend/src/App.tsx - Typed API client → Account access: gets token
frontend/src/api/client.ts - Company dashboard → Typed API client: requests data
frontend/src/pages/Dashboard.tsx - News dashboard → Typed API client: requests news
frontend/src/pages/News.tsx - Watchlists → Typed API client: requests watchlists
frontend/src/pages/Watchlist.tsx - Backtest view → Typed API client: requests evaluation
frontend/src/pages/Backtest.tsx - Typed API client → Versioned API: calls API
frontend/src/api/client.ts - Versioned API → API authorization: protects requests
- Versioned API → SEC ingestion: starts ingestion
- SEC ingestion → External data adapters: fetches facts
backend/app/services/ingestion_service.py - SEC ingestion → Financial normalization: normalizes payload
backend/app/services/ingestion_service.py - SEC ingestion → Research data models: persists filings
backend/app/services/ingestion_service.py - Financial normalization → Research data models: supplies metrics
- Versioned API → Drug valuation: serves valuation
- Versioned API → Explainable signals: serves signals
- Versioned API → Catalyst tracking: serves catalysts
- Versioned API → Analyst coverage: serves analyst data
- Versioned API → News intelligence: serves news
- Versioned API → Watchlist service: serves watchlists
- Versioned API → Signal evaluation: evaluates signals
- Versioned API → Evidence research: routes research
backend/app/api/v1/research.py - Evidence research → Gemini adapter: generates analysis
backend/app/api/v1/research.py - Clinical ML panel → Typed API client: requests ML
- Versioned API → Clinical ML workflow: serves trial model
- Clinical ML workflow → Model training: uses trained models
- Drug valuation → Research data models: reads research data
- Explainable signals → Research data models: persists signal runs