GenAMR demonstrates how qPCR screening, selective Sanger sequencing, and an interpretation layer could support an antimicrobial resistance (AMR) testing workflow. Explore six interactive clinical archetypes and 60 historical synthetic isolates through a responsive lab and bedside interface.
Open the live demo · Demo walkthrough · Documentation · Contributing
Research demonstration — synthetic data only. GenAMR is not a validated diagnostic service or treatment recommendation. Conventional phenotypic AST remains the clinical confirmation. Scores and retraining are deterministic simulations, not measured clinical performance.
| Workflow | Interactive features |
|---|---|
| 1. Collect | Synthetic sample intake, specimen label and collection timeline |
| 2. qPCR | 40-cycle amplification curves, target toggles and Ct controls |
| 3. Decide | Fast-track versus targeted-reflex routing, illustrative fees and turnaround |
| 4. Sanger | Four-color chromatograms, mutation toggles and a side-by-side PointFinder reference comparison with highlighted differences |
| 5. Report | Empirical-therapy mismatch alerts, 16-drug illustrative matrix and printable report |
| Validation | Phenotypic AST reconciliation and simulated model-version updates |
| Surveillance | Eight-division synthetic summaries and CSV/JSON exports |
The app includes desktop and mobile layouts, English/Bangla key labels, BDT/USD pricing, persona controls, local persistence and an installable PWA. Visited assets become available offline after an initial online visit.
- Open the live app and choose a case in the scenario controller.
- Compare GAMR-BD-001 (fast track; ৳1,200) with GAMR-BD-002 (Sanger reflex; ৳2,050).
- Change a qPCR target or Sanger mutation and inspect the updated report.
- Explore validation and surveillance using the synthetic records.
See the six-case walkthrough for more examples.
Use Node.js 20.9 or newer; Node.js 24 is recommended.
git clone https://github.com/rishadamin001/GenAMR-App.git
cd GenAMR-App
npm ci
npm run devOpen http://127.0.0.1:3000. No environment variables, paid APIs or cloud account are needed for local mode.
For a production build:
npm run build
npm run startNext.js 16 · React 19 · TypeScript · Tailwind CSS 4 · Recharts · HTML5 Canvas · optional Supabase
src/
app/ Next.js pages, styles and PWA manifest
components/ Workflow, layout, validation and surveillance UI
lib/
bioinformatics/ Deterministic qPCR, reflex, Sanger and interpretation engines
data/ Local-first repository and optional cloud synchronization
seed/ Synthetic Bangladesh AMR fixtures
public/ PWA assets
supabase/migrations/ Eight-table PostgreSQL schema and access policies
tests/ Engine tests and desktop/mobile browser tests
docs/ Guides, architecture, scope and verification evidence
The default store uses memory and browser localStorage. Optional Supabase integration uses authenticated, owner-scoped uploads; see the data and scientific scope guide and backend plan before enabling it. Live Supabase synchronization and RLS checks have not been verified against a connected project.
The live demo runs on Vercel. The current deployment was uploaded directly; GitHub pushes do not automatically deploy until the Vercel GitHub integration is connected. Local mode requires no paid services, cron jobs or cloud add-ons. Provider free-tier limits and account billing settings still apply.
npm run typecheck
npm test
npm run build
npx playwright install chromium
npm run test:e2eThe latest recorded verification passed 10 engine tests and 4 browser tests, including desktop (1440 × 900) and mobile (390 × 844) flows. Live production smoke tests also passed. These are software checks, not evidence of diagnostic accuracy.
Read the verification record for results and limitations.
-
Sanger database comparison — reference sources, highlighted differences and comparison scope
-
Documentation index — find the right guide
-
Demo walkthrough — six scenarios and interactive checks
-
Implementation scope — implemented behavior and boundaries
-
Backend plan — schema, access model and synchronization
-
Data and scientific scope — research limitations and cloud setup
-
Verification record — build, test and deployment evidence
-
Contributing — development and review expectations