Spaces:
Sleeping
Sleeping
| title: Fast Follow Structure Generator | |
| emoji: 🧪 | |
| colorFrom: green | |
| colorTo: green | |
| sdk: docker | |
| app_port: 7860 | |
| # Fast Follow Structure Generator | |
| Local research-assistive Web app for generating and ranking fast-follow small-molecule candidates from a first-in-class reference structure and target context. | |
| ## Disclaimer | |
| Outputs are computational research aids only. They are not clinical, toxicology, efficacy, patent, freedom-to-operate, or legal conclusions. | |
| ## Requirements | |
| - Python 3.11 or newer. The current verification environment used Python 3.14 with RDKit installed. | |
| - Node.js 20.19 or newer for the frontend tooling. The current verification environment used Node 25. | |
| ## Run Backend | |
| ```powershell | |
| cd backend | |
| python -m pip install -e ".[dev]" | |
| python -m uvicorn app.main:app --reload --host 127.0.0.1 --port 8000 | |
| ``` | |
| ## Run Frontend | |
| ```powershell | |
| cd frontend | |
| npm install | |
| npm run dev -- --port 5173 | |
| ``` | |
| Open `http://localhost:5173`. | |
| ## Example Input | |
| - Structure: `CCOc1ccc2nc(S(N)(=O)=O)sc2c1` | |
| - Format: `SMILES` | |
| - Target: `EGFR` | |
| - Patent notes: `Avoid sulfonamide substitutions and keep Tanimoto similarity below 0.95.` | |