SmartFilter-UI

SmartFilter-UI is a modern web interface for molecular filtering and analysis using SMARTS patterns.
It provides researchers, chemoinformaticians, and developers with an interactive way to test compounds against well-established substructure filters such as PAINS, Blake, Glaxo, ALARM NMR, Oprea, Toxicity, and MLSMR.

Built with React + TypeScript + Vite and powered by RDKit.js, SmartFilter-UI integrates seamlessly with the SmartFilter API.

---

Screenshot

---

🚀 Features

• Filter Mode – Test a set of molecules against selected filters and inspect failed matches.
• Analyze One Molecule Mode – Evaluate a single molecule against all filters with detailed matches.
• Supported SMARTS Sets:
  • PAINS – Pan-Assay Interference Compounds (Baell & Holloway, 2010).
  • Blake – From Array Biopharma (translated from SLN to SMARTS).
  • Glaxo – Includes unsuitable leads, natural products, reactive subsets, acids/bases, electrophilic, and nucleophilic subsets.
  • ALARM NMR – Abbott method for detecting reactive false positives.
  • Oprea – Multi-objective library fitness filters.
  • Toxicity – Based on published toxicophores.
  • MLSMR – NIH Roadmap screening filters.
• Expert Mode Options – Configure output with “Show Matches,” “Include Passes,” “Include Fails,” “Strict Mode,” etc.
• Batch Processing – Upload SMILES files and process thousands of molecules.
• Visualization – Molecules rendered via RDKit.js with highlighted substructure matches.
• Modern UI – Built with Vite, React, TailwindCSS, and Docker-ready.

---

🧪 Example

👉 Visit the live app here: SmartFilter-UI Deployment

👉 API Reference: SmartFilter API

---

🛠️ Tech Stack

• Frontend: React + TypeScript + Vite
• Cheminformatics: RDKit.js
• Styling: TailwindCSS, ShadCN/UI
• Deployment: Docker + Nginx (frontend), Apache (server proxy)
• API: SmartFilter API

---

📦 Installation & Development

Clone and run locally:

git clone https://github.com/unmtransinfo/SmartFilter-UI.git
cd SmartFilter-UI
npm install
npm run dev

App runs at: http://localhost:5173/smartsfilter

---

🐳 Docker Setup

Build and run with Docker:

docker build -t smartfilter-ui .
docker run -p 5173:5173 smartfilter-ui

---

🙌 Acknowledgments

• Supervision: Jeremy Yang
• Developer: Bivek Sharma Panthi
• Supporter: Jack Ringer
• Some UI ideas adapted from Badapple2-UI.
• Thanks to RDKit.js for cheminformatics functionality.

---

📄 License

This project is licensed under the BSD-3-Clause License.

---

👥 Audience

This project is intended for:

• Researchers – to filter out problematic compounds in screening libraries.
• Developers – to integrate filtering pipelines into cheminformatics workflows.
• Students & Educators – as an example of cheminformatics web app development with RDKit.js.