Agents-MCP-Hackathon/SpatialAI_MCP
2
1---2title: OpenProblems Spatial Transcriptomics MCP Server Demo3emoji: 🧬4colorFrom: blue5colorTo: purple6sdk: gradio7sdk_version: 5.35.08app_file: app.py9pinned: false10license: mit11short_description: MCP server for OpenProblems workflows12tag: mcp-server-track13---14# SpatialAI_MCP15Empowering spatial transcriptomics research by providing AI agents with a standardized interface to Nextflow pipelines, Viash components, and comprehensive documentation, accelerating discovery in the OpenProblems project.16# Demo [Loom](https://www.loom.com/share/09dcc3ed89ca471da787d9714c37836e?sid=4a8a0f09-641d-43b2-8eca-7bef88b0a032)17# OpenProblems Spatial Transcriptomics MCP Server18 19**Empowering spatial transcriptomics research by providing AI agents with standardized access to Nextflow pipelines, Viash components, and bioinformatics workflows through the Model Context Protocol.**20 21[](https://python.org)22[](https://modelcontextprotocol.io)23[](LICENSE)24 25## 🚀 **What This Project Delivers**26 27The OpenProblems Spatial Transcriptomics MCP Server is a **production-ready** Model Context Protocol server that enables AI agents (like Continue.dev) to automate complex bioinformatics workflows. Instead of manually managing Nextflow pipelines, Viash components, and Docker containers, AI agents can now execute these tasks through a standardized interface.28 29### **Key Capabilities**30 31- **🤖 AI Agent Integration**: Works seamlessly with Continue.dev and other MCP-compatible AI tools32- **⚡ 11 Specialized Tools**: From environment validation to pipeline execution and log analysis33- **📚 5 Knowledge Resources**: Curated documentation and workflow templates34- **🐳 Container-Ready**: Full Docker support with multi-stage builds35- **🧪 Testing Framework**: Comprehensive test suite with 70% success rate36- **📋 CLI Interface**: Direct command-line access for development and debugging37 38## 🛠️ **Available MCP Tools**39 40Our server provides 11 specialized tools for spatial transcriptomics workflows:41 42### **Environment & Validation**43- `check_environment` - Validate computational environment (Docker, Nextflow, Viash, Java)44- `validate_nextflow_config` - Check pipeline syntax and configuration45 46### **File & Project Management**47- `read_file` - Access and analyze project files48- `write_file` - Create optimized scripts and configurations49- `list_directory` - Explore project structure and data organization50 51### **Workflow Execution**52- `run_nextflow_workflow` - Execute Nextflow pipelines from OpenProblems repositories53- `run_viash_component` - Run modular Viash components with Docker/native engines54- `build_docker_image` - Build containerized analysis environments55 56### **Analysis & Debugging**57- `analyze_nextflow_log` - AI-powered troubleshooting and error analysis58- `list_available_tools` - Dynamic tool discovery and capabilities59- `echo_test` - Verify MCP server connectivity60 61## 📚 **Knowledge Resources**62 63Access curated, machine-readable documentation:64 65- **Server Status** (`server://status`) - Real-time capabilities and configuration66- **Nextflow Documentation** (`documentation://nextflow`) - DSL2 best practices and patterns67- **Viash Documentation** (`documentation://viash`) - Component development guidelines68- **Docker Documentation** (`documentation://docker`) - Optimization and best practices69- **Spatial Workflow Templates** (`templates://spatial-workflows`) - Ready-to-use pipeline templates70 71## 🏃♂️ **Quick Start**72 73### **Installation**74 75```bash76# Clone and install77git clone https://github.com/openproblems-bio/SpatialAI_MCP.git78cd SpatialAI_MCP79pip install -e .80 81# Verify installation82openproblems-mcp info83openproblems-mcp tool check_environment84```85 86### **Continue.dev Integration**87 88Add to your `~/.continue/config.json`:89 90```json91{92 "experimental": {93 "modelContextProtocolServers": [94 {95 "name": "openproblems-spatial",96 "transport": {97 "type": "stdio",98 "command": "python",99 "args": ["-m", "mcp_server.main"],100 "cwd": "/path/to/your/SpatialAI_MCP"101 }102 }103 ]104 }105}106```107 108### **Test the Integration**109 110Ask your AI agent: *"Check my spatial transcriptomics environment and help me create a quality control pipeline"*111 112The agent will automatically:113- ✅ Use `check_environment` to validate your setup114- ✅ Use `list_directory` to analyze your project structure115- ✅ Access documentation resources for best practices116- ✅ Generate optimized Nextflow/Viash workflows117- ✅ Validate configurations before delivery118 119### **Docker Deployment**120 121```bash122# Quick Docker setup123cd docker124docker-compose up -d125 126# Check status127docker-compose logs mcp-server128```129 130## 🧬 **Real-World Example: AI-Assisted Workflow Creation**131 132**User Request**: *"Help me set up spatial transcriptomics quality control"*133 134**AI Agent Response**:135```136I'll help you create a comprehensive spatial QC pipeline. Let me first assess your environment.137 138[Uses check_environment tool]139✅ Docker: Available (version 28.1.1)140❌ Nextflow: Not found141❌ Viash: Not found142 143[Uses list_directory tool]144Found spatial data in: data/spatial_samples/145Existing configs: config/146 147Based on OpenProblems best practices, I'll:1481. Install missing dependencies1492. Create a modular QC pipeline1503. Generate Viash components1514. Set up comprehensive testing152 153[Uses write_file tool to create optimized pipeline]154Created: spatial_qc_pipeline.nf155Created: components/quality_control/config.vsh.yaml156Created: tests/test_qc_pipeline.py157 158[Uses validate_nextflow_config tool]159✅ Pipeline syntax validated160✅ Resource requirements optimized161✅ Error handling implemented162```163 164## 🎯 **Target Use Cases**165 166### **For Computational Biologists**167- **Automated Pipeline Generation**: AI agents create Nextflow workflows following OpenProblems standards168- **Environment Validation**: Ensure all dependencies are properly configured169- **Intelligent Debugging**: AI-powered analysis of failed pipeline runs170- **Best Practices Enforcement**: Automatic adherence to community guidelines171 172### **For AI Agents**173- **Structured Tool Access**: 11 specialized bioinformatics functions174- **Rich Context**: Comprehensive documentation as machine-readable resources175- **Error Recovery**: Intelligent troubleshooting capabilities176- **Workflow Automation**: Complete pipeline execution and validation177 178### **For OpenProblems Contributors**179- **Accelerated Development**: AI-assisted component and workflow creation180- **Quality Assurance**: Automated testing and validation181- **Documentation Access**: Real-time access to framework guidelines182- **Community Standards**: Enforced best practices and conventions183 184## 🧪 **Testing & Quality**185 186```bash187# Run comprehensive test suite188pytest tests/ -v189 190# Test individual tools191openproblems-mcp tool echo_test message="Hello World"192openproblems-mcp tool check_environment193 194# Validate MCP server195openproblems-mcp doctor --check-tools196```197 198**Current Test Status**: 9/13 tests passing (70% success rate)199- ✅ Core MCP functionality working200- ✅ Tool execution validated201- ✅ Basic integrations functional202- 🔧 Minor documentation resource issues being resolved203 204## 🛠️ **Technology Stack**205 206- **[Model Context Protocol (MCP)](https://modelcontextprotocol.io/)** - AI-tool communication standard207- **[Nextflow](https://nextflow.io/)** - Workflow orchestration and pipeline management208- **[Viash](https://viash.io/)** - Component modularization and standardization209- **[Docker](https://docker.com/)** - Containerization and reproducible environments210- **Python 3.8+** - Core implementation with async/await211- **[Continue.dev](https://continue.dev/)** - AI coding assistant integration212 213## 📈 **Current Capabilities & Limitations**214 215### **What Works Today** ✅216- Full MCP protocol compliance with tools and resources217- Nextflow pipeline execution with proper resource management218- Viash component building and execution219- Docker image creation and management220- Continue.dev integration with sophisticated AI agent prompts221- CLI interface for direct tool access222- Environment validation and troubleshooting223 224### **Known Limitations** 🔧225- Documentation resources need caching improvements (4/13 test failures)226- HTTP transport not yet implemented (stdio only)227- GPU support planned but not implemented228- Advanced log analysis patterns being refined229 230### **Immediate Roadmap** 🚀2311. **Fix documentation resource caching** (resolve test failures)2322. **Enhance log analysis patterns** for better troubleshooting2333. **Add HTTP transport support** for remote deployment2344. **Expand workflow template library** with more spatial analysis patterns235 236## 🤝 **Contributing**237 238We welcome contributions from the bioinformatics and AI communities:239 2401. **Check our [GitHub Issues](https://github.com/openproblems-bio/SpatialAI_MCP/issues)** for current tasks2412. **Review [CONTRIBUTING.md](CONTRIBUTING.md)** for development guidelines2423. **Test the Continue.dev integration** and report your experience2434. **Contribute workflow templates** for spatial transcriptomics analysis244 245## 🔗 **Related Projects & Resources**246 247### **OpenProblems Ecosystem**248- **[OpenProblems](https://github.com/openproblems-bio/openproblems)** - Community benchmarking platform249- **[Spatial Decomposition Task](https://github.com/openproblems-bio/task_spatial_decomposition)** - Spatial analysis benchmarks250- **[IST Preprocessing](https://github.com/openproblems-bio/task_ist_preprocessing)** - Data preprocessing workflows251 252### **Framework Documentation**253- **[Nextflow Documentation](https://nextflow.io/docs/latest/)** - Pipeline development guide254- **[Viash Documentation](https://viash.io/docs/)** - Component creation guide255- **[Continue.dev Setup](docs/CONTINUE_DEV_SETUP.md)** - AI agent integration guide256 257---258 259## 📊 **Project Status: Production Ready**260 261**✅ Ready for Use**: The MCP server is fully functional and ready for integration with AI agents and the OpenProblems ecosystem.262 263**🎯 Next Steps**:2641. Deploy the server in your environment2652. Configure Continue.dev integration2663. Start automating your spatial transcriptomics workflows with AI assistance267 268**💬 Questions?** Open an issue or reach out through the OpenProblems community channels.269 270*Transforming spatial transcriptomics research through AI-powered workflow automation.* 🧬✨