gRNAde

You can run and query the gRNAde MCP Server to analyze RNA structures, evaluate sequences, design RNA sequences from structures, and perform batch processing. It exposes fast synchronous tools for quick results and asynchronous jobs for longer tasks, with built‑in job management and logs so you can monitor progress and retrieve results when ready.

How to use

To use the gRNAde MCP Server, you connect to it from an MCP client (such as Claude MCP or another compatible client). You will access a suite of tools for RNA structure analysis, sequence evaluation, inverse design, and batch workflows. You submit long tasks as jobs and monitor their status, then fetch results once complete. Start by registering the server in your MCP client, then run quick analyses or submit long‑running design or batch pipelines.

How to install

Prerequisites you need before installation include Python and a compatible MCP client. You will also ensure the required Python packages are installed so the MCP server can run reliably.

Step 1: Install required Python dependencies.

pip install fastmcp loguru

Step 2: Install with Claude Code CLI (Recommended). Run these commands to register the MCP server so you can access its tools from Claude.

cd /path/to/grnade_mcp
claude mcp add geometric_rna_design -- python $(pwd)/src/server.py

Step 3: Verify installation. Confirm the server appears as connected in Claude MCP listings.

claude mcp list | grep geometric-rna-design
# Should show: geometric-rna-design: ... - ✓ Connected

Step 4: Start using. In Claude, ask for available tools from geometric_rna_design and begin executing operations.

Alternatives to Claude Code CLI showcase different configuration methods.

Alternative: Claude Desktop adds the server to your desktop configuration so Claude can discover and run it.

{
  "mcpServers": {
    "geometric_rna_design": {
      "command": "python",
      "args": ["/absolute/path/to/grnade_mcp/src/server.py"]
    }
  }
}

Additional configuration and troubleshooting

If you use other MCP clients, you can configure the server similarly by providing the Python command and the absolute path to the server entry point. No environment variables are required unless you need to customize the Python path for your setup.

In all cases ensure you point to the exact server script path shown in your environment, for example: "/absolute/path/to/grnade_mcp/src/server.py".

Notes on usage and capabilities

The MCP server exposes two API styles: a quick Sync API for small tasks and a Submit API for long tasks. Use the Sync API for RNA structure analysis and sequence evaluation, and the Submit API for RNA inverse design and batch pipelines. Job management endpoints let you check status, retrieve results, view logs, or cancel jobs.

You can run small analyses, validate inputs, and retrieve example data with minimal delay. For design and batch workflows, prepare input parameters and submit a job, then monitor its progress and fetch the final results when the job completes.

Security and maintenance

Keep dependencies up to date and monitor logs to catch issues early. If you need to update dependencies or the MCP server itself, repeat the installation steps with updated versions and re-register if required by your MCP client.

Troubleshooting

If the server won’t start, verify syntax and imports, then confirm dependencies are installed. Check the Python runtime environment and ensure the MCP client can reach the server entry point.

If tools do not appear in your MCP client, verify the server registration and ensure the path to the server script is correct in your configuration.

Available tools

analyze_rna_structure

Analyze RNA secondary structures and statistics; supports optional sequence input and structure prediction with results saved to an output file if provided.

evaluate_rna_sequences

Evaluate RNA sequences against computational metrics, with optional target structure and output to CSV.

validate_rna_inputs

Validate RNA sequence and structure inputs to ensure compatibility before processing.

get_example_data

Provide example datasets and usage examples to illustrate tool usage.

submit_rna_inverse_design

Submit RNA inverse design tasks to generate sequences that fold into specified structures or fit 3D constraints; runs asynchronously.

submit_batch_rna_pipeline

Submit high-throughput batch pipelines to design and evaluate multiple RNA targets in parallel.

get_job_status

Check the progress of a submitted job by job_id.

get_job_result

Retrieve results for a completed job by job_id.

get_job_log

View execution logs for a specific job to diagnose issues.

cancel_job

Cancel a running job by job_id.

list_jobs

List all jobs with their current status.