RNAcentral MCP server
The RNAcentral MCP server exposes the RNAcentral database to AI assistants through the Model Context Protocol (MCP). It lets MCP-capable LLM clients, such as Claude Desktop and Claude Code, answer questions about non-coding RNA by calling RNAcentral directly, instead of you chaining together raw API requests.
Behind the scenes the server orchestrates several RNAcentral services (EBI Search, the RNAcentral REST API, Ensembl, and the sequence search and export microservices) so that a single question can be resolved with a single tool call.
What you can do with it
The server registers a set of tools that let an LLM:
Search RNAcentral by text or metadata, with enriched results including Rfam hits and GO annotations
Run sequence similarity searches from a raw RNA sequence
Map identifiers between RNAcentral URS IDs and external database IDs
Export sequences in bulk (FASTA or Parquet)
Find overlapping ncRNAs for a set of genomic coordinates or a gene
Retrieve secondary structure diagrams for a given URS ID
Get literature summaries describing an RNA
This makes it easy to ask natural-language questions such as “find ncRNAs overlapping this GWAS locus”, “build a training set of tRNA sequences”, or “show me the 2D structure and literature summary for this RNA” and have the assistant gather the answer for you.
Responsible use
The MCP server is the preferred interface for AI assistants carrying out RNAcentral research tasks. For direct API calls, bulk retrieval, or automated workflows, follow the responsible automated access guidance, including its rate limits and retry advice.
Quick start
The server is published on PyPI as
rnacentral-mcp-server and requires
Python 3.10+. With uv it can be fetched and run on demand with no
explicit install step:
uvx --from rnacentral-mcp-server run-server
Register that command with your MCP client. For example, in Claude Code:
claude mcp add rnacentral -- uvx --from rnacentral-mcp-server run-server
Full documentation
Installation guides for Claude Desktop, Claude Code, and other MCP clients, together with the per-tool reference and examples, are maintained separately:
Documentation: https://rnacentral.github.io/rnacentral-mcp-server/
Glama MCP directory listing: https://glama.ai/mcp/servers/RNAcentral/rnacentral-mcp-server (hosted install, schema browser, and a step-by-step usage guide)
Source code: https://github.com/RNAcentral/rnacentral-mcp-server