GSoC 2026
MCP Apps: Interactive Genomic Visualisations for AI Clients
This proposal outlines a plan to extend the cBioPortal MCP server with interactive genomic visualisation applications via MCP Apps that allow AI assistants like Claude and ChatGPT to render rich, interactive UI components directly within conversations. Rather than returning raw text or static data, the MCP server will serve four self-contained visualization apps (OncoPrint, Survival Analysis, Mutation Mapper, and Study Summary) as embeddable iframes, using bidirectional JSON-RPC communication via postMessage. This transforms the way researchers interact with cancer genomics data through AI, making cBioPortal the first genomics platform to offer interactive visual exploration inside AI chat interfaces.
Project details
Technologies
Not listed in the archive