Coda — connectome data analysis in the browser

A node-graph editor for querying and analysing connectomic data. Each step of an analysis (query, traverse, filter, aggregate, plot) is a node, and every node shows its result right on the canvas. Nothing to install, no notebook needed.

Coda reads neuPrint (hemibrain, MANC, optic-lobe, male-CNS), CAVE (FlyWire, MICrONS), CATMAID, precomputed Neuroglancer sources and your own CSV or Parquet files. Results export as CSV, SWC, images and Neuroglancer links, or as a Python or R script that runs the same pipeline locally.

Coda needs JavaScript and WebGL. These pages do not:

Feature overview: what Coda is, on one page.
Field guide: a ten-minute introduction that builds a real pipeline.
Video: your first Coda workflow: two minutes on YouTube, from an empty canvas to a shared dashboard on the male-CNS connectome.
Node guide: what every node does, what it takes and what it returns.
MCP server: build a Coda workflow from Claude, ChatGPT or any MCP client.
Source on GitHub: MIT licensed.