agentleFS
NeuroAIHubNeuroAIHub/awesome_cognitive_and_neuroscience_skills/skills/pycortex-guide/SKILL.md

pycortex-guide

Skill39 starsChanged 9 months agoMITView on GitHub

Domain-validated guidance for cortical surface visualization and brain surface rendering of fMRI data using pycortex: data types (Volume, Vertex, Dataset), 2D cortical flatmaps, 3D WebGL brain viewers, volume-to-surface mapping, FreeSurfer/fMRIPrep integration, ROI management, and surface analysis. Use this skill whenever the user mentions pycortex, `import cortex`, cortical surfaces, brain flatmaps, WebGL brain viewers, cortical surface mapping, or wants to visualize neuroimaging data on the cortex, even if they don't explicitly name pycortex.

  • Installs packages

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When saving datasets with pack=True, subject geometry and transforms are embedded in the HDF5 file for portability.

    line 123

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pycortex Guide

Purpose

Encodes domain knowledge for using pycortex — a Python library for visualizing fMRI and volumetric neuroimaging data on cortical surfaces. Covers data creation, 2D flatmap rendering, interactive 3D WebGL viewers, volume-to-surface mapping, subject database management, FreeSurfer/fMRIPrep integration, and surface geometry analysis.

When to Use This Skill

Activate when the user:

  • Wants to visualize fMRI data on cortical surfaces
  • Mentions pycortex, import cortex, flatmaps, or cortical surface visualization
  • Needs to create Volume, Vertex, or Dataset objects for brain data
  • Wants to generate 2D flatmaps or 3D interactive brain viewers
  • Needs to import FreeSurfer or fMRIPrep subjects into pycortex
  • Works with volume-to-surface mapping or cortical ROIs
  • Needs to align functional data to anatomical surfaces
  • Wants to compute surface properties (curvature, thickness, geodesic distance)

Reference Files

TopicFileWhen to Read
Data typesreferences/dataset-types.mdUser creates Volume, Vertex, RGB, 2D, or Dataset objects
Visualizationreferences/visualization.mdUser wants flatmaps, WebGL viewers, exports, or screenshots
Database & subjectsreferences/database-subjects.mdUser manages subjects, surfaces, transforms, or masks
Mapping & transformsreferences/mapping-transforms.mdUser maps volume↔surface, aligns data, or works with transforms
FreeSurfer & fMRIPrepreferences/freesurfer-fmriprep.mdUser imports subjects from FreeSurfer or fMRIPrep
Surface analysisreferences/surface-analysis.mdUser computes curvature, geodesic distance, ROIs, or distortion
MNI & utilitiesreferences/mni-utils.mdUser transforms to/from MNI space or uses volume utilities

Installation

pip install -U setuptools wheel numpy cython
pip install -U pycortex
# With headless rendering support:
pip install -U 'pycortex[headless]'

Requirements: Python 3.10+, Linux/macOS only. Key dependencies: numpy, scipy, matplotlib, nibabel, h5py, tornado, shapely, lxml.

Overview Pipeline

1. Import subject (FreeSurfer/fMRIPrep)  →  stored in cortex.db
2. Create data objects (Volume/Vertex)    →  wrap arrays with metadata
3. Visualize (quickflat / webgl / export) →  2D flatmaps or 3D viewers
4. Analyze surfaces (curvature, ROIs)     →  surface geometry tools

Quick Start

import cortex

# Create a random volume for demo subject "S1" with transform "fullhead"
vol = cortex.Volume.random("S1", "fullhead")

# 2D flatmap
fig = cortex.quickshow(vol, with_curvature=True, with_rois=True)
fig.savefig("flatmap.png")

# Interactive 3D viewer
cortex.webshow(vol)

# Create volume from your own data (3D numpy array)
import numpy as np
data = np.random.randn(31, 100, 100)  # must match transform shape
vol = cortex.Volume(data, "S1", "fullhead", cmap="RdBu_r", vmin=-2, vmax=2)
cortex.quickshow(vol)

# Save/load datasets
ds = cortex.Dataset(my_map=vol)
ds.save("results.hdf")
ds_loaded = cortex.load("results.hdf")

Key Data Types

ClassDescriptionKey Args
VolumeVolumetric data (3D/4D)data, subject, xfmname, cmap, vmin, vmax
VertexSurface vertex data (1D/2D)data, subject, cmap, vmin, vmax
VolumeRGBRGB per voxelred, green, blue, subject, xfmname, alpha
VertexRGBRGB per vertexred, green, blue, subject, alpha
Volume2DTwo volumes, 2D colormapdim1, dim2, subject, xfmname, vmin, vmax, vmin2, vmax2
Vertex2DTwo vertex maps, 2D colormapdim1, dim2, subject, vmin, vmax, vmin2, vmax2
DatasetContainer for multiple views**named_dataviews

Key Visualization Functions

FunctionDescription
cortex.quickshow(data, ...)2D flatmap → matplotlib Figure
cortex.quickflat.make_png(fname, data, ...)Save flatmap as PNG
cortex.quickflat.make_svg(fname, data, ...)Save flatmap as SVG
cortex.quickflat.make_gif(fname, volumes, ...)Animated GIF from volumes
cortex.webshow(data, ...)Interactive 3D WebGL viewer
cortex.webgl.make_static(path, data, ...)Static HTML 3D viewer
cortex.export.save_3d_views(vol, ...)Multi-angle PNG exports
cortex.export.plot_panels(vol, panels, ...)Multi-panel figure

Common Pitfalls

  1. Volume data shape must match the transform dimensions in the database. Use cortex.db.get_xfm(subject, xfmname) to check expected shape.
  2. Subject and transform names must exist in cortex.db before creating Volume objects. Import from FreeSurfer first.
  3. cortex.quickshow is an alias for cortex.quickflat.make_figure — they are the same function.
  4. Vertex data length must match total vertex count (left + right hemisphere). Single-hemisphere data is auto-padded with zeros.
  5. vmin/vmax default to 1st/99th percentile of data. Always set explicitly for consistent colorbars across subjects.
  6. WebGL viewer (webshow) starts a Tornado server — it blocks in scripts. Use in IPython/Jupyter or set autoclose=True.
  7. The pycortex filestore path is set in cortex.options.config. Check with cortex.database.default_filestore.
  8. For headless rendering (no display), install with pip install 'pycortex[headless]' and use cortex.export.save_3d_views(..., headless=True).
  9. cortex.db is a singleton — all operations share the same database instance.
  10. When saving datasets with pack=True, subject geometry and transforms are embedded in the HDF5 file for portability.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.