SciganticGuidesTopicsExplore

How to browse CZ CryoET Data Portal datasets in a Jupyter notebook

Every dataset is a numbered folder with a dataset_metadata.json describing the sample, and its runs hold tomograms as multi-scale OME-Zarr. In a Scigantic notebook the cryoet-data-portal-public bucket is mounted at /mnt/archive, so pathlib and json are enough to find a dataset and read its metadata.

The code on this page runs in a Scigantic hosted notebook, where the dataset is mounted read-only at /mnt/archive. The same path is in the SCIGANTIC_MOUNT_PATH environment variable and linked from ~/archive. The public source is s3://cryoet-data-portal-public (AWS us-west-2, public, no account needed).

Open CZ CryoET Data Portal on SciganticOpen the sample notebook: A cryo-EM tour of a real cell

Code

List datasets and read one dataset's metadata

import os
from pathlib import Path

mount = Path(os.environ.get('SCIGANTIC_MOUNT_PATH', '/mnt/archive'))

print(f'Mounted at:    {mount}  (exists: {mount.exists()})')

# Each top-level directory is a dataset id (an integer assigned by the portal).
# We sort and show just the first handful — there are hundreds.
dataset_dirs = sorted([p for p in mount.iterdir() if p.is_dir() and p.name.isdigit()],
                     key=lambda p: int(p.name))
print(f'First 8 datasets visible:')
for d in dataset_dirs[:8]:
    print(f'  {d.name}')
if len(dataset_dirs) > 8:
    print(f'... (more datasets follow)')

import json
import textwrap

dataset = mount / '10000'
meta = json.loads((dataset / 'dataset_metadata.json').read_text())

print(f"Title:    {meta['dataset_title']}")
print(f"Organism: {meta['organism']['name']}")
print(f"Sample:   {meta['sample_type']}")
print(f"Released: {meta['dates']['release_date']}")
print()
print('Description:')
for line in textwrap.wrap(meta['dataset_description'], width=88):
    print(f'  {line}')
print()
first = next(a for a in meta['authors'] if a.get('primary_author_status'))
print(f"First author: {first['name']}")

Copied from this dataset's starter notebook on Scigantic, the one that opens when you start a notebook from the archive page (2 cells, unchanged).

What the files are

The CZ CryoET Data Portal from the Chan Zuckerberg Initiative holds about 370 public cryo-electron tomography datasets. Each is a set of tilt-series runs reconstructed into tomograms, many with expert or machine-generated 3D annotations such as particle picks and segmentations.

It is the tomography counterpart to EMPIAR and EMDB, which are almost entirely single-particle data: here cells and organelles are imaged directly in 3D.

How the files are laid out

Path under the mountWhat it holds
<dataset_id>/dataset_metadata.jsonComplete metadata for the dataset: title, organism, sample type, authors, dates.
<dataset_id>/<run_name>/One run, for example 10000/TS_026, with folders for frames, tilt series, reconstructions and annotations.
<dataset_id>/<run_name>/Reconstructions/VoxelSpacing<apix>/Tomograms/<n>/<n>.zarrA multi-scale OME-Zarr tomogram, next to tomogram_metadata.json.
.../Reconstructions/VoxelSpacing<apix>/Annotations/<n>/One annotation per folder: a JSON description plus points (_point.ndjson) or a segmentation mask.

Size, format and license

FormatOME-Zarr tomograms, MRC, JSON metadata, NDJSON point annotations
SizeAbout 370 datasets, many with tens to thousands of runs
LicenseCC0
ProviderChan Zuckerberg Initiative Foundation, on AWS Open Data
Public sources3://cryoet-data-portal-public (AWS us-west-2, public, no account needed)
Notebook imageCryo-EM (scigantic_cryoet, copick and mrcfile preinstalled)

Things to know before you read it

Run it

Start a notebook from the CZ CryoET Data Portal archive page. The data is mounted when the notebook opens, so the code above runs as written and nothing is downloaded first. For a longer worked example, open the sample notebook A cryo-EM tour of a real cell.

Related guides

Last checked 2026-11-21. All dataset guides.