FaceBase: craniofacial research data (open access)
FaceBase is the NIH craniofacial research data hub, run by the FaceBase Consortium at the University of Southern California. It holds 1,074 open-access datasets (as of 2026-10-09) spanning micro-CT, light-sheet and confocal imaging, ChIP-seq, RNA-seq and single-cell sequencing, expression tables and morphometrics, mostly from mouse and zebrafish. About 995 of them carry files: 56,673 files and 19.7 TB. A further 91 datasets hold protected human-subjects data that FaceBase releases only through a data access request, and those are not readable here. That includes EarBase, the NIDCD temporal bone and hearing repository that lives on the same platform: its donor cases, specimens and slide images need a FaceBase account and an approved data access request, and only a single protected dataset record is visible without one. FaceBase files are not in S3, so nothing is mounted or copied. The preinstalled scigantic_facebase library searches the datasets, writes the citation FaceBase asks for, and reads files over HTTP, including a single slice of a zipped micro-CT stack without downloading the scan. The starter notebook compares zebrafish heads raised hypothyroid, euthyroid and hyperthyroid from an 81-scan atlas. Copyright to the data belongs to the contributing investigators and FaceBase's Terms of Use apply: cite each dataset's DOI and send other people to FaceBase rather than passing files on.
- Category
- Imaging
- Provider
- FaceBase Consortium (University of Southern California; NIDCR, NIDCD)
- License
- Per dataset. Copyright is held by the contributing investigators; FaceBase Terms of Use for Open Access Data apply (cite the dataset DOI, do not redistribute files).
- DOI
- 10.1177/00220345221107905
- Hosted on
- bucket www.facebase.org, us-east-1
- Access
- Free and public. No account with the data provider is needed.
- Source
- https://www.facebase.org/
Data structure and loading code
Files
Example file: 2E-YW20 (microCT scan of 17.5 mm hyperthyroid Danio rerio head): one 48.8 MB zip, HyperTH17.5mm1.zip, holding 596 TIFF slices at 10.5 um plus a SkyScan scan log; opening it costs 2 requests and 1.6 MB.
Load the data
import scigantic_facebase as fb
import matplotlib.pyplot as plt
f = fb.files("2E-YW20")[0]
with fb.microct.open_stack(f) as stack:
img = stack.read_slice(len(stack) // 2)
print(fb.count(), "open datasets;", fb.cite("2E-YW20"))
plt.imshow(img, cmap="gray"); plt.axis("off")Written for a Scigantic notebook, where /mnt/archive is the mounted dataset.
Analyze this dataset without downloading it
On Scigantic, FaceBase: craniofacial research data (open access) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
Related Imaging datasets
- Catalog of IDR images formatted as OME-NGFF
- Brain Image Library (BIL): BICCN whole-brain microscopy
- ExLLSM (Gao 2019)
- Image Data Resource (IDR): EMBL-EBI's archive of published cell and tissue imaging studies
- ExA-SPIM test acquisition 2026-08-04: four light-sheet tiles as OME-NGFF 0.5 / Zarr v3, sharded
- Curtin 2026
- Xu 2024
- ExA-SPIM raw acquisition 3163606: 30 light-sheet tiles of one expanded mouse brain as OME-Zarr 0.4 (Zarr v2)