Butler logo Butler logo

OME-Zarr Tools

Empowering the imaging community with the next generation of file formats

Next Generation File Formats (NGFF) are community-driven, FAIR-compliant, cloud-optimized file formats designed to address the scale and complexity of modern imaging data. NGFF overcomes the limitations of traditional monolithic file formats by enabling efficient storage, scalable access and integration across a growing ecosystem of imaging tools and platforms. 

A leading implementation of NGFF is OME-Zarr. Built on the chunked file format Zarr, OME-Zarr enables cloud-native, parallel data access, hence empowering researchers to manage, share and analyse large imaging datasets with unprecedented efficiency.

Why do we support OME-Zarr?

OME-Zarr offers a range of advantages for researchers such as

  • Fast, on-demand access to chunked data, enabling efficient streaming, subsetting and visualization 
  • Native support for multiscale datasets (image pyramids)
  • Efficient parallel writes, especially to remote storage
  • Rich, community-driven metadata layouts and specifications that integrate image data, segmentation masks, and analysis results in one unified, hierarchical structure
  • A growing community support

Image credits: 'monolithic-vs-chunked' by Henning Falk, ©2022 NumFOCUS, used under a CC BY 4.0 license.

A detailed comparison of OME-Zarr’s superior access performance versus traditional file formats is available for those interested in technical specifics here.

Our tools to enrich the OME-Zarr Ecosystem

To facilitate broad adoption and maximize the benefits of OME-Zarr, we are actively developing a set of tools that make working with OME-Zarr easier. All of our tools are open-source and ready for everyone to use!

EuBi-Bridge
Straightforward conversion into OME-Zarr

  • Python library and command-line tool
  • Support for conversion of a wide range of formats to to OME-Zarr (v0.4 or v0.5)
  • Parallelized conversion of image collections (optionally on HPC via Slurm)
  • Optional concatenation of images along specified dimensions during conversion
  • Isotropic downscaling 
  • Conveniences for metadata update or display

ome-zarr-pyramid
A Python library to work with OME-Zarr

  • Read, process, write and create OME-Zarr datasets
  • Represents an OME-Zarr dataset as a Python object with all the pyramidal layers and metadata
  • Provides a range of simple image filtering and array processing functions directly applicable to OME-Zarr

BatchConvert
Batch conversion into OME-Zarr via Nextflow

  • Command-line tool
  • Wide range of input formats through the use of Bio-Formats backend
  • Read/write from/to local paths or s3 buckets
  • Can write directly to BioStudies user space (facilitates submission to BioImage Archive)
  • Orchestrated by Nextflow: parallelization on Slurm, potentially using docker or singularity containers

Your questions answered

We have compiled an extensive list of questions and answers covering all our data services. You’ll find everything you need to get started right there. If your question isn’t answered, please don’t hesitate to contact us.