Open Software Tools to Accelerate Adoption of OME-Zarr (BioImagingUK)
Lead Research Organisation:
UNIVERSITY OF DUNDEE
Abstract
Bioimaging measures the structure, constitution and dynamics of molecules, cells, tissues and organisms in biological systems. Data are often GBytes or Tbytes in size, cover space, time and other dimensions, include complex metadata that describe experimental setups, acquisition parameters and analytic outputs and are stored in 100s of different proprietary file formats. The scale and complexity of these data are a consistent, unsolved block to researchers making their bioimaging data public and FAIR, despite the fact that open data sharing and publication are mandated by the UKRI and many other funding agencies.
OME is a globally recognised, open source bioimage data specification and software project (https://www.openmicroscopy.org). Since 2002, OME's OME-TIFF file format and Bio-Formats file translation library have served as the global de facto standard for accessing bioimage data, but the capabilities of these tools have been surpassed by the growing volume and complexity of bioimaging datasets. Starting in 2021, OME has proposed the next generation of bioimage file formats, a flexible data contained called OME-Zarr that will serve the needs of the rapidly evolving bioimaging community and feed new public data repositories and databases. OME-Zarr has been accepted by the global bioimaging community and initial examples of software that use these formats have been built by OME and others. However, the landscape is fractured with many choices and inconsistent development in different programming environments (e.g., Python implementations are much more advanced than Java), and support for different data types in different viewing applications.
The proposal is joined by the EMBL-EBI-based BioImage Archive team, which builds and runs a reference repository for bioimage data. When combined with OME’s established expertise in building APIs and access tools for bioimaging and the Image Data Resource team from OME, the combined team is one of the strongest in the world for delivering reference software implementations for bioimaging.
In this project, the OME team will build and release reference software implementations for multi-dimensional image data viewing in web browsers (essential for adoption in corporate and hospital environments). OME will also update, document and release a reference implementation for the Java programming ecosystem, which is especially important for users that depend on ImageJ, Fiji etc. In parallel, the BioImage Archive team will begin development of a new data access mechanism for OME-Zarr and associated metadata using RO-Crate, which is a UKRI Data Research Infrastructure resource that provides a common, domain-agnostic container for research data.
Together these developments will provide critical infrastructure resources for the bioimaging community based in the UK and elsewhere, and reinforce the position of the UK as a leader in the development and provision of imaging informatics resources for the global scientific community.
OME is a globally recognised, open source bioimage data specification and software project (https://www.openmicroscopy.org). Since 2002, OME's OME-TIFF file format and Bio-Formats file translation library have served as the global de facto standard for accessing bioimage data, but the capabilities of these tools have been surpassed by the growing volume and complexity of bioimaging datasets. Starting in 2021, OME has proposed the next generation of bioimage file formats, a flexible data contained called OME-Zarr that will serve the needs of the rapidly evolving bioimaging community and feed new public data repositories and databases. OME-Zarr has been accepted by the global bioimaging community and initial examples of software that use these formats have been built by OME and others. However, the landscape is fractured with many choices and inconsistent development in different programming environments (e.g., Python implementations are much more advanced than Java), and support for different data types in different viewing applications.
The proposal is joined by the EMBL-EBI-based BioImage Archive team, which builds and runs a reference repository for bioimage data. When combined with OME’s established expertise in building APIs and access tools for bioimaging and the Image Data Resource team from OME, the combined team is one of the strongest in the world for delivering reference software implementations for bioimaging.
In this project, the OME team will build and release reference software implementations for multi-dimensional image data viewing in web browsers (essential for adoption in corporate and hospital environments). OME will also update, document and release a reference implementation for the Java programming ecosystem, which is especially important for users that depend on ImageJ, Fiji etc. In parallel, the BioImage Archive team will begin development of a new data access mechanism for OME-Zarr and associated metadata using RO-Crate, which is a UKRI Data Research Infrastructure resource that provides a common, domain-agnostic container for research data.
Together these developments will provide critical infrastructure resources for the bioimaging community based in the UK and elsewhere, and reinforce the position of the UK as a leader in the development and provision of imaging informatics resources for the global scientific community.