[Workshop] Image Data Management in the Biological Sciences and Agriculture Research
Abstract
Veranstaltungsformat: Präsenz- und Online-Veranstaltung Veranstaltungsort: Hybrid event at University of Kassel (FB 10) Veranstalter: Universitätsbibliothek Forschungsdaten-Service NFDI4BIOIMAGE FAIRagro RTG Multiscale Clocks Zielgruppen: Promovierende/Doktorand:innen Absolvent:innen Wissenschaftliche Mitarbeitende Gruppenleitungen Postdoktorand:innen/Postdocs Mitarbeitende der Universität Kassel Workshop: Image Data Management in the Biological Sciences and Agriculture Research Do you work with image data and want to optimize your data management workflows? This workshop provides an overview of the key challenges associated with managing image data and presents available support services. You will learn about useful tools for image data management and how Bioimaging Research Data Management (RDM) can be integrated into your everyday practice. This workshop will not only provide knowledge in imaging data management but also create an opportunity to get to know each other, network, and exchange ideas. For the best learning outcome, we recommend in-person attendance. Content Module 1: General Introduction to Imaging Research Data Management 1.00 pm - 1.50 pm: for all audiences, incl. non-bioimaging stakeholders Module 2: NFDI4BIOIMAGE: Tools and Software for Bioimage Data Management FAIRagro: Observation data: From ground robots to UAVs 2.10 pm - 3.00 pm: for researchers at all career levels and tool providers Module 3: NFDI4BIOIMAGE: Bioimaging RDM in the everyday-practice FAIRagro: Satellite and grid data: the view from above 3.10 pm - 4.00 pm: for researchers at all career levels It is possible to attend individual modules, and your choices can be made during the registration process.
Full text
Workshop Module 2 Bioimage data management Tools and Software Michele Bortolomeazzi1, Christian Schmidt2 1 Single-cell open Lab, German Cancer Research Center (DKFZ), Heidelberg 2 Enabling Technology Department, German Cancer Research Center (DKFZ), Heidelberg With the exception of the used logos, and unless marked otherwise, the content of these slides is published under the terms of the Creative Commons Attribution 4.0 License (CC-BY) (http://creativecommons.org/licenses/by/4.0) Zenodo DOI: 10.5281/zenodo.17787401 2025/11/26
Image Data Management –Workshop Uni Kassel Status Quo of data handling? Adapted from: Schmidt C., Bortolomeazzi M., Boissonnet T., Fortmann-Grote C. et al. (2023). I3D:bio‘s OMERO trainingmaterial: Re-usable, adjustable, multi-purpose slides for local user training. Zenodo. DOI: 10.5281/zenodo.8323588If not stated otherwise, the content of this material (except for logos and the slide design) is published under CreativeCommons Attribution 4.0 license.
Image Data Management –Workshop Uni Kassel Schmidt, C., Bortolomeazzi, M., Boissonnet, T., Fortmann-Grote, C., Dohle, J., Zentis, P., Kandpal, N., Kunis, S., Zobel, T., WeidtkampPeters, S., & Ferrando-May, E. (2023). I3D:bio's OMERO training material: Re-usable, adjustable, multi-purpose slides for local user training. Zenodo. https://doi.org/10.5281/zenodo.8323588 Wet-lab scientist Pathologist Data Analyst •Performs the experiments •Acquires the images •Annotates the images •Identifies different areas •Processes the data •Builds models from annotations •How do I transfer/store the data? •File formats? •I cannot open the images? •From which sample is this image from? Challenges in handling image data
Image Data Management –Workshop Uni Kassel How do I even open the images ? Swedlow et al. (2003) Informatics and quantitative analysis in biological imaging. Science 300(5616), 100-2. DOI: 10.1126/science.1082602 CIDAS Göttingen · Josh Moore · Scalable Strategies (2025-01-23) 10.5281/zenodo. 14716546 Biological images are generally saved by the microscopes in vendor specific proprietary formats (> 150 different formats). The vendor can provide software for opening these images, but: - License? - Backwards compatibility? Need to convert the images to an open format, OME-TIFF, while preserving: - Data - Metadata
Image Data Management –Workshop Uni Kassel OMERO is a microscopy RDM platform Schmidt, C., Bortolomeazzi, M., Boissonnet, T., Fortmann-Grote, C., Dohle, J., Zentis, P., Kandpal, N., Kunis, S., Zobel, T., WeidtkampPeters, S., & Ferrando-May, E. (2023). I3D:bio's OMERO training material: Re-usable, adjustable, multi-purpose slides for local user training. Zenodo. https://doi.org/10.5281/zenodo.8323588
Image Data Management –Workshop Uni Kassel There many ways to use OMERO 62024-09-26 .insight .web
Image Data Management –Workshop Uni Kassel OMERO.insight 72024-09-26 .insight •Importing images •Viewing images •Annotating images
Image Data Management –Workshop Uni Kassel OMERO from your browser 82024-09-26 OMERO.figure OMERO.tagsearch OMERO.iviewer •Viewing images •Annotating images •Making Figures
Image Data Management –Workshop Uni Kassel https://YOUR_OMERO_WEB_ADDRESS Example: A file with a „multi-scene image“ Preview thumbnails Metadata •Managed data •Preview available •Access to metadata •User-friendly but machineaccessible How OMERO helps with data management
Image Data Management –Workshop Uni Kassel Image analysis workflow with OMERO Export Import Acquisition Import Analysis Images ROIs Metadata Tables Annotation
Image Data Management –Workshop Uni Kassel Connections with Image Analysis Software https://www.openmicroscopy.org/omero/features/analyze/ https://omero-guides.readthedocs.io/en/latest/external_tools.html Fiji CellProfiler napari QuPath Ilastick Orbit Non Exhaustive list of tools able to connect to OMERO directly or through plugins: Easing OMERO adoption with ezomero. Ratamero et al. bioRxiv 2023.06.29.546930; doi: https://doi.org/10.1101/2023.06.29.546930 OMERO python API ezOMERO rOMERO Gateway OMERO.matlab toolbox https://omero-guides.readthedocs.io/en/latest/api_usage.html
Image Data Management –Workshop Uni Kassel Fiji / ImageJ Three ways of working with Fiji and OMERO: •OMERO plugin for Fiji •Fiji Macros and OMERO.batch plugin https://github.com/GReD-Clermont/omero_batch-plugin •JIPipe https://jipipe.hki-jena.de/https://omero-guides.readthedocs.io/en/latest/fiji/docs/index.html
p 19 FiJi practical example https://www.youtube.com/watch?v=lRIOFpYYyCA
Image Data Management –Workshop Uni Kassel OMERO and Python with Jupyter Notebooks Bioimage analyst •OMERO python API: •https://docs.openmicroscopy.org/omero/5.6.0/developers/Python.html •ezomero: •https://thejacksonlaboratory.github.io/ezomero/index.html •omero-rois: •https://omero-rois.readthedocs.io/en/latest/ •omero2pandas: •https://pypi.org/project/omero2pandas/
Image Data Management –Workshop Uni Kassel Exploring the results with OMERO
Image Data Management –Workshop Uni Kassel Exploring data with OMERO https://wiki-biop.epfl.ch/data-management/omero/omero-tables#tables-on-dataset-level
Image Data Management –Workshop Uni Kassel Scatterplots with OMERO.parade https://wiki-biop.epfl.ch/data-management/omero/omero-tables#tables-on-dataset-level
Image Data Management –Workshop Uni Kassel Cooperative workflows: in OMERO terms •Images •ROIs •Annotations •Tables •Images •Metadata •Images •ROIs •Annotations
Image Data Management –Workshop Uni Kassel How about very large data? 15/12/22 Advancements in imaging technology: higher resolution, sensitivity, imaged area, … Images can have sizes in the order of multiple TBs! Mouse embryo (~2018) 5 TB, 1 volume McDole et al. Cell (10.1016/j.cell.2018.09.031) 2169 X 2048 X 988 (xyz) pixels 532 timepoints X 2 Channels X 16bit How can we deal with such large images: - Store - View + Analyse - Share -… “Monolithic vs. chunked” by Henning Falk https://github.com/zarr-developers/zarr-illustrations-falk-2022 ©2022 NumFOCUS (CC BY 4.0)
Image Data Management –Workshop Uni Kassel Converting images to OME-ZARR -Zarr bioformats2raw https://github.com/glencoesoftware/NGFF-Converter NGFF-Converter https://github.com/glencoesoftware/bioformats2raw bioformats2raw (CLI only)
Image Data Management –Workshop Uni Kassel Image Analysis with OME-ZARR n5j/BDV/MoBIE napari FiJi •Viewing •Conversion •Viewing •Conversion •Analysis And many more at: https://ngff.openmicroscopy.org/resources/tools/index.html Ilastick •Classification •Segmentation napari-ome-zarr Mastodon •Tracking •https://github.com/Bioconductor/ZarrExperiment •https://github.com/gdkrmr/zarr-R •…. •https://github.com/ome/ome-zarr-py •https://biovisioncenter.github.io/ngio/stable/ •… R libraries Python Libraries
Image Data Management –Workshop Uni Kassel Reproducible Image analysis guidelines Schmied, C., Nelson, M.S., Avilov, S. et al. Community-developed checklists for publishing images and image analyses. Nat Methods 21, 170–181 (2024). https://doi.org/10.1038/s41592-023-01987-9 Miura, K., & Nørrelykke, S. F. (2021). Reproducible image handling and analysis. The EMBO Journal, 40, e105889. https://doi.org/10.15252/embj.2020105889 https://quarep-limi.github.io/WG12_checklists_for_image_publishing/analysis_workflows/1_established_workflows.html •Document the settings and parameters used at every step •Document the version of all tools/packages and files •Environments / Containers are your friends •Include manually defined ROIs
Image Data Management –Workshop Uni Kassel Sharing Image analysis workflows Schmied, C., Nelson, M.S., Avilov, S. et al. Community-developed checklists for publishing images and image analyses. Nat Methods 21, 170–181 (2024). https://doi.org/10.1038/s41592-023-01987-9 Miura, K., & Nørrelykke, S. F. (2021). Reproducible image handling and analysis. The EMBO Journal, 40, e105889. https://doi.org/10.15252/embj.2020105889 https://quarep-limi.github.io/WG12_checklists_for_image_publishing/analysis_workflows/1_established_workflows.html •Prefer well documented and freely available open-source tools to proprietary software and undocumented packages •Use version control (git) and release your code / scripts / tools in a public repository •Github: e.g. https://github.com/labsyspharm/mcmicro •Nf-core: e.g. https://nf-co.re/mcmicro/dev/ •Make your trained models publicly available: https://bioimage.io/#/models
Image Data Management –Workshop Uni Kassel Module 3: Imaging RDM in the everyday-practice 3.10 pm - 4.00 pm: for researchers (all career levels) Conclusions •OMERO: •Central, backed-up storage for bioimaging data •Shared visualization and annotation for metadata enrichment •Figure creation and export directly with original images •Interfaces for image analysis available (e.g., Fiji, QuPath, etc.) •OME-ZARR: •Cloud-optimized file format for large image data •In active development but already widely used
Image Data Management –Workshop Uni Kassel Acknowledgments •NFDI4BIOIMAGE partners •I3D:bio •Collaboration partners •GerBI-GMB community 29/09/23 37 Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under the National Research Data Infrasstructure – NFDI 46/1 –501864659 The NFDI4BIOIMAGE consortium comprises legally independent partners and does not act autonomously towards third parties. The authors represent the contributions from their respective affiliated institutions and work together for the project. Contact NFDI4BIOIMAGE Coordination Office Inga Mohr & Christian Schmidt o[email protected] NFDI4BIOIMAGE Help Desk Data Stewardship Team https://nfdi4bioimage.de/help-desk I3D:bio Help Desk https://gerbi-gmb.de/i3dbio/i3dbiohelp-contact/ https://bsky.app/profile/nfdi4bioimage.bsky.social https://nfdi.social/@nfdi4bioimage https://www.linkedin.com/company/nfdi4bioimage/