Community-driven FAIR data management and reproducibility for the entire image-data life cycle

Biomedical advances crucially depend on the generation of high-quality Findable, Accessible, Interoperable, and Reproducible (FAIR; 10.1038/sdata.2016.18) datasets. This, in turn, requires the seamless integration of community-specified image documentation practices within the Research Data Management (RDM) processing pipelines required to ensure the execution, tracking, and documentation of the entire life cycle of data from sample preparation to publication (i.e., data provenance). This is important for microscopy, where data interpretation is crucially dependent on easy-to-use RDM software enabling the capture and reporting of knowledge that is collectively termed Image Metadata, and that consists of three key aspects: i) biological context (i.e., organism, growth conditions, sample-type); ii) image acquisition (i.e., microscope hardware/settings/quality-control); and iii) image processing (i.e., software, analysis steps). To illustrate these points, this presentation will first introduce recently published 4DN-BINA-OME-QUAREP community-driven Image Metadata specifications developed in the context of international bioimaging initiatives (10.1038/s41592-021-01327-9) and how they can be applied to typical light microscopy experiments. This will be followed by a deep dive into the importance of incorporating robust microscopy quality assessment and reporting procedures in the life cycle of light-microscopy data to ensure rigor, reproducibility, and reusability. The discussion will identify key stages in the pathway that includes image data acquisition, management, analysis, and dissemination and provide OMERO-based concrete and practical examples of how open-source tools and protocols developed by an international consortium of community initiatives led by QUality Assessment and REProducibility in Light Microscopy (QUAREP-LiMi), are being utilized in close collaboration with Canada BioImaging, at McGill University and UMass Medical School to capture and report the necessary quality-control metrics and metadata to support the reproducibility and reusability of image-based datasets. Finally, the presentation will also introduce the Micro-Meta App and MethodsJ2 software tools that allow researchers to collect detailed microscope hardware and acquisition settings metadata and generates draft methods text for publication.