Imaging-PHD: Empowering data reuse and reproducibility through microscopy-community defined Persistent Hardware Descriptors

1 Imaging-PHD: Empowering data reuse and reproducibility through microscopy-communitydefined Persistent Hardware Descriptors Lay Summary: Capturing the precise hardware configuration of an instrument that has generated scientific data is critical to understanding, reproducing and sharing the results. Though registries exist to provide unique identifiers for instrument models (e.g., RRID) and individual instruments themselves (e.g., PIDINST), no current method exists to capture the hardware configuration of an instrument with sufficient detail to promote data quality, reproducibility and FAIR sharing. This proposal introduces Persistent Hardware Descriptors (PHDs), specifically in light-microscopy, to solve the problem of inadequate metadata capture and storage, which hinders research reproducibility, data integrity, and cross-disciplinary collaboration. By providing a standardized framework for capturing this essential information, the project enhances the reliability of scientific results, makes advanced technologies more accessible, and supports long-term data preservation, ultimately advancing scientific research practices. Specifically, Imaging-PHD will (1) establish a vendor-friendly framework for describing hardware configurations that can be used in publications associated with a Digital Object Identifier (DOI), (2) create a collaborative space where configurations can be reviewed and edited before being published, including a user-friendly Graphical User Interface (GUI) for authoring microscopy configurations when vendor software does not create them automatically, and (3) launch a coordinated outreach effort to educate vendors and users, provide training for utilizing the resulting tools effectively, and generate a corpus of hardware descriptions to drive improvements in quality control and enable community-wide LLM approaches through high quality training data. All components of this proposal are designed to be interoperable and expandable for implementation at other multiple sites and within various scientific disciplines. Short Lay Summary: This proposal introduces Persistent Hardware Descriptors (PHDs) for microscopy to improve metadata capture and storage for scientific research. The objectives include establishing a standardized framework for documenting hardware configurations, creating a collaborative space for editing and publishing these configurations, and launching an outreach program to educate and train users on using these tools. This project aims to enhance research reproducibility, data integrity, and crossdisciplinary collaboration with a design that is interoperable across various research domains. 1 B. PROJECT SUMMARY Persistent Hardware Descriptors (PHDs) provide fault-tolerant, distributed, low-cost cyberinfrastructure (CI) for the citable description of the complex and changing configurations of instrument hardware captured during every data acquisition session, thus enhancing rigor, reproducibility, data reuse, and the recognition of technical staff, while democratizing access to advanced technologies and scientific information. This project addresses the critical need for improved data management through enhanced metadata capture and storage with possible applications to fundamental biology, physics, material, geological, earth, and environmental sciences and engineering. Overview: To demonstrate the feasibility of PHD, the Imaging-PHD project concentrates on light-microscopy and provides: (1) a vendor-friendly framework named Next-Generation Metadata (NGM) for automatically transferring hardware metadata from instrument manufacturers to researchers; (2) a user-friendly Graphical User Interface (GUI) named Micro-Meta Platform Frontend for authoring, reviewing and editing NGM-encoded Hardware Descriptors; (3) a service infrastructure named Micro-Meta Platform Backend for interlinking client applications with permanent identifier creation and repository deposition. MMPB will include exploratory LLM approaches for metadata extraction from instrument technical descriptions. (4) Outreach efforts to disseminate the project's results, educating the research community from multiple domains on the added value of linking hardware metadata with the persistent identification of instruments, samples, and data and training users to best utilize the generated CI. As added value, this effort will generate a corpus of hardware descriptors to drive improvements in quality control and enable community-wide LLM approaches. The flagship installation will be hosted at UMass Chan Med and provide persistent storage for Hardware Descriptors, an Elasticsearch-enabled dashboard, and smart visualization. Future installations will join a federated network, creating PHDs in microscopy and beyond. Keywords: CloudAccess, Community, Cross-disciplinary, Data reuse, Distributed, Equitable, FAIR, Hardware configurations, Machine-actionable, Microscopy Metadata, Persistent identifiers, Citable. Intellectual Merits: The capture and reporting of machine-actionable metadata are essential for researchers across multiple disciplines to generate high-quality, reproducible, and FAIR (Findable, Accessible, Interoperable, and Reusable) scientific data. This goal is furthered by documenting the full life-cycle (provenance) of data. This project contributes to producing and managing scientific outputs by expanding the current publishing infrastructure to enable persistent linking of samples, reagents, and data to rich, detailed, and citable hardware descriptors of individual instruments. The automation of these descriptions will eliminate tedious manual data entry, in turn improving reliability and efficiency. As a proof of principle, the project focuses on implementing these concepts in bioimaging. Nevertheless, all outputs will be developed to be extensible to and interoperable with other domains and fields of research. Broader Impacts: The PHD project’s broader impact centers on empowering researchers to produce data that can be harmonized, interpreted, pooled, and reanalyzed across different origins, thus expediting further scientific progress. To this aim, Imaging-PHD leverages community work carried out by the team in the context of BioImaging North America, QUality Assessment, and REProducibility for Instruments and Images in Light Microscopy, the Association of Biomolecular Resource Facilities, and the Open Microscopy Environment. By enabling public and citable descriptions of instrument instances used to produce individual datasets, Imaging-PHD will allow researchers to accurately report data generation methods in scientific publications and data repositories, thus fostering scientific validity and reliability. Imaging-PHD will continue to collaborate with global instrumentation manufacturers to improve technical standards and meet the changing needs of the research community and industry. At the same time, by sharing standardized information about research instrumentation, Imaging-PHD will improve the sustainability of the research ecosystem, ensure quality, and democratize access to scientific information and often prohibitively expensive research resources. Finally, the project will significantly impact scientists’ awareness of and access to research data management and sharing for rigor and data reuse. D. PROJECT DESCRIPTION D1. Intellectual Merits The advancement of science demands efficient re-utilization of high-quality FAIR (Findable, Accessible, Interoperable, and Reusable)1 data, which in turn is dependent on a deep understanding and trust of the methods that produced it. Since 2018, the global imaging community has built substantial traction towards unified reporting guidelines, metadata specifications and calibration protocols2–7. This is reflected in special journal issues on these topics and work by new community organizations like QUAREP6,8–10 (for abbreviations see Textbox 1). For many bioimaging organizations, transnational consortia, and scientific instrumentation companies, however, there is a common roadblock towards achieving standards and norms. The use of persistent and extensible rich metadata across these communities is needed to link samples, instrumentation, resulting data, publications and Research Data Management and Sharing (RDMS) platform in an open way that maximizes research integrity, data reuse and cross-disciplinary collaboration. Persistent Hardware Descriptors (PHD) provide this cyberinfrastructure (CI) enabling a persistent record of detailed instrument Hardware Descriptors within the publication ecosystem while expediting their discovery and re-use. Project Motivation and Impact Science-driven Scientific instruments play a pivotal role in research across diverse disciplines, spanning the life-sciences, physics, geology, chemistry and engineering. Accurate identification and description of individual instruments, enabled by globally unique Persistent Identifiers (PIDs) and metadata registries, are crucial for documenting experimental methods and linking instruments with samples, resulting data (i.e., data provenance)11–13, the scientific literature, RDMS platform, and the institutions, core facilities and individuals involved in research14–18 (Fig.1). Tracking each unique device, rather than solely recording its model (i.e., class), avoids tedious data entry, streamlines institutional inventories for strategic planning, reporting and funding purposes, promotes access to scientific information and advanced technologies, enhances the sustainability of core facilities through the recognition of their essential role in scientific advancement and optimizes resource utilization14,16,19,20. As such, instrument PIDs are essential for reproducibility, data trust and sharing21, credit attribution to core facilities, the equitable and efficient use of research resources and ensuring adherence to FAIR data principles1,14,16,17,20,22. Two projects provide key building blocks for the persistent identification and description of instruments. The RRID23–25-enabled CoreMarketplace26,68 (developed by AB & NH, governed by sample authorities such as the AntibodyRegistry.org and Cellosaurus.org; endorsed by ABRF, ARRIVE27, MDAR28, and NISO JATS29; adopted by unfund. collab. YCharOS30 and OGA31) is the platform of choice to uniquely identify core facilities32 and instrument classes (i.e., Models). This platform dovetails with the well-established RRIDs used by labs and core facilities to track metadata for reagents and samples such Textbox 1 – Glossary as antibodies, plasmids, cell lines, and organisms25,33,34 and link them to related data (e.g., validation data)30,31,35–37 and publications. PIDINST (developed by unfund. collab. Stocker)22,38 provides a robust metadata schema39 for the global identification of instrument instances, information retrieval, and workflow integration39. It has been endorsed by RDA22,40 and DRAC41–43, can be implemented44,45 as DOI46 via DataCite47, and was adopted for a wide variety of instrument types by NIST48, EUDAT B2INST49, and others50. Despite these advances, it is not possible to capture the detailed technical description of the hardware configuration of complex and evolving instruments at the time of data generation. PHD addresses this gap by leveraging robust technologies like RRID/CoreMarketplace and PIDINST, which were selected as they are the most widely adopted solutions for building PID frameworks for instrument Models and instances. The ultimate goal is to create a comprehensive provenance chain, persistently linking data to samples, instruments, analysis procedures, publications and RDMS systems. We seek funding to develop Imaging-PHD (Fig.2), focusing on light-microscopy where the potential scientific impact is massive for imaging scientists51 across different expertise levels, disciplines, geographical areas, and socioeconomic backgrounds, (see Sect. H of “Equipment, Facilities and Other Resources” - EFOR). Modern-day light-microscopy is a sophisticated, quantitative and rapidly evolving technology integral to multiple disciplines, including material sciences, nanotechnology and semiconductor engineering. In the life-sciences, microscopes are found at nearly every experimental laboratory across industry, academia and educational institutions as well as in one of the thousands of core facilities worldwide, where expert staff ensure the maintenance and utilization of advanced microscopes. This widespread distribution explains why images are included in 500 newly published articles/day, in the lifesciences alone52,53, and the expanding global microscope market size was estimated at USD 7-11 billion54,55. Despite the prevalence and potential impact of microscopes, the production of FAIR image data remains challenging. Researchers often lack sufficient expertise in microscopy56 and data management to produce reliable and quantitative image data, leading to difficulties in interpreting acquired data, reproducing experiments, and sharing results57–61. Experiments using microscopes vary in complexity and objectives, leading to different reporting and Quality Control (QC) requirements12. Additionally, metadata automatically recorded by microscopes from different manufacturers varies widely. This often leads to poor characterization of microscope configuration and performance during imaging. However, accurately tracking data provenance from acquisition through analysis and visualization is vital to ensure scientific rigor and maximizes the impact of image data.2,3,5,11,12,62,63. Although microscopes have a finite number of hardware components, the complexity arising from their combination grows quickly (e.g., EFOR, Fig.2)8,64,65. There are currently no universally adopted frameworks ensuring the capture of this complex space covering Hardware Specifications, Acquisitions Settings and Performance Metrics (collectively termed Microscopy Metadata - MM)2,12. The implementation and universal adoption of metadata frameworks and automated data management pipelines incorporating PIDs is essential for bridging the gap between sophisticated technology and researchers' expertise, thus expediting multidisciplinary scientific discovery. Innovation Textbox 2 – Imaging-PHD Use Cases Instrument Manufacturers Community Partnership: The over 650member strong QUAREP-LiMi consortium is actively collaborating with nine leading global microscopy camera manufacturers (Andor, Evident Olympus, Hamamatsu, Leica, Nikon, PCO, Roper Teledyne, Scientifica, Zeiss) to provide community-defined technical descriptions of cameras. This will be followed by similar work on other microscope components, aiming to enhance image data quality and reproducibility. The ImagingPHD framework will allow manufacturers to supply researchers with community-specified Hardware Descriptor files for instruments according to these technical specifications. This approach will facilitate transparency in instrument purchasing and ensure ongoing instrument performance monitoring after delivery. Industry partners: Microscope instrumentation vendors often need to monitor the usage patterns and performance of specific instrument models and their individual units to better understand their customers. The Imaging-PHD CI is ideally suited for these industry partners, as it allows for effective tracking of individual instrument instances as they are deployed in the field. Research Labs: Graduate students and undergrads may want to conduct complex imaging experiments that require microscope instruments not available to them. The Imaging-PHD extension of the CoreMarketplace capabilities would be transformative by enabling students to identify and access the appropriate microscope at a nearby core facility. Core Facilities: Facility managers often aim to increase revenue and demonstrate the broader impact of their microscopy core by attracting more users from within and outside their institutions. The powerful search engine to be included in the Imaging-PHD platform will transform customers' ability to find the instruments they need based on standardized technical descriptions, thereby increasing facility usage. To address these fundamental scientific challenges, Dr. Caterina Strambio-De-Castillia (CSDC) will lead the Imaging-PHD project, which involves recognized scientists in microscopy and RDMS (see Team composition and abbreviations in MCP), to deliver CI that can capture, search, and display complex but machine-actionable Hardware Descriptors of scientific instruments (Figs. 1-2) by leveraging existing key components of the data management ecosystem (PIDs, repositories). All senior members of the team play integral roles in global community initiatives that are spearheading the generation, stewardship and sharing of FAIR data. Such initiatives include 4DN66,67, ABRF68, BINA69, Canada BioImaging (CBI)70, DRAC41, German BioImaging (GerBi)71, Global BioImaging (GBI)72,73, OME74–76, the RRID-powered CoreMarketplace, PIDINST22,38,39, and QUAREP6,77 (see Sect. H, EFOR). Because of this multidisciplinary involvement, the ImagingPHD platform will be far-reaching across disciplines to promote data quality, reproducibility, and reuse as well as the efficient and equitable access to scientific information and research resources and the sustainability of the research enterprise. To maximize impact, Imaging-PHD was designed to address user-needs that were collected by team members through personal and community interactions (Textbox 2). The new platform will solve key issues within the light-microscopy domain while also laying the groundwork for a persistent mechanism to augment the current publication paradigm. PHD extends the existing network of public repositories (Figs.5&7) with semantically rich documents that can be queried and referenced. This innovative approach builds on the existing sustainability of libraries and repositories so that all PIDs will continue to resolve, independently of dedicated PHD funding, while enabling enhanced functionality like search and exploration as long as the scientific CI is maintained. D.2 Work Plan Outline Imaging-PHD simplifies the generation and publication of high-quality and FAIR microscopy datasets by enabling persistently-identified Hardware Descriptors captured using the Micro-Meta Platform to meet the community-defined 4DN-BINA-OME-QUAREP (NBO-Q)2 MM specifications (led by CSDC and DG)12. The connection with the OME project (led by JM) ensures instrument metadata is closely tied to acquired data, facilitating the smooth transfer of information from instrument manufacturers to microscope custodians and users. Leveraging the community CoreMarketplace/RRID infrastructure to persistently identify core facilities and instrument classes), PIDINST38 for instrument instances and the UMass Chan Library DataCite license for assigning DOIs, Imaging-PHD will furnish resolvable PIDs for instrument models, instances (individually configured microscopes in specific, unique locations) and Hardware Descriptors to ensure the inter-linking with institutional inventories, core facilities, microscopes, image data, reporting documents and scientific publications (Fig.1). The Imaging-PHD objectives (Fig.3) are: (1) Next-Generation Metadata (NGM): This composable metadata framework (Fig.4) enables the definition, capture and exchange of community-specified Hardware Descriptors. Based on the Resource Description Framework (RDF), NGM documents form an interlinked graph of metadata covering all states of acquisition systems. Designed for extension by instrument manufacturers, NGM can be user-authored or generated during acquisition, e.g., by embedding in the emerging bioimaging standard, Next-Generation Fig. 1: Persistent Hardware Descriptors (PHD) extend the capabilities of Research Resource Identifiers (RRID and Persistent Identifier for Instruments (PIDINST) by providing detailed configuration information. Permanently recording the configuration of unique devices rather than solely recording model information empowers both research scientists (blue) and core facilities personnel (orange) by promoting the association of instruments with: (A) the data they produce for tracking data provenance; (B) image data repositories for interoperability and FAIR data sharing; (C) performance metrics for QC; (D) scientific literature for the citations of research resources; (E) experimental protocols for reproducibility; (F) institutional inventories, funding proposals and financial reports for increasing their efficiency, accuracy and utility; (G) institutional and core facilities websites and inventories to promote the: (i) efficient selection and equitable access to resources(Brown et al., 2022; Meyn et al., 2022); (ii) financial sustainability and recognition of core facilities; and (iii) workforce educational and training. File Formats (NGFF)78,79. (2) User-friendly MicroMeta Platform Frontend (MMPF): A front-end Graphical User Interface (GUI) based on the existing Micro-Meta App (MMA)80,81 and Micro-Meta Explorer (MME)82 will be used for authoring microscope Hardware Descriptors2,12, finding microscopes with specific characteristics and comparing them across institutional and regional laboratories and facilities. (3) Re-usable Micro-Meta Platform Backend (MMPB): Service components (Fig.5) running at UMass Chan will provide pre-publication management of Hardware Descriptors associated with individual instruments and facilitate their citable publication (Fig.6) by supporting: (i) Persistently identifying instrument models and instances through integration with CoreMarketplace/RRID and PIDINST; (ii) linking persistently identified instruments with RRID-identified core facilities; (iii) minting PHDIdentifiers (PHD-IDs) implemented as DOIs for individual Hardware Descriptors through the institutional UMass Chan Library DataCite license; (iv) exploring LLM-efforts to automatically extract NBO-Q2 compliant metadata from instrument technical description files. MMPB will also leverage the Elasticsearchbased Foundry ETL-system83,84 for discovering published Hardware Descriptors online and aggregating them for search and exploration (Fig.7). (4) Coordinated Testing, Validation, Outreach and Dissemination (CTVOD): A coordinated strategy to promote adoption, training and education will be key to the broader impact of Imaging-PHD. This will include: (i) directly involving research scientists, corefacility personnel and data librarians during development to ensure the usability of produced software tools; (ii) enacting a targeted plan of action to educate users about the importance of FAIR practices, including the use of PIDs and metadata registries for the success of the research enterprise, and provide training for the effective utilization of the resulting CI; (iii) creating a collection of technical descriptions for microscopy-related hardware to facilitate future LLM-based community-learning. The Imaging-PHD project will demonstrate how vexing gaps in the current research data landscape can be bridged by offering a comprehensive CI solution for the persistent identification and tracking of Hardware Descriptors in microscopy. By providing reusable components for researchers, core facilities, and vendors to consistently document and share detailed descriptions of any hardware configuration, the PHD project drastically improves data quality and reuse, enhances collaboration, promotes the recognition of the key role of scientists and core facilities in the research enterprise, and promote a more efficient use of resources. Furthermore, Imaging-PHD is designed to mitigate interoperability risks across diverse instruments, modalities, and disciplines: (1) The rapid pace of advancements in light microscopy (e.g., spatial omics, AI-based acquisition modalities, etc.) makes it essential to develop solutions that are inherently extensible and adaptable. (2) The choice of interoperable solutions such as CoreMarketplace/RRID and PIDINST, ensures that Imaging-PHD will be widely applicable. (3) W3C standards, modular architecture, and Representational State Transfer (REST) API protocols will enhance extensibility. (4) Collaboration with imaging, PID, and RDMS initiatives (see EFOR, Sect. H) will promote consensus, adoption and interoperability across communities. In short, this proposal will contribute to the development of a cohesive and standardized approach to managing complex data, ultimately improving the overall quality and impact of research as a whole. D.3 Background The Need for FAIR Data Principles in the Life Sciences To be broadly and equitably impactful, advances in the life sciences depend on the generation of FAIR1,85 datasets that can be reused in a democratized manner. Funding agencies and publishers increasingly Fig. 2: PHDs provide a record of hardware changes over time: Beyond the out-of-the-box model and basic metadata information provided by RRID and PIDINST, respectively, each PHD faithfully records the configuration of an instrument, at a given time. For example, PHD-ID:0000 is the initial configuration on setup. PHD-ID:0001 records the replacement of the lightsource, while PHD-ID:0002 additionally records a new stage. emphasize RDMS throughout the research lifecycle. Given the growing volume and complexity of data, computational support is essential for the progress of biological research. FAIR principles facilitate automated data discovery, support reuse by other scientists across diverse backgrounds, geographical areas and financial statuses, facilitate knowledge discovery, and enhance research quality. They are imperative as data interpretation hinges on a thorough understanding of its provenance (from the sample to acquisition to quantitative result extraction)11,12,6263,86 and quality (including instrument performance)6,77. Implementing "FAIRness" (including PIDs, standardized metadata schemas and interoperable RDMS systems) is a necessity to catalyzes scientific discovery and ensures the highest research quality. Impact of Persistent Identification of research entities The use of PIDs22,25,46,87–90 is widely recognized to be key for research integrity42,48,91 and have demonstrated financial and efficiency benefits14,16. In addition, PIDs have been shown to more broadly impact equity in research resource allocation, sustainability, automation, and strategic decision making, ultimately enhancing the open use of science (see LoC from unfund. collab. Lariviers, UNESCO Chair for Open Science). However, identifiers alone do not sufficiently capture all necessary information needed to reconstruct scientific experiments, especially in microscopy. Addressing challenges in microscopy data management For reliable and reproducible extraction of quantitative information, microscopes must be well-described using community-accepted metadata specifications2,5,12,76,80,92,93. Challenges in producing high-quality FAIR image data include the limited enforcement of existing guidelines by vendors, researchers, funders, and scientific publishers and the rapid pace of imaging technical development regularly introduces new metadata needs. Tackling these challenges with CI such as the proposed Imaging-PHD is critical for ensuring that the full potential of microscopy data is realized in the life sciences and beyond. Community Initiatives Promoting Microscopy Quality, Reproducibility and Data Reuse Community initiatives like QUAREP6,77 bring together imaging scientists to promote consensus on best practices for quality, reproducibility, and data reuse in light microscopy. Involving key stakeholders (instrument manufacturers, journal editors, and professional associations)69,94–99, these initiatives advance FAIR practices through shared QC best-practices, metadata standards, and community outreach. As proofof-principle, CSDC and DG are leading an effort with nine global vendors (Andor, Hamamatsu, Roper Teldyne, Scientifica, and unfund. collabs. Evident Olympus, Leica, Nikon, PCO, Zeiss) to produce consensus camera metadata specifications based on NBO-Q2 and quality criteria (Textbox 2). These initiatives foster standardization of Hardware Descriptors and improve microscopy data quality and reuse. D.4 Preliminary Data Results from prior NSF support: NSF 1917206 to DG. Work products are teaching materials for microscopy courses including 3D printed microscopes for classroom assembly, asynchronous online learning modules, 3D microscope assembly game (online, assessment tool) and contributions to multiple publications2,3,80. This grant supported one post doc, two PhD students and six REU/RAHSS students in the lab. The REU/RAHSS program developed here will be further supported as part of this CSSI application. D.5 Cyberinfrastructure plan Building on existing, recognized capabilities DICOM, OME and the Status of image data and metadata standards in light microscopy Fig. 3: PHD Objectives Gantt Chart. When relevant a distinction is made between phases of active development versus support. Whereas standards for RDMS have been introduced for some biological research methods, such as genomics or cytometry100–104, a complete solution for optical microscopy in the life sciences does not yet exist. Though DICOM105 has succeeded in the development of standard image data formats and workflows for medical imaging technologies, this has not resulted in wide adoption in the fundamental life sciences, despite the recent addition to data repositories106 in support of some microscopy modalities107,108 like whole slide imaging109,110. A more complete covering of life science modalities is provided by OME which has worked since 2000 to establish the de facto standard for meeting the combined needs of industry and academia. Led by team member JM and unfund. collab. J. Swedlow, OME develops the OME Data Model74,75, the OME-TIFF file format111, the Bio-Formats conversion library112, the OMERO repository76 and more recently, a NGFF78,79 initiative to create a bioimaging file format standard113 (OME-Zarr79). Together these tools enable the open exchange, FAIR RDMS, and processing of data obtained from a wide range of microscopy modalities and technologies and stored in over 160 formats today. However, to fully capture the hardware metadata produced by thousands of daily users thus truly enabling a standard file format, the OME framework requires an extensible metadata framework, such as NGM, to capture a wide array of instrument types and data modalities across different scientific disciplines possibly even serving as a bridge for future harmonization with DICOM. 4DN-BINA-OME Microscopy Metadata Specifications and REMBI guidelines As a first example of building community specifications for capturing hardware metadata for the ever increasing variety of light microscopy technologies, the 4DN66,67,114 Imaging WG, under the leadership of CSDC and DG, launched a collaboration with BINA115 to expand the OME Data Model74,75 to ensure the quality, reproducibility and reuse of FISH omics experiments12,116. This resulted in the publication of the NBO MM specifications2,117, which are modular, tiered, and designed to support multiple modalities2,12. NBO has since been adopted by the QUAREP global initiative6,77,118 leading to the NBO-QUAREP (NBOQ)2 revision in partnership with microscope hardware manufacturers119,120 (Textbox 2). As such, NBO-Q fulfills one of the 8 metadata categories defined in Recommended Metadata For Biological Images (REMBI)5. Co-authored by 46 community members including CSDC, DG, and JM, these recommendations cover a broad range of imaging modalities, describe all imaging datasets deposited to BioImage Archive121, and form the basis for further multi-disciplinary harmonization through the JM-led foundingGIDE initiative122. NBO-Q builds on REMBI demonstrating the need and feasibility of NGM and MMPF80. Micro-Meta App and Explorer While the establishment of data formats and metadata standards are important, software tools that expedite the capture, visualization and exploration of MM2,12 are essential to maximize scientific impact. MMA80,81 provides a user-friendly GUI and implements NBO-Q2 to capture and organize hardware specification and image acquisition settings for image datasets produced using light microscopes. It provides an ideal platform for training users on the intricacies of hardware configurations and was adopted by 4DN66,67,114,123 and HuBMAP124–126, whose research coordination center is led by unfund. collab P. Blood (see EFOR, Sect. H). The MME82 provides complementary query and comparison functionalities to identify appropriate microscopes and evaluate their capabilities. As Javascript React components, both MMA and MME can be integrated in third-party frameworks, e.g. the 4DN-Data Portal114,123. As such, MMA and MME are ideally suited to serve as a frontend for the proposed Micro-Meta Platform. ABRF CoreMarketplace The Vermont Biomedical Research Network127 created CoreMarketplace26 in partnership with ABRF as a central registry to assign RRIDs23–25 to research core facilities across various institutions and scientific disciplines. This promotes the discoverability of facilities and instruments, attracts potential customers, encourages collaboration, expedites usage monitoring, and enhances the recognition of research activities needed for sustainability and funding. Furthermore, by leveraging the USED.it database128, CoreMarketplace extends the use of RRIDs to instrument Models (i.e., classes), thus facilitating the tracking of data quality & provenance and scientific citation. Because of its ability to link with the full imaging data provenance chain and inherent interoperability CoreMarketplace/RRID provides an ideal foundation for the Imaging-PHD project. Foundry Technology and Elasticsearch engine The FAIR Data Informatics Lab at UCSD has created the Foundry, an open source set of cloud-native applications and services based on the Elasticsearch technology that was developed for DataMed.org83. The Foundry is a simple, federated system on which the Imaging-PHD project will be based. It uses open REST interfaces and a common schema for interoperability with other RDMS systems. The Foundry is currently used by several other projects showing scalability and robustness, including: the federated RRID system (~30 databases aggregated and constantly updated), dkNet.org (a data search platform for digestive and kidney disorder researchers), and the Neuroscience Information Framework (a data search platform for neuroscience researchers). The Elasticsearch-based Foundry technology will be used to stand up a federated search system that will serve the needs of the proposed CI. Enterprise cloud-based solutions for research data at UMass Chan UMass Chan's cloud infrastructure, hosted on Amazon Web Services (AWS; for more details, see EFOR, Sect. E.1.3 and “Cloud Computing Resources” - CCR), is managed by professional staff to comply with federal and international regulations, including SOC1 Type II, SOC2 Type II, and ISO 27001. Tailored for enhanced security, the platform includes single sign-on, secure public interfaces, scalable high-availability resources, and fine-grained access controls via AWS Identity and Access Management (IAM). AWS's multi-region redundancy safeguards against data loss and disruption, while comprehensive policies ensure resource efficiency, security, and streamlined access. Together, these features underpin the PHD CI. The proposed CI incorporates a cloud risk mitigation strategy to address potential challenges such as cost fluctuations, service changes, and vendor lock-in, ensuring long-term sustainability, data portability, and operational resilience (see CCR). Collaborations among stakeholders In addition to the leadership role of CSDC, DG, JJC, JL, and JM in QUAREP6,77,118, Imaging-PHD senior personnel (see Team Composition in MCP), are internationally recognized leading members of global community initiatives that have provided LoCs (see Sect. A&H, EFOR) and together represent hundreds of core facilities and thousands of individual researchers and their needs. For over 30 years, these communities have brought together researchers, CI professionals, and hardware vendors, specifically for the purpose of fostering collaboration and decision making. Interactions with the vendors have led to detailed discussions about the configurations of microscope hardware components119,120 and their capture in the inherently multi-modal NBO-Q2,6. This work has laid the foundation for the proposed NGM specifications using open, common internet data formats (Obj.1). These discussions are currently centered on bioimaging, but the close interaction with vendors across communities incentivizes adoption of PHD in other disciplines (e.g., material sciences). Furthermore, the reliance of the Imaging-PHD CI on both the well-known PIDINST50 and CoreMarketplace/RRID frameworks, the advisory role of AB and NH, and the workshop participation of CSDC in the NSF Research Coordination Network (RCN) entitled, “FAIR Instruments and core facilities”129 mitigate the risk of overlooking the needs of stakeholders across a diverse array of domains (i.e., earth sciences, physics, astronomy, etc. See Mayernik, Mundoma, Johnson and Julian LOCs). The leadership team has met for over a year to share use-cases (Textbox 2) which have emerged from the individual organizations. This extensive preparation now ideally positions the Imaging-PHD team to ensure that this project is successful and addresses the needs of stakeholders across the research community, as demonstrated by diverse LOCs from research scientists, core facilities, instrument manufacturers, data management experts, IT and library departments, standards organizations and community organizations (see Sect. H, EFOR). Measurable Outcomes Objective 1: Definition of Next-Generation Metadata (NGM) The NGM framework (Fig.4) and reference implementations will enable manufacturers and research projects to create and extend specifications representing their key outputs. In close connection with the NGFF community, led by team member JM and J. Swedlow, this work permits the automatic capture and storage of Hardware Descriptors along with associated instrument instances and acquired data. The development of dedicated hardware specifications like NBO-Q2,6 have demonstrated the need to extend the base specifications within the community, while the Bio-Formats112 translation tool has shown that there is release. As needed, the team will employ the same mechanisms for community feedback solicitation described in 4-1. Measurable outcome: Qualitative metrics based on user surveys, in combination with continual consultations (~20/yr) and validation by the Obj.4 team as an essential service to the development team. Deliverable 4-3: Training and Education (T&E) Strategy - Objective: Educate research scientists on the importance of FAIR RDMS, PIDs, and MM documentation. Starting from year 3, the main tasks of the Obj.4 team will be the creation of outreach strategies and materials for the Imaging-PHD CI. JJC has trained >4000 individuals and found that new tasks are better learned one-on-one or by short video walkthroughs. Hence, virtual drop-in office hours will be advertised, scheduled and posted online (in anonymized fashion) via the teams’ extensive networks and community organizations (see EFOR, Sections A&H), in addition to presentations and workshops organized at conferences. DG and MH have extensive experience in the creation of independent learning modules (ILMs) for light microscopy and will support JJC in creating asynchronous online teaching modalities and FAQs. Mr. Horrman and other undergraduate students working primarily in DG lab will ensure that the produced training material is tailored to student needs. In addition, opportunities will be sought for existing COOP-students in the DG lab to contribute to the project as appropriate. Measurable outcome: Outreach strategies, ILMs, training videos, workshop/conference/community presentations. Qualitative metrics based on user surveys will be used to evaluate reaching T&E goals. Deliverable 4-4: Curated corpus of microscope technical material - Objective: As adoption of the CI will increase during year 4, end-users authoring Hardware Descriptors files will be encouraged to contribute to community-wide Machine Learning efforts by depositing related technical microscopy documentation into the MMPB database (see Obj.3). This material will be curated by the Obj.4 team to ensure accuracy and usability for future LLM training efforts. Measurable outcome: A growing corpus of curated microscopy technical material for future LLM efforts. 4-2Sustained and sustainable impacts (Obj.4): Consulting, Testing and Validation of systems directly contribute to the sustainability of the other objectives. Additional sustainable impacts, resulting from outreach and dissemination are: (1) Increased awareness: the extensive education and training effort conducted in Obj.4 will increase awareness of the importance of FAIR data, PIDs and Metadata Standards for increasing the rigor, reproducibility, and reuse of image data and for improving the societal impact, efficiency, equitability and sustainability of the research enterprise. (2) Training materials and Documentation: asynchronous T&E material (videos, slide decks, ILMs, extended documentation) will be sustained by advertising throughout the community and deposition to the BINA MicroscopyDB136, eScholarship@UMassChan137, Zenodo, and community YouTube channels. In addition to graduate microscopy course material, simplified starter-packs for K12 teachers, high-school and undergraduate students will be developed, tested and distributed via existing school outreach programs at UMass Chan, like ScienceLIVE138. (3) Community development: early NIH 4DN consortium standardization efforts have been sustained by fueling activities like BINA and QUAREP and providing the basis for community standards development2,5,80. Similarly much of the sustained impact of the products of Obj.4 will manifest through community propagation. (4) LLM training resources: The authoritative corpus of microscopy-related hardware technical descriptions will be used as an outside knowledge base for RAG-supported133 response generation in the future, creating a robust foundation for domain-specific applications and augmented search capabilities. Metrics for all Objectives: Please reference the DMCUM supplemental document. Long-term Sustainability for all Objectives: Different strategies are required to ensure the sustainable impact of the different PHD CI deliverables: (1) NGM will be adopted by OME and QUAREP, including its large industry membership. (2) MMPF is a centerpiece of CSDC’s work to expand NBO-Q2 in partnership with QUAREP6,9,119,120 and community and industry unfund. collab. (Sect. H, EFOR). As such its functionalities will remain available as part of the stand-alone Micro-Meta App80,131 and through integration in third-party data portals (e.g., OMERO, 4DN, HubMAP)123,126,132. (3) MMPB responds to the needs of diverse stakeholders in the community (Textbox 2; Sect. H, EFOR) to pool resources to build a nationwide FAIR infrastructure for imaging data as demonstrated by several opinion papers and responses to pertinent NIH Requests for Information (RFIs) 139–141. As such, Imaging-PHD will continue to be supported through UMass Chan Library and IT resources (see Sect. H, EFOR) and through its integration with CoreMarketplace and RRID as demand for interoperable, extensible, flexible, and openly accessible solutions will only increase in the future. In addition, its reusable tiered architecture will facilitate sustainability through adoption by institutions (i.e., academic libraries and manufacturers) and community initiatives (e.g., OME open-source sustainability strategy). The SciCrunch Antibody Registry (led by AB)24, iRODS142 & the NIST Genome Editing Consortium143 are successful examples of transitioning from federal funding to self-sufficiency and balancing open access with the need to generate revenue through a cost-sharing partnership with industry. This blueprint will be leveraged to develop an exploratory business model for the long-term financial sustainability of the PHD CI. Specifically, the strong relationship all members of the Imaging-PHD team have with institutions, community initiatives, and vendors will be exploited to explore the revenue potential of our scalable CI while upholding open access: (1) Manufacturer Partnerships for Revenue and Support: As cornerstone of our financial strategy we will investigate establishing partnerships in which manufacturers will contribute financially or in-kind to the CI in exchange for access. This would not only generate consistent revenue but also foster collaborative technical advancements. (2) Diversified Revenue Streams: While protecting open access for academic institutions and researchers, we will explore alternative revenue streams, including specialized services, consulting, and technical support to industry partners, as well as potential licensing opportunities for technology developed within the Imaging-PHD system. (3) Efficient Modular Design for Reduced Operational Costs: In general, the modular design of our CI allows for economical maintenance and upgrading, significantly lowering operational expenses and ensuring the efficient use of financial resources. D.6 Broader impacts Beyond addressing urgent community needs, the PHD CI impacts Sustained Scientific Innovation by enabling the generation and sharing of reproducible, well-documented datasets that are persistently linked from the full provenance chain. Additional impact dimensions are: (1) Advancing Research Integrity and Scientific Discovery: Standardized, persistent and citable Hardware Descriptors containing comprehensive metadata contribute to research integrity by minimizing errors in data acquisition and analysis thus providing a solid foundation for expediting scientific progress. (2) Facilitating Interdisciplinary Research: The adoption of FAIR data principles and the development of NGM provides a common language and framework for sharing data across different imaging modalities, data types and scientific domains. (3) Democratizing Access to Technology: The CI open-access platform for authoring and publishing Hardware Descriptors, along with powerful search capabilities, is crucial to democratize recognition for scientific development and access to needed technologies for researchers in resourceconstrained environments. To further increase the PHD CI's equitability impact, a collaboration was initiated with the Africa PID Alliance90 effort to develop PIDs for the global south produced research resources and output (see Owango LOC; Sect. H, EFOR) (4) Empowering core facility Staff: Recognizing the vital role played by core facility staff in research, the project elevates their contributions by facilitating tracking their work in reports, grants and publications. (5) Enhancing Data Management and Sharing: The project aligns with the evolving expectations of funding agencies and publishers for well-documented, reusable datasets that benefit the broader scientific community. (6) Educational and Training Opportunities: The coordinated outreach and training strategy of the project equips students and researchers with the knowledge and skills needed to effectively utilize the PHD CI, thus contributing to workforce development in an increasingly sought-after field. This educational component has a multiplier effect as trained researchers foster a culture of data stewardship and quality in their respective institutions. (7) Efficiency Benefits: Existing cost-benefit analyses14,16,91 underscore the multiplicative economic advantages of adopting PIDs and standardized metadata practices. (8) Global Collaboration: Through collaboration with international community initiatives, the project fosters a global network of stakeholders dedicated to advancing microscopy and RDMS. 1 E. REFERENCE CITED 1. Wilkinson, M. D. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 3, 160018 (2016). 2. Hammer, M. et al. Towards community-driven metadata standards for light microscopy: tiered specifications extending the OME model. Nature Methods (https://doi.org/10.1038/s41592-02101327-9) 18, 1427–1440 (2021). 3. Schmied, C. et al. Community-developed checklists for publishing images and image analyses. Nat. Methods (2023) doi:10.1038/s41592-023-01987-9. 4. Wittner, R. et al. Toward a common standard for data and specimen provenance in life sciences. Learn. Health Syst. (2023) doi:10.1002/lrh2.10365. 5. Sarkans, U. et al. REMBI: Recommended Metadata for Biological Images - realizing the full potential of the bioimaging revolution by enabling data reuse. Nature Methods (https://doi.org/10.1038/s41592-021-01166-8) 18, 1418–1422 (2021). 6. Boehm, U. et al. QUAREP-LiMi: a community endeavor to advance quality assessment and reproducibility in light microscopy. Nature Methods (https://doi.org/10.1038/s41592-021-01162-y) 18, 1423–1426 (2021). 7. Montero Llopis, P. et al. Best practices and tools for reporting reproducible fluorescence microscopy methods. Nat. Methods 18, 1463–1476 (2021). 8. Larsen, D. D., Gaudreault, N. & Gibbs, H. C. Reporting reproducible imaging protocols. STAR Protoc 4, 102040 (2023). 9. Marx, V. Caterina Strambio-De-Castillia. Nat. Methods 18, 1413 (2021). 10. Reporting and reproducibility in microscopy. Nature https://www.nature.com/collections/djiciihhjh (2021). 11. Ram, S. & Liu, J. A Semantic Foundation for Provenance Management. J. Data Semant. 1, 11–17 (2012). 12. Huisman, M. et al. A perspective on Microscopy Metadata: data provenance and quality control. arXiv [q-bio.QM]: https://arxiv.org/abs/1910.11370 (2021). 13. Wittner, R. et al. Lightweight Distributed Provenance Model for Complex Real-world Environments. Sci Data 9, 503 (2022). 14. Brown, J., Jones, P., Meadows, A. & Murphy, F. Incentives to invest in identifiers: A cost-benefit analysis of persistent identifiers in Australian research systems. Zenodo (2022) doi:10.5281/zenodo.7100578. 15. Klump, J. & Huber, R. 20 years of persistent identifiers – which systems are here to stay? Data Sci. J. 16, (2017). 16. Brown, J., Jones, P., Meadows, A. & Murphy, F. Revised cost-benefit analysis for the UK PID Support Network. Zenodo (2022) doi:10.5281/zenodo.7356219. 17. McCafferty, S. et al. Best Practices: PIDs for Instruments. Zenodo (2023) doi:10.5281/zenodo.7759201. 18. Borgman, C. L. Big Data, Little Data, No Data: Scholarship in the Networked World. (The MIT Press, 2015). doi:10.7551/mitpress/9963.001.0001. 19. Meyn, S. M. et al. Addressing the Environmental Impact of Science Through a More Rigorous, 2 Reproducible, and Sustainable Conduct of Research. J. Biomol. Tech. 33, (2022). 20. Cousijn, H. et al. Connected Research: The Potential of the PID Graph. Patterns (N Y) 2, 100180 (2021). 21. Buck, J. J. H. et al. Ocean data product integration through innovation-the next level of data interoperability. Front. Mar. Sci. 6, (2019). 22. Darroch, L. et al. Persistent Identification of Instruments (PIDINST) White Paper. https://docs.pidinst.org/en/latest/white-paper/index.html (2023). 23. Bandrowski, A. et al. The Resource Identification Initiative: A cultural shift in publishing. F1000Res. 4, 134 (2015). 24. Bandrowski, A., Pairish, M., Eckmann, P., Grethe, J. & Martone, M. E. The Antibody Registry: ten years of registering antibodies. Nucleic Acids Res. 51, D358–D367 (2023). 25. Bandrowski, A. A decade of GigaScience: What can be learned from half a million RRIDs in the scientific literature? Gigascience 11, (2022). 26. Herzog, N. The CoreMarketplace. CoreMarketplace https://coremarketplace.org/ (2018). 27. ARRIVE. 9a. Experimental procedures - explanation. ARRIVE (Animal Research: Reporting of In Vivo Experiments) https://arriveguidelines.org/arrive-guidelines/experimentalprocedures/9a/explanation. 28. Mellor, D. T. et al. Materials Design Analysis Reporting (MDAR) Checklist for Authors. Center for Open Science https://osf.io/bj3mu (2019). 29. Mulberry Technologies, Inc. NISO JATS Version 1.2d2 (ANSI/NISO Z39.96-2015)_Element: Resource Identifier. Journal Archiving and Interchange Tag Library (2018). 30. Biddle, M. S. & Virk, H. S. YCharOS open antibody characterisation data: Lessons learned and progress made. F1000Res. 12, 1344 (2023). 31. Virk, H., Krockow, E. M. & Biddle, M. OGA - only good antibodies community. OGA - only good antibodies community https://onlygoodantibodies.com/ (2024). 32. Bandrowski, A. & French, A. Understanding RRID and ROR for Facilities. Research Organization Registry (ROR) https://ror.org/blog/2024-11-26-rrid-ror-facilities/ (2024). 33. SciCrunch. RRID Antibodies. SciCrunch https://www.scicrunch.com/faq-rrid-antibody (2023). 34. SciCrunch. RRID Organism. SciCrunch https://www.scicrunch.com/faq-rrid-organism (2023). 35. SciCrunch. Purified anti-Tubulin β 3 (TUBB3) record: RRID:AB_2313773. FDI Lab SciCrunch Infrastructure http://n2t.net/RRID:AB_2313773. 36. HuBMAP Consortium. Human Reference Atlas Portal - Organ Mapping Antibody Panels (OMAPs) 5th HRA Release (v1.4). Human Atlas Portal https://humanatlas.io/omap (2023). 37. Quardokus, E. M. et al. Organ Mapping Antibody Panels: a community resource for standardized multiplexed tissue imaging. Nat. Methods 20, 1174–1178 (2023). 38. Stocker, M. et al. Persistent identification of instruments. Data Sci. J. 19, (2020). 39. Krahl, R. et al. Metadata schema for the persistent identification of instruments. Preprint at https://doi.org/10.15497/RDA00070 (2021). 40. Research Data Alliance. Research Data Alliance (RDA). rd-alliance.org https://www.rdalliance.org/ (2011). 3 41. DRAC. Digital Research Alliance of Canada. Digital Research Alliance of Canada https://alliancecan.ca/ (2021). 42. Khair, S. et al. The Current State of Research Data Management in Canada. https://alliancecan.ca/en/document/293 (2020). 43. Baker, D. et al. Research Data Management in Canada: A Backgrounder. https://alliancecan.ca/en/document/293 (2019) doi:10.5281/ZENODO.3341596. 44. Darroch, L. et al. Persistent Identification of Instruments — PIDINST documentation. https://docs.pidinst.org/en/latest/index.html (2023). 45. PID consortium. Persistent identifiers for eResearch (ePIC). https://www.pidconsortium.net/ https://www.pidconsortium.net/ (2009). 46. Paskin, N. Digital Object Identifier (DOI®) System. in Encyclopedia of Library and Information Sciences, Third Edition 1586–1592 (CRC Press, 2009). doi:10.1081/e-elis3-120044418. 47. Vierkant, P. Connecting research, advancing knowledge. DataCite https://datacite.org/ (2023). 48. Hanisch, R. et al. NIST Research Data Framework (RDaF). http://dx.doi.org/10.6028/nist.sp.150018r2 (2024) doi:10.6028/nist.sp.1500-18r2. 49. EUDAT. EUDAT - B2INST. EUDAT - B2INST https://b2inst.gwdg.de/ (2021). 50. Darroch, L. et al. PIDINST: Adoption. ReadTheDocs https://docs.pidinst.org/en/latest/adoption/index.html (2024). 51. BioImagingUK Network. What is an Imaging Scientist? Imaging Scientist http://www.imagingscientist.com/ (2018). 52. Schmied, C. & Jambor, H. K. Effective image visualization for publications – a workflow using open access tools and concepts. F1000Research 9, (2020). 53. Lee, P.-S., West, J. D. & Howe, B. Viziometrics: Analyzing Visual Information in the Scientific Literature. IEEE Transactions on Big Data 4, 117–129 (2018). 54. Grand View Research. Microscope Market Size, Share & Trends Analysis Report. Grand View Research https://www.grandviewresearch.com/industry-analysis/microscopes-industry (2022). 55. Markets and Markets. Microscopy Market. MarketsandMarkets https://www.marketsandmarkets.com/Market-Reports/world-microscopy-399.html (2022). 56. Marqués, G., Pengo, T. & Sanders, M. A. Imaging methods are vastly underreported in biomedical research. Elife 9, (2020). 57. Sheen, M. R. et al. Replication Study: Biomechanical remodeling of the microenvironment by stromal caveolin-1 favors tumor invasion and metastasis. Elife 8, (2019). 58. Prinz, F., Schlange, T. & Asadullah, K. Believe it or not: how much can we rely on published data on potential drug targets? Nat. Rev. Drug Discov. 10, 712 (2011). 59. Errington, T. M., Denis, A., Perfito, N., Iorns, E. & Nosek, B. A. Challenges for assessing replicability in preclinical cancer biology. Elife 10, (2021). 60. Errington, T. M. et al. Investigating the replicability of preclinical cancer biology. Elife 10, (2021). 61. Baker, M. & Penny, D. 1,500 scientists lift the lid on reproducibility. Nature vol. 533 452–454 Preprint at https://doi.org/10.1038/533452a (2016). 62. Wittner, R. et al. ISO 23494: Biotechnology - provenance information model for biological specimen and data. in Provenance and Annotation of Data and Processes. International Provenance 4 and Annotation Workshop (IPAW) 2020, IPAW 2021 (eds. Glavic, B., Braganholo, V. & Koop, D.) vol. 12839 222–225 (Springer, Cham, 2021). 63. Bialy, N. et al. A guide to accurate reporting in digital image processing - can anyone reproduce your quantitative analysis? J. Cell Sci. 134, (2021). 64. Gaudreault, N. et al. Illumination Power, Stability, and Linearity Measurements for Confocal and Widefield Microscopes v2. https://www.protocols.io/view/illumination-power-stability-and-linearitymeasure-b68prhvn (2022) doi:10.17504/protocols.io.5jyl853ndl2w/v2. 65. Faklaris, O. et al. Quality assessment in light microscopy for routine use through simple tools and robust metrics. J. Cell Biol. 221, (2022). 66. Dekker, J. et al. The 4D nucleome project. Nature 549, 219–226 (2017). 67. Dekker, J. et al. Spatial and temporal organization of the genome: Current state and future aims of the 4D nucleome project. Mol. Cell (2023) doi:10.1016/j.molcel.2023.06.018. 68. Hight, R. ABRF: Association of Biomolecular Resource Facilities. https://www.abrf.org/ (1989). 69. BioImaging North America. https://www.bioimagingna.org (2018). 70. Canada BioImaging. Canada BioImaging. Canada BioImaging https://www.canadabioimaging.org (2022). 71. Hanne, J. GerBI-GMB Home. German BioImaging https://gerbi-gmb.de (2016). 72. Global BioImaging. Global BioImaging. Global BioImaging https://globalbioimaging.org/ (2015). 73. Swedlow, J. R. et al. A Global View of Standards for Open Image Data Formats and Repositories. Nature Methods (https://doi.org/10.1038/s41592-021-01113-7) 18, 1440–1446 (2021). 74. Goldberg, I. G. et al. The Open Microscopy Environment (OME) Data Model and XML file: open tools for informatics and quantitative analysis in biological imaging. Genome Biol. 6, R47 (2005). 75. Swedlow, J. R., Goldberg, I. G., Brauner, E. & Sorger, P. K. Informatics and Quantitative Analysis in Biological Imaging. Science 300, 100–102 (2003). 76. Allan, C. et al. OMERO: flexible, model-driven data management for experimental biology. Nat. Methods 9, 245–253 (2012). 77. Nelson, G. et al. QUAREP-LiMi: A community-driven initiative to establish guidelines for quality assessment and reproducibility for instruments and images in light microscopy. J Microsc (https://doi.org/10.1111/jmi.13041) 284, 56–73 (2021). 78. Moore, J. et al. OME-NGFF: a next-generation file format for expanding bioimaging data access strategies. Nature Methods (https://doi.org/10.1038/s41592-021-01326-w) 18, 1496–1498 (2021). 79. Moore, J. et al. OME-Zarr: a cloud-optimized bioimaging file format with international community support. Histochem. Cell Biol. (2023) doi:10.1007/s00418-023-02209-1. 80. Rigano, A. et al. Micro-Meta App: an interactive tool for collecting microscopy metadata based on community specifications. Nat. Methods 18, 1489–1495 (2021). 81. Rigano, A. et al. Micro-Meta App - React component. (Github - https://github.com/WU-BIMAC, 2021). doi:10.5281/zenodo.7255584. 82. Rigano, A. & Strambio-De-Castillia, C. Micro-Meta Explorer - REACT Component. (Github, 2023). 83. Ozyurt, I. B. & Grethe, J. S. Foundry: a message-oriented, horizontally scalable ETL system for scientific data integration and enhancement. Database 2018, (2018). 5 84. biocaddie. Foundry-ES: Biocaddie Data Processing Pipeline. (Github, 2019). 85. Go-Fair Consortium. FAIR principles. go-fair.org https://www.go-fair.org/fair-principles/ (2017). 86. Schmied, C. et al. Community-developed checklists for publishing images and image analysis. ArXiv (2023) doi:10.48550/arXiv.2302.07005. 87. Haak, L. L., Fenner, M., Paglione, L., Pentz, E. & Ratner, H. ORCID: a system to uniquely identify researchers. Learn. Publ. 25, 259–264 (2012). 88. Shillum, C. et al. From Vision to Value: ORCID’s 2022–2025 Strategic Plan. Preprint at https://doi.org/10.23640/07243.16687207.V1 (2021). 89. Gould, M. ROR and organizational identifier interoperability in publishing systems. in ROR and Organizational Identifier Interoperability in Publishing Systems (ScienceOpen, 2023). doi:10.14293/s2199-ssp-am23-01030. 90. Ksibi, N., Owango, J. & Sara. Africa PID alliance digital object Identifiers registration concept note. Preprint at https://doi.org/10.5281/ZENODO.7924069 (2023). 91. Canadian Persistent Identifiers (PID) Advisory Committee (CPIDAC). Persistent identifier (PID) governance. Canadian Research Knowledge Network https://www.crkn-rcdr.ca/en/persistentidentifier-pid-governance. 92. Ryan, J. et al. MethodsJ2: A Software Tool to Capture and Generate Comprehensive Microscopy Methods Text and Improve Reproducibility. Nature Methods (https://doi.org/10.1038/s41592-02101290-5) 18, 1414–1415 (2021). 93. Kunis, S. et al. MDEmic: a metadata annotation tool to facilitate FAIR image data management in the bioimaging community. Nature Methods 18, 1415–1416 (2021). 94. BioImagingUK. https://bioimaginguk.org/ (2009). 95. Euro Bioimaging. https://www.eurobioimaging.eu/ (2019). 96. ELMI Initiative. European Light Microscopy Initiative. https://www.embl.org (2001). 97. German BioImaging. https://www.gerbi-gmb.de/ (2009). 98. The Royal Microscopy Society. https://www.rms.org.uk/ (1839). 99. Global BioImaging. https://www.globalbioimaging.org/ (2016). 100. Brazma, A. et al. Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365–371 (2001). 101. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012). 102. Davis, C. A. et al. The Encyclopedia of DNA elements (ENCODE): data portal update. Nucleic Acids Res. 46, D794–D801 (2018). 103. Lee, J. A. et al. MIFlowCyt: the minimum information about a Flow Cytometry Experiment. Cytometry A 73, 926–930 (2008). 104. Spidlen, J., Breuer, K. & Brinkman, R. Preparing a Minimum Information about a Flow Cytometry Experiment (MIFlowCyt) compliant manuscript using the International Society for Advancement of Cytometry (ISAC) FCS file repository (FlowRepository.org). Curr. Protoc. Cytom. Chapter 10, Unit 10.18 (2012). 105. National Electrical Manufacturers Association, Rosslyn, VA, USA. NEMA PS3 / ISO 12052, Digital Imaging and Communications in Medicine (DICOM) Standard. http://medical.nema.org/. 6 106. Fedorov, A. et al. National Cancer Institute Imaging Data Commons: Toward Transparency, Reproducibility, and Scalability in Imaging Artificial Intelligence. Radiographics 43, e230180 (2023). 107. Gupta, Y., Costa, C., Pinho, E. & Bastião Silva, L. A. A DICOM Standard Pipeline for Microscope Imaging Modalities. in 2021 IEEE Symposium on Computers and Communications (ISCC) 1–6 (2021). doi:10.1109/ISCC53001.2021.9631529. 108. Singh, R., Chubb, L., Pantanowitz, L. & Parwani, A. Standardization in digital pathology: Supplement 145 of the DICOM standards. J. Pathol. Inform. 2, 23 (2011). 109. Herrmann, M. D. et al. Implementing the DICOM Standard for Digital Pathology. J. Pathol. Inform. 9, 37 (2018). 110. Fischer, M. et al. DICOM Whole Slide Imaging for Computational Pathology Research in Kaapana and the Joint Imaging Platform. in Bildverarbeitung für die Medizin 2022 273–278 (Springer Fachmedien Wiesbaden, 2022). doi:10.1007/978-3-658-36932-3_58. 111. Besson, S. et al. Bringing Open Data to Whole Slide Imaging. in Digital Pathology 3–10 (Springer International Publishing, 2019). doi:10.1007/978-3-030-23937-4_1. 112. Linkert, M. et al. Metadata matters: access to image data in the real world. J. Cell Biol. 189, 777– 782 (2010). 113. Peng, H. Bioimage informatics: a new area of engineering biology. Bioinformatics 24, 1827–1836 (2008). 114. Reiff, S. B. et al. The 4D Nucleome Data Portal as a resource for searching and visualizing curated nucleomics data. Nat. Commun. 13, 2365 (2022). 115. Strambio-De-Castillia, C. et al. BioImaging North America (BINA): Quality Control and Data Management WG (QC-DM-WG). Bioimaging North America Site https://www.bioimagingnorthamerica.org/qc-dm-wg/ (2019). 116. Huisman, M. et al. Minimum Information guidelines for fluorescence microscopy: increasing the value, quality, and fidelity of image data. arXiv [q-bio.QM]: https://arxiv.org/abs/1910.11370v3 (2019) doi:10.48550/arXiv.1910.11370. 117. Rigano, A. et al. 4DN-BINA-OME (NBO) Tiered Microscopy Metadata Specifications - v2.01. (https://github.com/WU-BIMAC/NBOMicroscopyMetadataSpecs, 2021). doi:10.5281/zenodo.4710731. 118. Fallisch, A. QUAREP-LiMi website. Quality Assessment and Reproducibility for Instruments & Images in Light Microscopy http://QuAREP.org (2020). 119. Marx, V. Imaging standards to ease reproducibility and the everyday. Nat. Methods 19, 784–788 (2022). 120. Marx, V. The making of microscope camera standards. Nat. Methods 19, 788–791 (2022). 121. Ellenberg, J. et al. A call for public archives for biological image data. Nat. Methods 15, 849–854 (2018). 122. foundingGIDE Consortium. Founding a global image data ecosystem (GIDE). FoundingGIDE https://founding-gide.eurobioimaging.eu/ (2024). 123. Rigano, A. et al. Micro-Meta App – 4DN Data Portal. (https://data.4dnucleome.org/tools/micrometa-app, 2021). doi:10.5281/zenodo.5140157. 124. Snyder, M. P. & HuBMAP Consortium. The human body at cellular resolution: the NIH Human 7 Biomolecular Atlas Program. Nature 574, 187–192 (2019). 125. Jain, S. et al. Advances and prospects for the Human BioMolecular Atlas Program (HuBMAP). Nat. Cell Biol. 25, 1089–1100 (2023). 126. Strambio-De-Castillia, C. & Csonka, C. HuBMAP Microscope Hardware Descriptor collection. HuBMAP https://docs.hubmapconsortium.org/microscopy-metadata-links. 127. Vermont Biomedical Research Network. Vermont Biomedical Research Network https://vbrn.org/ (2020). 128. University Communications. Universal Scientific Equipment Discovery Tool (USEDit). https://myweb.fsu.edu/aglerum/usedit/usedit-about.html (2019). 129. Johnson, A. et al. FAIR Facilities and Instruments. FAIR Facilities and Instruments https://ncar.github.io/FAIR-Facilities-Instruments/ (2022). 130. Soiland-Reyes, S. et al. Packaging research artefacts with RO-Crate. Data Sci. 5, 97–138 (2022). 131. Rigano, A. et al. Micro-Meta App - Electron. (Github - https://github.com/WU-BIMAC, 2021). doi:10.5281/zenodo.4750765. 132. Rigano, A., Moore, W., Ehmsen, S., Alver, B. & Strambio-De-Castillia, C. Micro-Meta App - OMERO Plug-In. (Github - https://github.com/WU-BIMAC, 2021). doi:10.5281/zenodo.4750762. 133. Lewis, P. et al. Retrieval-augmented generation for knowledge-intensive NLP tasks. arXiv [cs.CL] (2020) doi:10.48550/ARXIV.2005.11401. 134. Bellve, K., Rigano, A., Fogarty, K. & Strambio-De-Castillia, C. Example Microscopy Metadata JSON files produced using Micro-Meta App to document the acquisition of example images using the custom-built TIRF Epifluorescence Structured Illumination Microscope. Zenodo https://doi.org/10.5281/zenodo.4891883 (2021). 135. Rigano, A. et al. Example Microscopy Metadata JSON files produced using Micro-Meta App to document example microscopy experiments performed at individual core facilities. Zenodo https://doi.org/10.5281/ZENODO.5847477 (2022). 136. Orr, V., Members of BioImaging North America & Bialy, N. MicroscopyDB. MicroscopyDB https://microscopydb.io (2022). 137. UMass Chan Lamar Soutter Library. EScholarship@UMassChan. eScholarhip @ UMassmed.edu https://repository.escholarship.umassmed.edu/ (2021). 138. ScienceLIVE outreach program. UMass Chan Medical School https://www.umassmed.edu/rti/outreach/Science-LIVE/ (2020). 139. Bagheri, N., Carpenter, A. E., Lundberg, E., Plant, A. L. & Horwitz, R. The new era of quantitative cell imaging-challenges and opportunities. Mol. Cell 82, 241–247 (2022). 140. Bajcsy, P. et al. Enabling Global Image Data Sharing in the Life Sciences. ArXiv (2024) doi: https://doi.org/10.48550/arXiv.2401.13023. 141. Bialy, N. et al. Harmonizing the generation and pre-publication stewardship of FAIR image data. arXiv (2024) doi:10.48550/ARXIV.2401.13022. 142. Russell, T. & iRODS Team. IRODS Consortium. iRODS https://irods.org/about/ (2018). 143. Maragh, S. et al. NIST Genome Editing Consortium. https://www.nist.gov/programs-projects/nistgenome-editing-consortium https://www.nist.gov/programs-projects/nist-genome-editing-consortium (2018). 8 144. Calcul Quebeq. Calcul Quebec. Calcul Quebec https://www.calculquebec.ca/ (2024). 145. Navaroli, D. M. et al. Rabenosyn-5 defines the fate of the transferrin receptor following clathrinmediated endocytosis. Proceedings of the National Academy of Sciences 109, E471–80 (2012). 146. Ott, A. et al. ABRF: Commitee on Core Rigor and Reproducibility (CCoRRE). abrf.org https://www.abrf.org/core-rigor-and-reproducibility-ccorre- (2020). 147. Abrams, B. et al. ABRF light microscopy research group study 3: The development and implementation of new tools for microscope quality assurance testing. J. Biomol. Tech. 31, S33 (2020). 148. Stack, R. F. et al. Quality assurance testing for modern optical imaging systems. Microsc. Microanal. 17, 598–606 (2011). 149. Cole, R. W. et al. International test results for objective lens quality, resolution, spectral accuracy and spectral separation for confocal laser scanning microscopes. Microsc. Microanal. 19, 1653–1668 (2013). 150. Abrams, B. et al. Tissue-like 3D standard and protocols for microscope quality management. Microsc. Microanal. 29, 616–634 (2023). 151. Cuéllar, R. & Pimentel, A. Laboratorio Nacional de Microscopía Avanzada. Laboratorio Nacional de Microscopía Avanzada https://lnma.unam.mx/wp/index.php (2020). 152. Schmidt, C. et al. Research data management for bioimaging: the 2021 NFDI4BIOIMAGE community survey. F1000Res. 11, 638 (2022). 153. Grebnev, G. & Global BioImaging. Global BioImaging training platform. Global BioImaging https://globalbioimaging.org/international-training-courses/repository (2021). 154. Rueden, C. T. et al. Scientific Community Image Forum: A discussion forum for scientific image software. PLoS Biol. 17, e3000340 (2019).