The Unfunded Ally of Open Science: Research Data Management and Sharing

The consensus is clear: publicly funded research data should be shared. Globally, the benefits of sharing research data are well-recognized. It drives transparency and validation, reduces redundant efforts, accelerates discovery, increases equitability, and amplifies research impact through collaboration and efficient resource use. As a result, funding agencies worldwide are increasingly mandating data sharing. However, these mandates typically lack funding for the substantial resources required to make data sharing a reality.

Not All Data are Created Equal: The Unique Challenges of Sharing Image Data

Image data can come from diverse types of instruments and technologies such as optical and electron microscopes and pre-clinical and clinical imagers like MRI and CT scans. This makes image data uniquely complex to manage, but solving the image data management problem will give rise to solutions that can be applied to other less complex data formats. Advanced technologies generate large, multimodal, multiplexed datasets, spanning multiple targets, across different spatial-temporal scales. These datasets stem from diverse sources, including optical and electron microscopy and medical imaging, each with unique properties. Managing this complexity without global standardization is a formidable task.

Data sharing requires not only the images themselves but also comprehensive metadata detailing experiments, samples, acquisition strategies, instrument parameters, and analysis processes. Yet researchers lack: 1) the tools to collect and link this metadata to images, 2) the resources to store images securely, and 3) available and easily accessible public repositories for image data archiving and sharing.

Even when data is available, researchers face challenges in finding and reusing it due to a lack of awareness of existing repositories and inadequate search tools. AI researchers, in particular, who rely on high-quality, annotated image data to train and test models, struggle to find data that meets these needs. Without community standards, investments in curation, and software tools for metadata collection and linkage to images, the potential of scientific image data generated with public funding remains untapped.

Canada’s Data Sharing Landscape: A Critical Gap in Support and Infrastructure

In Canada, the gap in funding for data-sharing infrastructure is stark. Researchers are mandated to comply with data-sharing policies, yet they receive little support for education, data curation, storage, maintenance, tool development, or dissemination. Individual institutions are left to develop generic data management plans, a task often shouldered by overburdened research officers with limited specialized expertise. This approach leads to duplicated efforts, inconsistent practices, and fragmented solutions, which hinder effective data sharing.

Rather than isolated institutional efforts, Canada needs a unified, national approach that pools resources and expertise. A coordinated strategy would streamline processes, minimize redundancy, and allow for better resource allocation to meet shared challenges.

Leveraging Networks for Collaborative Solutions

Existing networks, such as the Canadian Network of Scientific Platforms ¹ (CNSP) and Global BioImaging ² (GBI), demonstrate the value of cross-institution collaboration in overcoming shared challenges. Events like the recent Image Data Horizons GBI meeting and the foundingGIDE³ initiative show that collaboration across borders can drive consensus and establish streamlined, effective approaches to data sharing. The power of a community driven collaborative approach was demonstrated in a 2021 Nature Methods collection focused on Reporting and Reproducibility in Microscopy.

By following the examples set by these international initiatives, Canadian institutions, imaging platforms, and government labs can pool resources and expertise to create a national strategy for standardized data-sharing processes and common robust software tools. This can be expanded beyond imaging and act as a roadmap enabling the creation of national data repositories, providing a place for researchers to put curated datasets that can be shared to maximize research impact.

A Call to Action: Building a Collaborative Research Data Ecosystem in Canada

The UNESCO recommendation on open science calls for international and multi stakeholder co-operation to reduce digital, technological and knowledge gaps⁴. UNESCO includes open science as Outcome 4 in its recently published 10-year Strategic Plan for the Implementation of the International Decade of Sciences for Sustainable Development⁵ stressing that open science needs to be widely and equitably used to democratise scientific progress and access to scientific knowledge. For Canada to meet its commitment to the sustainable development goals, including open science, there is a unique opportunity for a comprehensive strategy for sharing bioimaging data. It is essential that Canada transition from isolated institutional efforts to a coordinated, national strategy. Infrastructure networks like the Canadian Network of Scientific Platforms play a key role in bringing together the community invested in research infrastructure that generates image data. To meet data-sharing mandates effectively, Canada needs investment in infrastructure, education, and funding for human resources dedicated to consensus-building, standardization, curation, tool development and dissemination. The value of shared research data is undeniable. By building on networks and collective expertise, Canada can establish a national data-sharing infrastructure that accelerates innovation, enhances research quality, and maximizes the returns on public investments in science.

The time to act is now—our researchers, institutions, and broader community must unite to make this vision a reality and position Canada as a global leader in research data management.

References:

1- https://cnsp-rcps.ca/

2- https://globalbioimaging.org/

3- https://founding-gide.eurobioimaging.eu/

https://www.nature.com/collections/djiciihhjh

5- https://unesdoc.unesco.org/ark:/48223/pf0000383771

6- https://www.interacademies.org/sites/default/files/inline-files/Strategic%20Plan%20for%20the%20Implementation%20of%20the%20IDSSD%202024-08-25%2006_57_09.pdf