HDR-UK Federated Analytics is a research collaboration within HDR UK’s Infrastructure and Services programme, funded by the second quinquennial review period (HDR QQ2).

Health Data Research UK (HDR UK) is a national institute dedicated to improving health outcomes and advancing medical research through the use of data. Established in 2018, HDR UK is a partnership between leading universities, research institutions, and the National Health Service (NHS) in the UK. The primary mission of HDR UK is to harness the power of health data to drive scientific discoveries, develop innovative treatments and interventions, and ultimately improve patient care. By integrating and analysing vast amounts of health data, including electronic health records, genomic data, and other relevant information, HDR UK aims to generate insights that can transform healthcare delivery, policy, and research.

The University of Manchester, the University of Nottingham, Swansea University and the University of Dundee have established a collaborative network to explore innovative ways to use health data for research and healthcare improvement. These initiatives involve developing new analytical techniques, data linkage methods, and tools for data sharing while ensuring patient privacy and data security.

eScience Lab involvement

At the University of Manchester, we want to better understand the challenges of computational reproducibility and FAIR data sharing within HDR UK federated data infrastructures, especially focussing on technology potential, limitations, integrity measures and handling of sensitive data.

We are building on the outputs from the TRE-FX project including the Five Safes RO-Crate profile. A PhD studentship was granted.

We’re leading WP3 on FAIR and Transparency.

Our key objective is to improve transparency around the technical engineering and governance of federated analytics to make it easier for developers from other disciplines to get involved. Integrating FAIR data capabilities into the technical framework and architecture supports the governance requirements of Trusted Research Environments (TREs). We propose to use Research Objects, packaged as RO-Crates, to organise and describe the inputs, methods and outputs of an analysis. By cataloguing this information, a TRE provider has a record of the data provenance, context and understanding of how to reproduce a particular analysis of a particular dataset.

  • Objective 1: Observe, collect and review FAIR practices (data, tools, services, infrastructure) present in the key use cases within the Trusted Research Environment (TRE) ecosystem.
  • Objective 2: Practical FAIR for work to be implemented or adopted
  • Objective 3: Co-create RO-Crate profiles to facilitate interoperability: overarching RO-Crate profiles that accommodate community-specific solutions
  • Objective 4: Document and Disseminate the Framework for FAIR and RO-Crate usage

References

Kay Snowley, Lara Edwards, Ben Crosby, Helen Tatlow (2023):
Integrating Our Community. Year 1
Health Data Research UK (report) https://www.hdruk.ac.uk/wp-content/uploads/2023/10/Integrating-Our-Community_v1-Oct-2023-compressed.pdf