UKRI Digital Research Infrastructure Fund has awarded £3.52 million to eight new research exemplar projects to demonstrate how emerging capabilities within and between UK Trusted Research Environments (TREs) can support real-world sensitive data research.
Funded as part of DARE UK Phase 2, the projects will run for up to 12 months (from April 2026 – 2027), applying new TRE capabilities enabled through TREvolution and tested by early adopter projects to answer complex research questions.
The awards follow a competitive funding call launched in September 2025 that invited proposals from research teams across the UK. Applications were reviewed by an independent panel, including members of the public, which considered scientific ambition, feasibility, governance, and public benefit.
Putting TRE capabilities into practice
The challenge of analysing sensitive data at scale across multiple TREs while preserving privacy and public trust is being addressed, but TRE innovations responding to the challenge still need to be explored in practice to ensure they meet the needs of the researchers using them.
The DARE UK Research Exemplars portfolio focuses on this piece, using the emerging capabilities in real-world, complex data research projects that directly test them and assess their efficacy.
Across the portfolio, the funded research projects will explore innovations enabling:
- Federated data access across multiple TREs
- Distributed analytics pipelines that run securely across environments
- Enhanced disclosure control for AI and machine learning outputs
- Secure connections to advanced computing infrastructure
- Cross-sector data linkage combining health, environmental and geographic datasets
The projects will generate practical insights into how TRE tools and methods perform in real-world research settings, identifying barriers and supporting enhancements where needed.
The funded projects
The eight funded projects are:
- CONNECT-AF will establish a federated research network linking the SAIL Databank, the East of England Secure Data Environment and the Dementias Platform UK TRE to study observational and self-reported outcomes associated with Atrial Fibrillation (AF). Using a distributed analysis approach where patient-level data remains within each TRE, the project will analyse like-for-like AF data assets across regions, demonstrating how federated research can support national learning while maintaining strong privacy protections.
FRAME: Federated Research on Anti-hypertensives Maternity Emulation Trial – £446,795
- FRAME will explore how federated health data can be used to emulate clinical trials investigating antihypertensive treatments during pregnancy. By analysing data across multiple TREs, the project aims to determine whether treatment effects vary by ethnicity and assess potential long-term developmental outcomes for children exposed to these medications before birth.
BRAID: Brain Frailty Integrated through Data Federation – £454,345
- BRAID will investigate how federated data infrastructure can support research into “brain frailty”, the cumulative impact of structural and functional brain damage associated with poorer health outcomes. Using advanced analytics and machine learning across TREs, the project will evaluate how federated approaches enable large-scale analysis of complex neurological datasets while maintaining strong governance protections.
- HEAL-Scot will demonstrate how health data can be securely linked with housing, environmental and geographic datasets across multiple Scottish Safe Havens. The project will build federated pipelines connecting DataLoch and the Health Informatics Centre in Dundee, enabling research into how living conditions, environment and location influence health outcomes and inequalities.
SAFEVID: Spasms Analysis using Federated Learning from Videos Across Multiple TREs – £451,368
- SAFEVID will explore how federated machine learning can support early detection of neurological conditions by analysing infant movement videos. By enabling models to learn across multiple TRE datasets without sharing raw video data, the project aims to improve the prediction of conditions such as infantile epileptic spasms.
MELODY: Federated Machine Learning for Dermatology – £486,340
- MELODY will develop and test federated machine learning approaches for dermatology research using clinical images from NHS Tayside and Oxford University Hospitals. By training AI models across multiple TREs without centralising data, the project aims to support the development of more inclusive and representative dermatology AI systems.
AIRR-BRIDGE: Bridging Research Infrastructure for Data Governance Extension on AIRR – £487,300
- AIRR-BRIDGE will explore how Trusted Research Environments can securely connect to the UK’s national AI supercomputers, including the Dawn and Isambard-AI systems within the AI Research Resource. The project will demonstrate how sensitive health datasets can be analysed using high-performance computing infrastructure while maintaining TRE governance standards.
TransPECT – Securing AI NLP-Transformer Models for Safe Release in TREs – £422,343
- TransPECT will address the challenge of safely exporting trained AI models from TREs. Focusing on transformer-based language models trained on sensitive free-text data, the project will develop methods and tools to evaluate disclosure risks and ensure that models can be safely released to enable further development and validation work.
Public Involvement and Engagement is embedded across all projects, with patients, communities and members of the public helping to shape governance frameworks, inform acceptable uses of data, and guide how research outputs are communicated.
By demonstrating how emerging TRE capabilities can support real research studies, the DARE UK Research Exemplars will generate evidence to inform the future development of secure, federated data research infrastructures across the UK.
Reflecting on the awards, DARE UK Interim Director, Prof Emily Jefferson, said:
“Trusted Research Environments already enable vital research using sensitive data, but many research questions require collaboration across datasets, organisations and regions. These exemplar projects will show how TRE capabilities, such as federation, AI-supported disclosure control and access to high-performance computing, can work in practice, expanding research possibilities while maintaining strong privacy protections and public trust.”
Findings and outputs from the projects will be shared with the wider data research community to help guide the future development of TRE capabilities and research projects involving data held across multiple TREs.
Public webinar
Join us for a public webinar on Thursday, 14 May 2026, from 11:00 – 12:30 to introduce the eight DARE UK Research Exemplars and what they’ve set out to achieve. The event will feature short presentations from each project, followed by a Q&A session.
During the session, project teams will present their proposed research and explain how they are applying emerging TRE capabilities to address real-world research challenges. The webinar will also highlight how Public Involvement and Engagement is helping to shape new approaches for sensitive data research.
