Home Blogs & News

TREvolution Project Update: Building User-Friendly Tools for Federated Research

Ross blogged about our involvement in the DARE UK TREvolution project in late 2025, and you can read the interim report that we submitted to DARE here. This blog provides an update on our work under this project, and details our remaining objectives.

DARE Interim Report:

One of the pillars of the TREvolution project is enabling federated analysis between Trusted Research Environments (TRE), specifically with federation components which are based on the GA4GH Task Execution Service (TES). As detailed in previous blogs, the open source Federated Node (FN) developed by Aridhia is based on TES.

The interim report details how we successfully integrated the FN with a DRE Workspace, and were then able to run federated tasks on data hosted in the workspace. It also detailed the current limitation of this approach, primarily the heavily manual set up process which in practice would prevent most users from adopting it.

We also detailed what steps are required to improve this, and make it possible for users to self-serve a connection between a secure workspace and federated infrastructure. In summary these were:

1. Deploying the Federated Node
2. Mapping the data
3. Securing the workspace
4. Enabling data access requests

At present we don’t believe the first can be self-service, and the fourth is already available as a feature of the DRE. We believe that there are small improvements that could be made to item three, primarily by adding a federation condition to our existing conditions and restrictions framework, but this is not central to the process as the DRE already offers users a strong suite of security controls. Therefore, we are focusing our efforts on the second item, mapping the data.

Mapping the Data:

We want to create a workflow that allows data owners to do the following without any external assistance:

• Deploy their data to a workspace database
• Associate the workspace with deployed FN
• Map the deployed data to the FN

The first is already a central feature of the DRE, so no development work was required. We believe the second can be resolved by an extension of the existing Workspace Secrets feature, and the third by developing custom apps for interacting with a Federated Node from a workspace.

Workspace Secrets:

Workspace Secrets allow users to securely store auth tokens for general use within a particular workspace, making them available to all apps within it. We are now working to extend this, making it possible to manage secrets at an organisation level within a DRE hub.

This means that where a DRE organisation has a deployed Federated Node, auth tokens for interacting with it can be stored at an org level, and made available to individual workspaces as required by the organisation’s administrators. Where this org level secret is available within a workspace users will now be able to interact with their organisations deployed FN.

Custom Apps:

Users can interact with the FN directly via its API, all its functions including managing associated datasets can be carried out this way. However, we appreciate this is not the best user experience and not all users will be comfortable with API only interactions. Therefore, we are developing apps which can be deployed into a workspace that will allow users to easily manage the process of mapping data hosted in the workspace database to the organisations FN.

As with the FN itself, when they are ready, we will make these apps available under an open source license form the Aridhia Open Source Github project.

The project runs until 31 March, when we provide our final report to DARE UK. As with the interim report we will make this available on our website and on the Aridhia Zenodo community.

If you’d like to know more about our work on data federation please contact us here.