Our softwareTRE Agent

TRE Agent

The TRE Agent handles the communication between the Submission Layer and a Trusted Research Environment. The TRE Agent polls out to the Submission Layer, gets a task, then sends the task to an implementation of the GA4GH Task Execution Service. Once the task is finished executing the TRE Agent sends the results to Egress.

The code for TRE Agent is open source, and can be found on Github.

Core Functionality

  • Polls the Submission Layer to retrieve new tasks
  • Submits tasks to a GA4GH TES implementation
  • Monitors task execution and collects results
  • Sends task outputs to the Egress service for approval
  • Maintains communication flow between the Submission Layer, TES, and Egress

Use in Federated Research

TRE Agent is used within the Five Safes TES weave, and is situated within the Control Plane.

Integration and Extensibility

TRE Agent provides a API that can be used to integrate and build against.

Ideal Uses Cases

  • Orchestrating task execution inside a TRE
  • Connecting submission workflows to TES backends
  • Enabling controlled communication between TRE components
  • Supporting federated task execution across TREs
SubmissionEgress