People and AI see only the data they are approved to use.
Access in Foundry120 is granted per study and per lab group. Helix inherits the same limits. Sample history cannot be edited or deleted.
Three levels, granted per study and per lab group.
Access is granted study by study and lab by lab, so two people with the same job title can hold different access. A coordinator on MUSIC works with MUSIC samples and cannot see MARVEL's. A lab operator manages the boxes in their own lab. Every study and every lab group uses the same three levels.
Viewer
Read sample records and study IDs on a study, or boxes and experiments in a lab group. A viewer cannot change anything.
Operator
Create and update records: samples and study IDs on a study, or boxes, experiments, and lab-governed datasets in a lab group.
Admin or lead
Manage access to that study or lab group, review its audit trail, and control who can use its datasets. Each study and lab group has its own admins.

The AI assistant runs inside the same limits.
Helix reads what your account is approved to read, and its working can be inspected so you can verify the answer.
- Helix answers from your samples, your approved datasets, and your documents
- Every answer names the dataset it queried and the filters it applied, so you can check the working
What gets recorded.
Research integrity depends on being able to say what happened, when, and who did it.
- Every sample change is logged with the time, the account that made it, and the value before and after
- Dataset access is recorded the same way whether the data was read through Helix or manual download
Your data stays yours.
Foundry120 holds your research data; it does not take ownership of it. You can take it out at anytime in open formats.
- Research teams keep full ownership of their samples and datasets
- Export sample and study-tracking records as CSV at any time
- Download every dataset file your account is approved to see
Where your data lives.
- Core platform data is held in the University of Glasgow's Microsoft Azure environment, in the UK South region
- Custom deployment possible where the arrangement fits
- Helix AI processing runs in Microsoft Azure regions in the EU
- Research data is not used to train external AI models
- Pseudonymised research data only
Interested in using Foundry120, or exploring a research collaboration?
We’d be glad to hear from research groups and institutions exploring the platform, or holding data that has outgrown spreadsheets and shared drives.
Request a platform walkthrough →