Provide Data to OXFORDIA¶

This page is for data administrators — the people who decide what data lives on an OXFORDIA Node and who is allowed to compute what against it.

Data administrators work through the node's web admin interface, logged in as an authorized administrator on their institution's identity provider.

1. Log in to the dashboard¶

Navigate to the URL of your institution's OXFORDIA Node. Log in with your institutional account.

The dashboard lists all datasets currently loaded on the node.

The OXFORDIA data dashboard showing two datasets with Copy data url, View Admin Access, and View Researcher Access actions

2. Upload a dataset¶

Click + Create to upload a new dataset.

Currently, data must be provided as a standardized CSV matching the schema expected by the relevant Data Plugin. More flexible ingestion methods are planned for future releases.

Once uploaded, the CSV is processed, converted to RDF, and stored in the node's triplestore. The dataset becomes visible in the dashboard and can be viewed in the plugin's data table view.

Nemaline Myopathy Assessments data table showing participant records with columns for cluster, genetic group, baseline age, ambulation, MFM scores, and KM event data

To allow other administrators at your institution to manage a dataset, use Resource Sharing.

Open the dataset, click Views, and open the Resource Sharing dialog. Add collaborators by WebID or contact. You can set their access level (Viewer, Editor, etc.).

Resource Sharing Preferences dialog showing Public Access, Authenticated Agents, and Individual Agents fields with WebID entry

Resource Sharing vs. Statistic Access Rules

Resource Sharing grants access to the raw data. Use it only for colleagues at your own site who need to manage the dataset. To allow external researchers to compute against the data without seeing individual records, use Statistic Access Rules instead.

4. Author Statistic Access Rules¶

The most important configuration on any OXFORDIA Node is the Statistic Access Rule (SAR) editor. This is where your institution makes its actual policy choices: which researchers may compute which statistics against which fields, and under what constraints.

Open a dataset and click Edit Statistic Access Rule.

What a rule specifies¶

Each SAR covers one dataset and contains:

Field	Description
Allowed Agents	WebIDs of researchers (or named groups) who may query this dataset
Statistic	Which computation is authorized — `mean`, `kaplan-meier`, etc.
Graph Path	Which field the statistic may operate on — e.g., `BaselineAge`
Min Count	Minimum number of records that must match before any result is returned

Example policy¶

Researchers Alice and Bob may compute means and Kaplan–Meier curves against BaselineAge and EventTime, as long as at least 10 records match. No one may retrieve a single row of data.

This is the policy posture that most rare-disease consortia would want to adopt. OXFORDIA makes it the default.

Statistic Access Rule editor showing Allowed Agents field and mean statistic configuration with graph paths BaselineAge, KaplanMeierEvent, and KaplanMeierTime

Adding a rule¶

Enter the researcher's WebID in the Allowed Agents field and click Add Agent.
Under the statistic section (e.g., 1. mean), set the Graph Path and Min Count.
Repeat for each statistic and graph path you wish to authorize.
Click Save Changes.

Min Count¶

The minCount constraint prevents a rule from being used to read out individual records. If a query would return a result derived from fewer than minCount records, the node rejects it and returns nothing.

Set minCount to at least 5 for any externally-accessible dataset. For sensitive cohorts or tail analyses (survival curves), use a higher threshold.

5. Copy the dataset URL¶

Each dataset has a stable URL that researchers use to query it. Click Copy data url next to the dataset in the dashboard. Share this URL with researchers who will be querying your node.