Dataset management
The Datasets management page provides users with the ability to register, view, manage, and update datasets along with their metadata and schema. It includes features to perform various dataset-specific actions such as creating data marts, running quality checks, and updating schemas.
Table structure
| Column | Description |
|---|---|
| Dataset ID | Unique identifier of the dataset (UUID), with an option to copy to clipboard. |
| Name | Descriptive label of the dataset. |
| Schema Name | Name of the applied schema (with namespace), copyable. |
| Schema Version | Version of the schema currently applied to the dataset. |
| Latest Available Schema | Latest schema version available for the dataset. |
| Data Model | The model the dataset adheres to (e.g., OMOP5-4), along with its plugin. |
| Actions | Dropdown menu to trigger various dataset-specific operations. |
Page actions
| Button | Functionality |
|---|---|
| Add dataset | Opens a dialog to register a new dataset into the system (more information). |
| Update dataset metadata | Enables bulk or individual metadata updates for datasets. |
Dataset action menu
Accessible via the "Select action" dropdown for each dataset:
| Action | Description |
|---|---|
| Update dataset | Edit metadata such as name, description, or tags for the dataset. |
| Create data mart | Generates a filtered or transformed version of the dataset for a specific use case. |
| Permissions | Manage access control and dataset sharing options. |
| Resources | View and manage associated files or storage locations for the dataset. |
| Update schema | Apply the latest schema version to the dataset. |
| Delete dataset | Permanently remove the dataset and all associated resources. |
| Create release (disabled) | Placeholder for releasing datasets (not currently active). |
| Run data quality | Execute quality assessment routines to validate dataset consistency (more information). |
| Run data characterization | Run characterization scripts to generate descriptive statistics and metadata. |
| Create cache | Generate cached views for faster access or downstream processing. |
note
- Schema versioning is tightly coupled with data quality and compatibility. Always verify before applying updates.
- Dataset names should be unique and semantically meaningful.
- Some actions (e.g., "Create release") may be disabled depending on platform configuration.
Permissions
- Only users with appropriate dataset or platform-level roles (e.g.,
Admin) can execute most actions. - Dataset-specific permissions (via the "Permissions" action) can be adjusted to allow individual users access to the data.
Best practices
- Regularly run data quality and characterization checks after schema updates or data mart creation.
- Use caching for datasets with heavy read operations to improve performance.
- Maintain version control and documentation externally to support traceability.