Definitions and Data Hierarchy¶
Define Data Schema and Definitions
Overview: Define metadata structures and data types (data schema) for projects via a Editors. Specify terms and URLs to relate to ontologies in a separate GUI.
Data Schema Editor¶
Located under "Project Group," the editor allows users to:
Select, edit, or create data types.
Specify metadata in a table, including: - Description - Unit - Mandatory status - List: Link metadata to fixed lists (e.g., devices: "Oven," "Fume hood") or other data types (e.g., procedures).
Note: Metadata must be unique and include a name, tags, and comments in the general tab.
Definition Overview¶
Scientific properties require detailed definitions. For example:
temperature_A : 97
Here, temperature_A is the key, and 97 is the value. However, additional details like:
Scientific unit
Description
PURL/IRI
are essential. A complete representation:
temperature_A : {
'value': 97,
'unit': 'C',
'description': 'Temperature inside the instrument’s right side',
'PURL': 'https://www.wikidata.org/wiki/Q11466'}
Additional metadata can include:
Uncertainty: Margin of error.
Data type: Integer, text, date, etc.
Allowable values: E.g., temperatures cannot be below 0 K.
PASTA-ELN balances usability and flexibility, allowing most metadata in the value or description fields.
Handling Multiple Datasets and Data Types¶
To avoid duplication, PASTA-ELN stores description and PURL for each key in a centralized table (definitions). The value and unit can be stored independently or inherit from the master definition. If both individual and master definitions specify a unit, the individual value's unit takes precedence.
This ensures each key is unique. For example:
"Height" defined for instruments applies to samples, sharing the same description and PURL but potentially differing in units.
Unique keys are required for distinct descriptions.