Each table build should be accompanied by metadata providing, at minimum, the following:
- The date of build
- Source versions of each ontology (or at least of PHENIO)
- Source versions of each association table
- Parameters for similarity metrics (e.g., resnik score cutoff)
- File manifest
- Description of columns in each file
Ideally this would be in a nice, flexible but consistent format like Datasheets for Datasets: https://github.com/bridge2ai/data-sheets-schema
Each table build should be accompanied by metadata providing, at minimum, the following:
Ideally this would be in a nice, flexible but consistent format like Datasheets for Datasets: https://github.com/bridge2ai/data-sheets-schema