Vocabularies are flat controlled lists managed under the Vocabularies section in the CMS. Unlike cross-collection links, vocabulary values are stored as plain strings — not UUIDs. Renaming a vocabulary term does not automatically update existing entries that use it.
| Vocabulary | File | Structure | Used in |
|---|---|---|---|
| Datatypes | content/data/datatypes.yaml |
string list | Datasets |
| Sources / Hosts | content/data/sources.yaml |
string list | Datasets, Subtypes |
| Data Origins | content/data/data_origins.yaml |
string list | Datasets |
| Detection Methods | content/data/detection_methods.yaml |
string list | Datasets |
| Countries | content/data/countries.yaml |
string list | Datasets, Research Labs |
File Format
Every vocabulary file is a simple string list:
datatypes:
- 18s
- Genome
- Transcriptome
Entries store the string value directly (e.g. a dataset's countries holds country names).