ga4gh / ga4gh/phenopacket-schema
Expand temporal fields in PhenotypicFeature to support episodic occurrences
- Dominant language
- Java
- Stars
- 101
- Forks
- 34
- PR merge metrics
- No merged PRs in 30d
Description
## Background
Certain phenotypic features are episodic and can occur multiple times throughout a patient's life (e.g., Seizures). We currently lack the granularity to specify the precise times or ages at which these multiple discrete events occur.
Currently, `PhenotypicFeature` only supports `Onset` and `Resolution`. This creates ambiguity in clinical reporting: it is unclear whether `Onset` refers exclusively to the first time a symptom was observed, or to the onset of a specific episode within a recurring condition.
## Proposed Changes:
Add new fields to `PhenotypicFeature` to properly model recurrent events, bringing it into closer alignment with how we handle `Measurement`.
**Things to discuss:**
- Naming: Should the new field be named `Time_Observed` or `Age_Observed`?
- Granularity: To capture the clinical reality accurately, we may need to introduce two or more additional fields to separate:
- Date of Onset
- Date Recorded
- Date Occurred
- Documentation: We will need to update the docs to provide strict guidance on how implementers should differentiate between the initial onset of a condition vs. the observation of subsequent episodes.
Contributor guide
Research direction
Start by reviewing the PhenotypicFeature and Measurement definitions and the existing documentation that explains onset and resolution. Resolve the field naming and temporal granularity questions with maintainers, then update the schema and guidance so initial onset and later episodic observations are unambiguous.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- backend-api-design, documentation
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100