ga4gh-metadata / schemablocks Goto Github PK

@allisonheath has detailed use cases we should use as worked examples of schema blocks usage

https://github.com/ga4gh-discovery/ga4gh-discovery-search/blob/master/json_schema/schemas_source/definitions.yaml

See https://github.com/ga4gh-metadata/schemas/blob/master/main/doc/individual.md

Ideally we'd have created and updated together at the end

Could we have a dataset object of which a calmest would be a specific instantiation?

The link in this pepositories description leads to a 404.

https://ga4gh-metadata.github.io/schemas/

Guess this needs updating, as well as the link at https://ga4gh-metadata.github.io/schemablocks.html

The docs mention use of CURIEs for ontology classes, but there are some invalid CURIEs used in this context, e.g

In other cases, the CURIE has been lowercased from it's standard form:

CURIEs/URIs are case-sensitive so it's important to use the standard form. I realize there is inconsistency between OBO and identifiers.org with the former using NCIT and the latter using ncit but nobody lowercases the leading c in the local fragment portion.

Also, why not extend the use of CURIEs for all identifiers, rather than just ontology classes?

Where did 'Biocharacteristics' come from? I think you're getting this mixed-up with the intention of 'Phenotypes'.

https://github.com/ga4gh-metadata/schemas/blob/5a72ed16dc736ef41a4bf7b0cec425927f06a609/main/yaml/individual.yaml#L48-L63

https://github.com/phenopackets/phenopacket-schema/blob/4ce66acabfd3cc0f66e0a33bc5199bb80c4b2c87/src/main/proto/org/phenopackets/schema/v1/core/base.proto#L108-L170

At https://github.com/ga4gh-metadata/schemas/blob/master/main/yaml/biosample.yaml#L16 the definition of individual_id reads " In a complete data model "individual_id" represents the identifier of this biosample in the "individuals" collection."

I don't quite understand this. If we think in terms of donor_id (eg the initial patient the sample was taken from) and biosamples_id (eg SAME1234567 in EMBL-EBI BioSamples) - is this meant to be the donor_id?

Trying to align with Phenopackets, https://github.com/phenopackets/phenopacket-schema/blob/master/src/main/proto/org/phenopackets/schema/v1/core/base.proto#L108, where individual is defined as "An individual (or subject) typically corresponds to an individual human or another organism. FHIR mapping: Patient (https://www.hl7.org/fhir/patient.html)."

In DUO, you are able to add secondary and primary use cases.
Also, you might want to add two or more data use cases (cancer and diabetes research for example)

If a biosample can have one data use code, should a biosample have more than one schemablock? thats a data management headache. having a set of data use codes instead of one might solve it

@mbrush has a model for provenance in the context of variant annotation - can we port over?

IMO, "$ref" is good for defining a neat structure, but not convenient to implement. At the implementation level, I would recommend using URL as plain string.