incatools/ontology-term-usage

determines how an ontology term has been used

ontology_term_usage

Python wrapper to different clients to determine how a particular term is used.

One use case is for ontology developers who want to obsolete a term but need to know how it is used.

For example, if GO wanted to obsolete "apoptosis", GO editors could query for GO:0006915 and see which ontologies use it.

For more context see:

• obo#1630

Demo server (NOT STABLE):

• https://v61vbincq3.execute-api.us-east-1.amazonaws.com/staging/docs
• https://v61vbincq3.execute-api.us-east-1.amazonaws.com/staging/usage/GO:0006915

Services wrapped:

• ontobee
• ubergraph (subset of OBO)
• go (for annotations and go-cams)
• uniprot sparql endpoint (for annotations)

Planned:

• bioportal

For more details, see inline docs and tests

No CLI yet. We will implement using click

Demo:

poetry run pytest -s

Caveats

It is actually quite hard to determine whether a term is used. Many ontologies do not put their "complete" versions forward as the main URL that is loaded into services like OLS, Ontobee, Bioportal, in particular, the inter-ontology axioms this tool queries may be missing.

Ubergraph gets around this partially by having its own configuration that may load additional axioms - but ubergraph may not have all of obo

API

To run the API locally:

make app

• Docs: http://127.0.0.1:8000/docs
• Find usages: http://127.0.0.1:8000/usage/GO:0007588

Deployment on AWS via serverless:

Follow

https://adem.sh/blog/tutorial-fastapi-aws-lambda-serverless

for serverless/lambda deployment

Type

make deploy

Future plans

• It may be better to wrap this into a generic obolibrary python package

ontology-term-usage's Issues

Retire this and bring functionality into OAK

move main.py from top level

I would prefer main in ontology_term_usage.app but serverless.yaml seems to expect a top-level package