Add mechanism to retrieve missing CHEBI ids and include them in build

[caufieldjh]

No description provided.

[caufieldjh]

This isn't working quite right yet

[caufieldjh]

Still isn't working - the terms get this far:
tmp/monarch_chebi_terms.txt -> imports/chebi_terms.txt -> imports/chebi_terms_combined.txt -> imports/merged_terms_combined.txt
but then they still aren't present in merged_import.owl.

The target is this:

$(IMPORTDIR)/merged_import.owl: $(MIRRORDIR)/merged.owl $(IMPORTDIR)/merged_terms_combined.txt
	if [ $(IMP) = true ]; then $(ROBOT) merge -i $< \
		remove  --select "<http://birdgenenames.org/cgnc/*>" remove  --select "<http://flybase.org/reports/*>" remove  --select "<http://purl.obolibrary.org/obo/CARO_*>" remove  --select "<http://purl.obolibrary.org/obo/ECTO_*>" remove  --select "<http://purl.obolibrary.org/obo/ENVO_*>" remove  --select "<http://purl.obolibrary.org/obo/FAO_*>" remove  --select "<http://purl.obolibrary.org/obo/GOCHE_*>" remove  --select "<http://purl.obolibrary.org/obo/MFOMD_*>" remove  --select "<http://purl.obolibrary.org/obo/MF_*>" remove  --select "<http://purl.obolibrary.org/obo/MOD_*>" remove  --select "<http://purl.obolibrary.org/obo/NCBITaxon_Union_*>" remove  --select "<http://purl.obolibrary.org/obo/NCIT_*>" remove  --select "<http://purl.obolibrary.org/obo/OBI_*>" remove  --select "<http://purl.obolibrary.org/obo/OGMS_*>" remove  --select "<http://purl.obolibrary.org/obo/PCO_*>" remove  --select "<http://purl.obolibrary.org/obo/PO_*>" remove  --select "<http://purl.obolibrary.org/obo/TS_*>" remove  --select "<http://www.informatics.jax.org/marker/MGI:*>" remove  --select "<https://swisslipids.org/rdf/SLM_*>" remove  --select "<https://w3id.org/biolink/vocab/*>" remove  --select "<http://www.ensembl.org/id/*>" remove  --select "<http://purl.obolibrary.org/obo/OPL_*>" remove  --select "<https://w3id.org/linkml/*>" remove  --select "<https://bioregistry.io/lipidmaps*>" remove  --select "<http://www.ncbi.nlm.nih.gov/gene/*>"  \
		extract -T $(IMPORTDIR)/merged_terms_combined.txt --force true --copy-ontology-annotations true --individuals exclude --method BOT \
		remove $(patsubst %, --term %, $(ANNOTATION_PROPERTIES)) -T $(IMPORTDIR)/merged_terms_combined.txt --select complement --select annotation-properties \
		query --update ../sparql/inject-subset-declaration.ru --update ../sparql/inject-synonymtype-declaration.ru --update ../sparql/postprocess-module.ru \
		$(ANNOTATE_CONVERT_FILE); fi

Is the ROBOT merge expecting a URI instead of a CURIE?

[matentzn]

When you say "does not work" remember that none of the IDs in this list are in that CHEBI slim yet - the PR is still open there.

[matentzn]

no need for this, ROBOT can deal with CURIEs just fine

[caufieldjh]

Great, I thought it should be able to handle CURIEs.

[caufieldjh]

When you say "does not work" remember that none of the IDs in this list are in that CHEBI slim yet - the PR is still open there.

Right - does this mean they won't be referenced at all in merged_import.owl? I was expecting them to be present there, even if they don't get propagated to the phenio.owl since they aren't in the chebi source.

[caufieldjh]

Looks like the core issue here should be covered by obophenotype/chebi_obo_slim#35, but I'm going to merge the code here as it will allow for switching over to other versions of CHEBI if needed