build robot templates for ENVO subset membership

[turbomam]

This process will require some input about which EnvO classes go into which subsets. We could do that by

• adding a see also, or some similar relationship, in EnvO between the subset property (like http://purl.obolibrary.org/obo/ENVO_03605013 in https://github.com/EnvironmentOntology/envo/blob/4fd549cce4284265e06d5064f23915de35ad5bfb/src/envo/envo-edit.owl) and the IRI for the NMDC enumeration (EnvBroadScaleSoilEnum -> nmdc_sub_schema:EnvBroadScaleSoilEnum -> https://example.com/nmdc_sub_schema/EnvBroadScaleSoilEnum, not web resolvable!) or the NMDC enumeration page, like https://microbiomedata.github.io/submission-schema/EnvBroadScaleSoilEnum/
• adding a see also, or some similar annotation, in submission-schema between the enumeration and the CURIe for the EnvO subset property... but @sierra-moxon could that be overwritten by the enumeration generation process ?

  • submission-schema/project.Makefile

    Lines 188 to 213 in 55c0b28

    	########### ENV Triad PV generation ##########
    	# Specify that 'ingest-triad' is a phony target, meaning it doesn't correspond to an actual file
    	.PHONY: ingest-triad temp_target
    	WATCHED_DIR := notebooks/environmental_context_value_sets
    	# Match only .tsv files with the specific naming convention
    	WATCHED_FILES := $(shell find $(WATCHED_DIR) -type f \( \
    	-name 'post_google_sheets_*_local_scale.tsv' -o \
    	-name 'post_google_sheets_*_broad_scale.tsv' -o \
    	-name 'post_google_sheets_*_medium.tsv' \))
    	# Define an intermediate target to prevent re-triggering
    	.INTERMEDIATE: temp_target
    	# Main target to run ingestion commands, depending on the intermediate target
    	ingest-triad: temp_target
    	# Intermediate target that dynamically processes matching TSV files
    	temp_target: $(WATCHED_FILES) src/nmdc_submission_schema/schema/nmdc_submission_schema.yaml
    	@echo "Processing all matching TSV files in $(WATCHED_DIR)..."
    	@for tsv_file in $(WATCHED_FILES); do \
    	echo "Processing $$tsv_file..."; \
    	$(RUN) inject-env-triad-terms -f $$tsv_file \
    	-i src/nmdc_submission_schema/schema/nmdc_submission_schema.yaml \
    	-o src/nmdc_submission_schema/schema/nmdc_submission_schema.yaml || exit 1; \
    	done
    	@touch temp_target

Partially related to this: we don't want to insert axioms into EnvO that assert: some subject that isn't defined in EnvO is in an EnvO subset. So I think we should have access to "the latest" stable EnvO, presumably as a sem-sql file. For my development, I will be building an EnvO sem-sql file locally and placing it into my oaklib cache ~/.data/oaklib/

Presumably this will also require that robot is on the path

[turbomam]

see also

• Associate EnvO classes with NMDC environmental context subset properties EnvironmentOntology/envo#1596