Draft skeleton pipeline

src/ontology/mondo-ingest.Makefile

@@ -250,3 +250,14 @@ ALL_COMPONENT_SIGNTAURE_REPORTS=$(foreach n,$(ALL_COMPONENT_IDS), reports/mirror
 .PHONY: signature_reports
 signature_reports: $(ALL_MIRROR_SIGNTAURE_REPORTS) $(ALL_COMPONENT_SIGNTAURE_REPORTS)
 	echo "Finished running signature reports.."
+
+slurp/:
+	mkdir -p $@
+
+slurp/%.tsv: components/%.owl tmp/mondo.sssom.tsv reports/mirror-signature-mondo.tsv | slurp/
+	python $(SCRIPTSDIR)/migrate.py -i $< --mapping-file tmp/mondo.sssom.tsv --min-id 123000 --mondo-terms reports/mirror-signature-mondo.tsv --output $@
+	# Feel free to change the signature. Min ID is the next available Mondo ID.
+
+slurp-%: slurp/%.tsv
+
+slurp: slurp-omim

src/scripts/migrate.py

@@ -0,0 +1,37 @@
+# Migration pipeline
+
+#### THIS IS PSEUDO CODE NOT PYTHON OR ANYTHING
+#### THIS IS PSEUDO CODE NOT PYTHON OR ANYTHING
+#### THIS IS PSEUDO CODE NOT PYTHON OR ANYTHING
+#### THIS IS PSEUDO CODE NOT PYTHON OR ANYTHING
+
+#Inputs:
+source_ontology #e.g. omim
+sssom_map # e.g. mondo.sssom.tsv
+min_id
+termlist_mondo
+
+#Outputs:
+data = []
+
+for t in source_ontology:
+    if t not in sssom_map['object_id']:
+        parents = []
+        migrate = True
+        for p in oak.get_direct_parents(t):
+            if p not in sssom_map['object_id']:
+                migrate = False
+                break
+            elif sssom_map[sssom_map['object_id']==p]['predicate_id'] = 'skos:exactMatch' \
+                or sssom_map[sssom_map['object_id']==p]['predicate_id'] = 'skos:narrowMatch':
+                # In other words, if the parent is mapped, and the mapping is either exact or narrower
+                parents.append(sssom_map[sssom_map['object_id']==p]['subject_id'])
+            else:
+                # Its fine, just continue looking for other parents in this case
+        if migrate and parents:
+            next_mondo_id = determine_next_available_mondo_id(min_id, termlist_mondo) # satrting from min_id, then counting up and checking if it does not already exist.
+            label = oak.get_label(t)
+            definition = oak.get_definition(t)
+            data.append({'mondo_id':next_mondo_id, 'xref': t, 'label': label, 'definition': definition})
+
+pandas.DataFrame(data).to_csv(fn, sep="\t")