Now produces spreadsheet report.

requirements.txt

@@ -0,0 +1,2 @@
+pandas
+openpyxl

src/ontology/Makefile

@@ -1,13 +1,13 @@
 ROBOT = robot
 URIBASE = http://purl.obolibrary.org/obo
 
-all: aba-uberon.owl.gz
+all: aba_uberon.owl report.xlsx
 
 IMP=true # Global parameter to bypass import generation
 MIR=true # Global parameter to bypass mirror generation
 
 #  Pulling full Uberon for now and using
-mirror/uberon.owl: 
+mirror/uberon.owl:
 	@if [ $(MIR) = true ] && [ $(IMP) = true ]; then $(ROBOT) convert -I $(URIBASE)/uberon.owl -o $@.tmp.owl && mv $@.tmp.owl $@; fi
 
 # TODO: move to making uberon slice.
@@ -19,17 +19,19 @@ sources_merged.owl:
 
 # Generate seed for pulling relevant subset of Uberon
 terms.txt: sources_merged.owl
-	$(ROBOT) query --use-graphs true -f csv -i $< --query ../sparql/terms.sparql $@
+	$(ROBOT) query --use-graphs true -f tsv -i $< --query ../sparql/terms.sparql $@
 
-# Generate slice of uberon.
+# Generate slice of uberon, removing disjoints
 # Using BOT for now - but this could potentially be more minimal
 
 uberon_slice.owl: mirror/uberon.owl terms.txt
-	$(ROBOT) extract --method BOT -i mirror/uberon.owl -T terms.txt -o $@
+	$(ROBOT) extract --method BOT -i mirror/uberon.owl -T terms.txt remove --axioms disjoint -o $@
 
 
+# merge uberon + sources, reason & relax (EC -> SC)
+
 tmp.owl: uberon_slice.owl sources_merged.owl
-	robot merge --input uberon_slice.owl --input sources_merged.owl annotate --ontology-iri $(URIBASE)/$@ -o $@
+	robot merge --input uberon_slice.owl --input sources_merged.owl reason relax annotate --ontology-iri $(URIBASE)/$@ -o $@
 
 # Make a json file for use in geneating ROBOT template
 tmp.json: tmp.owl
@@ -38,7 +40,7 @@ tmp.json: tmp.owl
 # Build robot  template - with linkouts and prefLabels
 
 ../robot_templates/linkouts.tsv: tmp.json
-	python ../scripts/gen_linkout_template.py
+	python ../scripts/gen_linkout_template.py $<
 
 # generate OWL from template
 
@@ -49,9 +51,15 @@ linkouts.owl: tmp.owl ../robot_templates/linkouts.tsv
 aba_uberon.owl: linkouts.owl
 	robot merge --input $< --input tmp.owl annotate --ontology-iri $(URIBASE)/$@ convert -f ofn -o $@
 
+report.tsv: aba_uberon.owl
+	robot query -i $< -f tsv -q ../sparql/aba_mapping_report.sparql $@
+
+report.xlsx: report.tsv
+	python3 ../scripts/mapping_spreadsheet_gen.py $< $@
+
 # Compress for release to get below GitHub file size restrictions.
-aba-uberon.owl.gz: aba_uberon.owl
-	gzip $<
+# aba-uberon.owl.gz: aba_uberon.owl
+#	gzip $<
 
 prepare_release: aba_uberon.owl
 	cp $< ../../.

src/robot_templates/README.md

@@ -0,0 +1 @@
+Add robot templates here

src/scripts/gen_linkout_template.py

@@ -2,8 +2,14 @@
 from ruamel.yaml import YAML
 from string import Template
 import pandas as pd
+import argparse
 
-f = open('../ontology/tmp.json', 'r')
+parser = argparse.ArgumentParser(description='Process some integers.')
+parser.add_argument('filepath',
+                    help='Path to json version of ontology for input')
+args = parser.parse_args()
+
+f = open(args.filepath, 'r')
 j = json.loads(f.read())
 g = j['graphs'][0]
 conf = open('../config/db_graph_atlas.yaml', 'r')

src/scripts/mapping_spreadsheet_gen.py

@@ -0,0 +1,35 @@
+import pandas as pd
+import argparse
+
+parser = argparse.ArgumentParser(description='Process some integers.')
+parser.add_argument('infile',
+                    help='Path to csv for input')
+parser.add_argument('outfile', help='Path to output ')
+args = parser.parse_args()
+
+df = pd.read_csv(args.infile, sep='\t')
+
+# remove rows with generic classification  (regional part of brain)
+unwanted_uberon_mappings = ['<http://purl.obolibrary.org/obo/UBERON_0002616>']
+df.rename(columns={'?sup': 'superclass_iri', '?supname': 'superclass_name',
+           '?sub': 'subclass_iri', '?subname': 'subclass_name'}, inplace=True)
+filtered_df = df[~df.superclass_iri.isin(unwanted_uberon_mappings)]
+
+
+def gen_ols_xls_link(row):
+    ont_name = 'uberon'
+    link = 'https://www.ebi.ac.uk/ols/ontologies/%s/terms?iri=%s' \
+           '' % (ont_name,
+        row['superclass_iri'].replace('>', '').replace('<', ''))
+    return'=HYPERLINK("%s","%s")' % (link, row['superclass_name'])
+#urllib.parse.quote()
+
+filtered_df.insert(0, 'superclass_name_linked',
+                   value=filtered_df.apply(
+                         lambda row: gen_ols_xls_link(row), axis=1))
+
+filtered_df.drop(columns=['superclass_name'], inplace=True)
+
+filtered_df.to_excel(args.outfile, header=True, index=False)
+
+

src/sparql/aba_mapping_report.sparql

@@ -0,0 +1,16 @@
+PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
+PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
+
+SELECT DISTINCT ?supname ?sup ?subname ?sub
+WHERE {
+  { ?sub rdfs:subClassOf ?sup .
+    ?sub rdfs:label ?subname .
+    ?sup rdfs:label ?supname .
+    }
+   FILTER (STRSTARTS(STR(?sup), "http://purl.obolibrary.org/obo/UBERON_"))
+   # Can't get double filter to work!
+   # FILTER (!STRENDS(STR(?sup), "UBERON_0002616")) # ignore generic uberon classification
+   FILTER (!STRSTARTS(STR(?sub), "http://purl.obolibrary.org/obo/UBERON_"))
+    }
+ORDER BY ?supname