Adding lexical matches to uPheno pipeline

.gitignore

@@ -244,3 +244,6 @@ src/ontology/upheno.db
 src/ontology/semsim/upheno-0.4.semsimian.tsv
 src/ontology/imports/all_phenotype_terms.txt
 src/ontology/config/prefixes.csv
+src/ontology/upheno-test.db
+src/ontology/upheno-test.owl
+upheno-test.owl

src/ontology/Makefile

@@ -10,7 +10,7 @@
 # More information: https://github.com/INCATools/ontology-development-kit/
 
 # Fingerprint of the configuration file when this Makefile was last generated
-CONFIG_HASH=                1e77a49593b43cb269ead22e0f03aa7e85284fb88e0e8b01cb02248f733fccd6
+CONFIG_HASH=                bd81b2a27db2a8712339e7c75ddfc54af797ede910a207485ad4179a01f54550
 
 
 # ----------------------------------------
@@ -65,7 +65,7 @@ PATTERN_RELEASE_FILES=      $(PATTERNDIR)/definitions.owl $(PATTERNDIR)/pattern.
 MAPPINGDIR=                 ../mappings
 MAPPING_TESTER=             sssom validate
 SSSOMPY=                    sssom
-MAPPINGS=                   upheno-oba upheno-species-independent upheno-cross-species nbo-go uberon 
+MAPPINGS=                   upheno-oba upheno-species-independent upheno-lexical upheno-cross-species nbo-go uberon 
 MAPPING_RELEASE_FILES=      $(foreach n,$(MAPPINGS), $(MAPPINGDIR)/$(n).sssom.tsv)
 
 
@@ -205,7 +205,7 @@ all_subsets: $(SUBSET_FILES)
 # ----------------------------------------
 
 
-MAPPINGS =  upheno-oba upheno-species-independent upheno-cross-species nbo-go uberon
+MAPPINGS =  upheno-oba upheno-species-independent upheno-lexical upheno-cross-species nbo-go uberon
 
 MAPPING_FILES = $(patsubst %, $(MAPPINGDIR)/%.sssom.tsv, $(MAPPINGS))
 
@@ -893,6 +893,10 @@ $(MAPPINGDIR)/upheno-oba.sssom.tsv:
 $(MAPPINGDIR)/upheno-species-independent.sssom.tsv:
 	test -f $@
 
+# This mappingset is manually curated, so we only check that the file actually exists.
+$(MAPPINGDIR)/upheno-lexical.sssom.tsv:
+	test -f $@
+
 # This mappingset is manually curated, so we only check that the file actually exists.
 $(MAPPINGDIR)/upheno-cross-species.sssom.tsv:
 	test -f $@

src/ontology/config/upheno-match-rules.yaml

@@ -0,0 +1,100 @@
+rules:
+  - description: default
+    postconditions:
+      predicate_id: skos:closeMatch
+      weight: 0.0
+
+  - description: exact to exact
+    preconditions:
+      subject_match_field_one_of:
+        - oio:hasExactSynonym
+        - rdfs:label
+        - skos:prefLabel
+      object_match_field_one_of:
+        - oio:hasExactSynonym
+        - rdfs:label
+        - skos:prefLabel
+    postconditions:
+      predicate_id: skos:exactMatch
+      weight: 2.0
+
+  - description: >-
+     label to label; note this is additive with the exact to exact rule,
+      so the score just represents an additional small boost
+    preconditions:
+      subject_match_field_one_of:
+        - rdfs:label
+      object_match_field_one_of:
+        - rdfs:label
+    postconditions:
+      predicate_id: skos:exactMatch
+      weight: 0.5
+
+  - description: xref match
+    preconditions:
+      subject_match_field_one_of:
+        # - oio:hasDbXref
+        - skos:exactMatch
+      object_match_field_one_of:
+        # - oio:hasDbXref
+        - skos:exactMatch
+    postconditions:
+      predicate_id: skos:exactMatch
+      weight: 4.0
+
+  - preconditions:
+      subject_match_field_one_of:
+        - oio:hasExactSynonym
+        - rdfs:label
+      object_match_field_one_of:
+        - oio:hasBroadSynonym
+    postconditions:
+      predicate_id: skos:broadMatch
+      weight: 2.0
+
+  - preconditions:
+      subject_match_field_one_of:
+        - oio:hasExactSynonym
+        - rdfs:label
+      object_match_field_one_of:
+        - oio:hasNarrowSynonym
+    postconditions:
+      predicate_id: skos:narrowMatch
+      weight: 2.0
+
+  - synonymizer:
+      description: Remove box brackets bound info from the label.
+      match: '\[[^)]*\]'
+      match_scope: "*"
+      replacement: ""
+
+  - synonymizer:
+      description: Remove variant from the label (for matching wbphenotype).
+      match: 'aberrant'
+      match_scope: "*"
+      replacement: ""
+
+  - synonymizer:
+      description: Remove abnormally from the label.
+      match: 'abnormally'
+      match_scope: "*"
+      replacement: ""
+
+  - synonymizer:
+      description: Remove abnormal from the label.
+      match: 'abnormal'
+      match_scope: "*"
+      replacement: ""
+
+  - synonymizer:
+      description: Remove aberrant from the label (for matching dicty).
+      match: 'aberrant'
+      match_scope: "*"
+      replacement: ""
+
+  - synonymizer:
+      description: Replace "'s" by "s" in the label.
+      match: "[']s"
+      match_scope: "*"
+      replacement: "s"
+

src/ontology/upheno-odk.yaml

@@ -265,6 +265,8 @@ sssom_mappingset_group:
       maintenance: manual
     - id: upheno-species-independent
       maintenance: manual
+    - id: upheno-lexical
+      maintenance: manual
     - id: upheno-cross-species
       maintenance: manual
     - id: nbo-go

src/ontology/upheno.Makefile

@@ -1,13 +1,34 @@
 
 SSPOS = mp hp zp dpo wbphenotype xpo planp ddpheno fypo apo mgpo phipo
 
+%.db: %.owl
+	@rm -f $*.db
+	@rm -f .template.db
+	@rm -f .template.db.tmp
+	@rm -f $*-relation-graph.tsv.gz
+	RUST_BACKTRACE=full semsql make $*.db -P config/prefixes.csv
+	@rm -f .template.db
+	@rm -f .template.db.tmp
+	@rm -f $*-relation-graph.tsv.gz
+	@test -f $*.db || (echo "Error: File not found!" && exit 1)
+
+.PRECIOUS: %.db
+
 ###############################
 #### Mappings and reports #####
 ###############################
 
 $(TMPDIR)/upheno-species-lexical.csv: upheno.owl
 	$(ROBOT) query -f csv -i $< --query ../sparql/phenotype-classes-labels.sparql $@
 
+$(TMPDIR)/upheno-species-lexical-oak.sssom.tsv: upheno.db
+	runoak -i sqlite:$< lexmatch -R config/upheno-match-rules.yaml -o $@
+
+# Currently only the oak lexical match is used for the upheno-lexical.sssom.tsv
+# Should this be a merge of the upheno-species-independent.sssom.tsv and oak lexical match?
+$(MAPPINGDIR)/upheno-lexical.sssom.tsv: $(TMPDIR)/upheno-species-lexical-oak.sssom.tsv
+	sssom filter $< -o $@ --predicate_id skos:exactMatch
+
 $(TMPDIR)/upheno-mapping-logical.csv: upheno.owl
 	$(ROBOT) query -f csv -i $< --query ../sparql/cross-species-mappings.sparql $@
 	#echo "SKIP $@"
@@ -18,8 +39,8 @@ $(REPORTDIR)/upheno-associated-entities.csv: upheno.owl
 	#$(ROBOT) query -i tmp/mat_upheno.owl -f csv --query ../sparql/phenotype_entity_associations.sparql $@
 	touch $@
 
-$(MAPPINGDIR)/upheno-oba.sssom.tsv: upheno.owl
-	robot query -i $< --query ../sparql/pheno_trait.sparql $@
+$(MAPPINGDIR)/upheno-oba.sssom.tsv: #upheno.owl
+	robot query -i upheno.owl --query ../sparql/pheno_trait.sparql $@
 	sed -i 's/[?]//g' $@
 	sed -i 's/<http:[/][/]purl[.]obolibrary[.]org[/]obo[/]/obo:/g' $@
 	sed -i 's/>//g' $@
@@ -56,19 +77,6 @@ $(MAPPINGDIR)/upheno-cross-species.sssom.tsv: $(TMPDIR)/upheno-species-lexical.c
 $(MAPPINGDIR)/%.sssom.owl: $(MAPPINGDIR)/%.sssom.tsv
 	sssom convert -i $< -O owl -o $@
 
-%.db: %.owl
-	@rm -f $*.db
-	@rm -f .template.db
-	@rm -f .template.db.tmp
-	@rm -f $*-relation-graph.tsv.gz
-	RUST_BACKTRACE=full semsql make $*.db -P config/prefixes.csv
-	@rm -f .template.db
-	@rm -f .template.db.tmp
-	@rm -f $*-relation-graph.tsv.gz
-	@test -f $*.db || (echo "Error: File not found!" && exit 1)
-
-.PRECIOUS: %.db
-
 semsim/upheno-0.4.semsimian.tsv: upheno.db $(IMPORTDIR)/all_phenotype_terms.txt
 	runoak --stacktrace -vvv -i semsimian:sqlite:upheno.db similarity -p i \
 	--set1-file $(IMPORTDIR)/all_phenotype_terms.txt \