From 4256b6f6492f3dc46ff986df53f0b621391d6091 Mon Sep 17 00:00:00 2001
From: Anita Caron <anitac@ebi.ac.uk>
Date: Fri, 1 Sep 2023 16:05:59 +0100
Subject: [PATCH] Add label synonym polysemy check (#2112)

* add label synonym polysemy check

* update annotation to use obo:cl#abbreviation in sparql qc

---------

Co-authored-by: Anita Caron <anitacaron@users.noreply.github.com>
---
 src/ontology/Makefile                         |  2 +-
 src/ontology/cl-odk.yaml                      |  1 +
 .../label-synonym-polysemy-violation.sparql   | 36 +++++++++++++++++++
 3 files changed, 38 insertions(+), 1 deletion(-)
 create mode 100644 src/sparql/label-synonym-polysemy-violation.sparql

diff --git a/src/ontology/Makefile b/src/ontology/Makefile
index b604bab1e..cb1fd9e53 100644
--- a/src/ontology/Makefile
+++ b/src/ontology/Makefile
@@ -41,7 +41,7 @@ REPORT_FAIL_ON =            ERROR
 REPORT_LABEL =              
 REPORT_PROFILE_OPTS =       
 OBO_FORMAT_OPTIONS =        
-SPARQL_VALIDATION_CHECKS =  equivalent-classes owldef-self-reference nolabels pmid-not-dbxref obsolete-replaced_by obsolete-alt-id orcid-contributor illegal-annotation-property 
+SPARQL_VALIDATION_CHECKS =  equivalent-classes owldef-self-reference nolabels pmid-not-dbxref obsolete-replaced_by obsolete-alt-id orcid-contributor illegal-annotation-property label-synonym-polysemy 
 SPARQL_EXPORTS =            basic-report 
 ODK_VERSION_MAKEFILE =      v1.4.1
 
diff --git a/src/ontology/cl-odk.yaml b/src/ontology/cl-odk.yaml
index 5d8f91b7d..ace3b84e4 100644
--- a/src/ontology/cl-odk.yaml
+++ b/src/ontology/cl-odk.yaml
@@ -90,6 +90,7 @@ robot_report:
     - obsolete-alt-id
     - orcid-contributor
     - illegal-annotation-property
+    - label-synonym-polysemy
   custom_sparql_exports :
     - basic-report
 components:
diff --git a/src/sparql/label-synonym-polysemy-violation.sparql b/src/sparql/label-synonym-polysemy-violation.sparql
new file mode 100644
index 000000000..803e2201d
--- /dev/null
+++ b/src/sparql/label-synonym-polysemy-violation.sparql
@@ -0,0 +1,36 @@
+PREFIX obo: <http://purl.obolibrary.org/obo/>
+PREFIX oboInOwl: <http://www.geneontology.org/formats/oboInOwl#>
+PREFIX owl: <http://www.w3.org/2002/07/owl#>
+PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+
+SELECT ?entity ?property ?value WHERE {
+  VALUES ?property {
+    obo:IAO_0000118
+    oboInOwl:hasExactSynonym
+    rdfs:label
+  }
+  { SELECT DISTINCT ?iname (COUNT(DISTINCT ?entity) AS ?cnt) WHERE {
+      VALUES ?property {
+        obo:IAO_0000118
+        oboInOwl:hasExactSynonym
+        rdfs:label
+      }
+      ?entity ?property ?name .
+      OPTIONAL {
+        ?abbr a owl:Axiom;
+              owl:annotatedSource ?entity;
+              owl:annotatedProperty oboInOwl:hasExactSynonym;
+              owl:annotatedTarget ?name;
+              oboInOwl:hasSynonymType <http://purl.obolibrary.org/obo/cl#abbreviation>.
+      }
+
+      BIND(UCASE((?name)) AS ?iname)
+      FILTER (!isBlank(?entity) && !BOUND(?abbr))
+      FILTER NOT EXISTS { ?entity owl:deprecated true }
+    } GROUP BY ?iname HAVING (?cnt > 1)
+  }
+  ?entity ?property ?value .
+  FILTER (!isBlank(?entity))
+  FILTER NOT EXISTS { ?entity owl:deprecated true }
+  FILTER (UCASE(?value) = ?iname)
+} ORDER BY ?entity