Disease-Gene Exclusions

- Update: Added Sabrina's ORCID to exclusions file.
- Bug fix: Was still not filtering exlcusions correctly; was missing a logical condition.
- Bug fix?: Added RO:0003302 entries to disease-gene-relationships (.sparql / .tsv). I think these were previously left out by mistake.
- Update: A comment to be more clear

data/exclusions-disease-gene.tsv

@@ -1,8 +1,8 @@
 omim_id	mondo_id	mondo_label	orcid	exclusion_reason_comment
-OMIM:603956	MONDO:0002974	cervical cancer'		evidence of various genes involved
-OMIM:619151	MONDO:0030894	AMED syndrome, digenic'		digenic
-OMIM:158901	MONDO:0008031			digenic
-OMIM:108770	MONDO:0007171	atrial standstill 1'		digenic
-OMIM:620040	MONDO:0031057	dyskeratosis congenita, digenic'		digenic
-OMIM:619478	MONDO:0030355	facioscapulohumeral muscular dystrophy 4, digenic'		digenic
-OMIM:300818	MONDO:0010438	paroxysmal nocturnal hemoglobinuria 1		disease caused by a somatic mutation, therefore a gene association stating this is due to a germline mutation should not be added
+OMIM:603956	MONDO:0002974	cervical cancer'	https://orcid.org/0000-0002-4142-7153	evidence of various genes involved
+OMIM:619151	MONDO:0030894	"AMED syndrome, digenic'"	https://orcid.org/0000-0002-4142-7153	digenic
+OMIM:158901	MONDO:0008031		https://orcid.org/0000-0002-4142-7153	digenic
+OMIM:108770	MONDO:0007171	atrial standstill 1'	https://orcid.org/0000-0002-4142-7153	digenic
+OMIM:620040	MONDO:0031057	"dyskeratosis congenita, digenic'"	https://orcid.org/0000-0002-4142-7153	digenic
+OMIM:619478	MONDO:0030355	"facioscapulohumeral muscular dystrophy 4, digenic'"	https://orcid.org/0000-0002-4142-7153	digenic
+OMIM:300818	MONDO:0010438	paroxysmal nocturnal hemoglobinuria 1	https://orcid.org/0000-0002-4142-7153	"disease caused by a somatic mutation, therefore a gene association stating this is due to a germline mutation should not be added"

omim2obo/main.py

@@ -394,11 +394,13 @@ def omim2obo(use_cache: bool = False):
             if not p_mim or p_map_key == '1':
                 continue
 
-            # Add restrictions: Gene->Disease non-causal (disease-defining) relationships
+            # Add restrictions: Gene->Disease non-causal / non-disease-defining relationships
             # - RO:0003302 docs: see MORBIDMAP_PHENOTYPE_MAPPING_KEY_PREDICATES
             # - Mapping key 3 = 'causal' (disease-defining). Handled separately below.
             if p_map_key != '3' or p_mim_excluded:
-                g2d_pred = MORBIDMAP_PHENOTYPE_MAPPING_KEY_PREDICATES[p_map_key] if len(assocs) == 1 else RO['0003302']
+                g2d_pred = MORBIDMAP_PHENOTYPE_MAPPING_KEY_PREDICATES[p_map_key] \
+                    if len(assocs) == 1 and not p_mim_excluded \
+                    else RO['0003302']
                 orcid: Optional[URIRef] = exclusions_p_mim_orcid_map[p_mim] if p_mim_excluded else None
                 add_subclassof_restriction_with_evidence_and_source(
                     graph, g2d_pred, OMIM[p_mim], OMIM[gene_mim], evidence, orcid)

sparql/disease-gene-relationships.sparql

@@ -24,6 +24,7 @@ WHERE {
 
   FILTER(
     ?PredUri IN (
+      <http://purl.obolibrary.org/obo/RO_0003302>,
       <http://purl.obolibrary.org/obo/RO_0003303>,
       <http://purl.obolibrary.org/obo/RO_0003304>,
       <http://purl.obolibrary.org/obo/RO_0004013>,