Search in descriptions & link to GARD and Orphanet (#679)

Mostly I wanted the search in description, but then I found myself chasing down doing _something_ so that we'll have GARD & Orphanet links working in the UI
monarch-initiative · Apr 22, 2024 · ac4e7b4 · ac4e7b4
1 parent fbdda4c
commit ac4e7b4
Show file tree

Hide file tree

Showing 16 changed files with 238 additions and 222 deletions.
diff --git a/backend/src/monarch_py/implementations/solr/solr_query_utils.py b/backend/src/monarch_py/implementations/solr/solr_query_utils.py
@@ -283,14 +283,20 @@ def blank_search_boost() -> str:
     return f"product({boost})"
 
 
-def entity_query_fields():
+def autocomplete_query_fields():
     """
-    Shared query field list between search and autocomplete,
-    since the field list and boosts are currently the same
+    Fields and boosts used for autocomplete
     """
     return "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac"
 
 
+def entity_query_fields():
+    """
+    Fields and boosts used for entity search, includes autocomplete fields and expands upon them beyond pure name fields
+    """
+    return f"{autocomplete_query_fields()} description_t"
+
+
 def association_search_query_fields():
     """
     Shared field list for free text search on associations (e.g. for the association table)

diff --git a/backend/src/monarch_py/service/curie_service.py b/backend/src/monarch_py/service/curie_service.py
@@ -9,3 +9,5 @@
 # this is a magic keyword that represents the "merged" context from Chris M's algorithm
 # (https://github.com/linkml/prefixmaps/blob/main/src/prefixmaps/data/merged.csv)
 converter = load_converter("merged")
+converter.add_prefix("GARD", "https://rarediseases.info.nih.gov/diseases/")
+converter.add_prefix("Orphanet", "https://www.orpha.net/en/disease/detail/", merge=True)
diff --git a/backend/src/monarch_py/utils/utils.py b/backend/src/monarch_py/utils/utils.py
@@ -68,7 +68,15 @@ def set_log_level(log_level: str):
 
 def get_links_for_field(field: List[str]) -> List[ExpandedCurie]:
     # TODO should be able to remove curie.replace("PMID", "PUBMED")) since the converter should handle prefix synonyms
-    return [ExpandedCurie(id=curie, url=converter.expand(curie.replace("PMID", "PUBMED"))) for curie in field]
+    expanded_curies = [ExpandedCurie(id=curie, url=get_link_for_curie(curie)) for curie in field]
+    return expanded_curies
+
+
+def get_link_for_curie(curie: str) -> ExpandedCurie:
+    url = converter.expand(curie.replace("PMID", "PUBMED"))
+    if curie.startswith("GARD:"):
+        url += "/index"
+    return url
 
 
 def get_provided_by_link(provided_by: str) -> ExpandedCurie:

diff --git a/backend/tests/fixtures/association_counts_response.py b/backend/tests/fixtures/association_counts_response.py
@@ -5,7 +5,7 @@
 def association_counts_response():
     return {
         "responseHeader": {
-            "QTime": 0,
+            "QTime": 1,
             "params": {
                 "facet.query": [
                     '(category:"biolink:DiseaseToPhenotypicFeatureAssociation") AND (subject:"MONDO:0020121" OR subject_closure:"MONDO:0020121")',

diff --git a/backend/tests/fixtures/autocomplete_query.py b/backend/tests/fixtures/autocomplete_query.py
@@ -12,7 +12,7 @@ def autocomplete_query():
         "facet_fields": [],
         "facet_queries": [],
         "filter_queries": [],
-        "query_fields": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac",
+        "query_fields": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac description_t",
         "def_type": "edismax",
         "q_op": "AND",
         "mm": "100%",

diff --git a/backend/tests/fixtures/autocomplete_response.py b/backend/tests/fixtures/autocomplete_response.py
@@ -5,13 +5,13 @@
 def autocomplete_response():
     return {
         "responseHeader": {
-            "QTime": 0,
+            "QTime": 1,
             "params": {
                 "mm": "100%",
                 "q": "fanc",
                 "defType": "edismax",
                 "facet_min_count": "1",
-                "qf": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac",
+                "qf": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac description_t",
                 "start": "0",
                 "q.op": "AND",
                 "boost": 'product(if(termfreq(category,"biolink:PhenotypicFeature"),1.1,1),if(termfreq(category,"biolink:Disease"),1.3,1),if(and(termfreq(in_taxon,"NCBITaxon:9606"),termfreq(category,"biolink:Gene")),1.1,1),if(termfreq(deprecated,"true"),0.1,1))',

diff --git a/backend/tests/fixtures/node.py b/backend/tests/fixtures/node.py
@@ -42,20 +42,20 @@ def node():
             {"id": "DOID:9884", "url": "http://purl.obolibrary.org/obo/DOID_9884"},
             {"id": "ICD10CM:G71.0", "url": "https://icd.codes/icd10cm/G71.0"},
             {"id": "NCIT:C84910", "url": "http://purl.obolibrary.org/obo/NCIT_C84910"},
-            {"id": "Orphanet:98473", "url": None},
+            {"id": "Orphanet:98473", "url": "https://www.orpha.net/en/disease/detail/98473"},
             {"id": "SCTID:73297009", "url": "http://identifiers.org/snomedct/73297009"},
             {"id": "UMLS:C0026850", "url": "http://identifiers.org/umls/C0026850"},
             {"id": "MESH:D009136", "url": "http://identifiers.org/mesh/D009136"},
         ],
         "external_links": [
             {"id": "DOID:9884", "url": "http://purl.obolibrary.org/obo/DOID_9884"},
-            {"id": "GARD:7922", "url": None},
+            {"id": "GARD:7922", "url": "https://rarediseases.info.nih.gov/diseases/7922/index"},
             {"id": "ICD10CM:G71.0", "url": "https://icd.codes/icd10cm/G71.0"},
             {"id": "ICD9:359.1", "url": None},
             {"id": "MESH:D009136", "url": "http://identifiers.org/mesh/D009136"},
             {"id": "MedDRA:10028356", "url": None},
             {"id": "NCIT:C84910", "url": "http://purl.obolibrary.org/obo/NCIT_C84910"},
-            {"id": "Orphanet:98473", "url": None},
+            {"id": "Orphanet:98473", "url": "https://www.orpha.net/en/disease/detail/98473"},
             {"id": "SCTID:73297009", "url": "http://identifiers.org/snomedct/73297009"},
             {"id": "UMLS:C0026850", "url": "http://identifiers.org/umls/C0026850"},
         ],

diff --git a/backend/tests/fixtures/object_formatted.py b/backend/tests/fixtures/object_formatted.py
diff --git a/backend/tests/fixtures/search.py b/backend/tests/fixtures/search.py
@@ -6,7 +6,7 @@ def search():
     return {
         "limit": 20,
         "offset": 0,
-        "total": 101,
+        "total": 117,
         "items": [
             {
                 "id": "MONDO:0019391",

diff --git a/backend/tests/fixtures/search_query.py b/backend/tests/fixtures/search_query.py
@@ -12,7 +12,7 @@ def search_query():
         "facet_fields": [],
         "facet_queries": [],
         "filter_queries": ["name:*"],
-        "query_fields": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac",
+        "query_fields": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac description_t",
         "def_type": "edismax",
         "q_op": "AND",
         "mm": "100%",

diff --git a/backend/tests/fixtures/search_response.py b/backend/tests/fixtures/search_response.py
@@ -5,13 +5,13 @@
 def search_response():
     return {
         "responseHeader": {
-            "QTime": 0,
+            "QTime": 3,
             "params": {
                 "mm": "100%",
                 "q": "fanconi",
                 "defType": "edismax",
                 "facet_min_count": "1",
-                "qf": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac",
+                "qf": "id^100 name^10 name_t^5 name_ac symbol^10 symbol_t^5 symbol_ac synonym synonym_t synonym_ac description_t",
                 "start": "0",
                 "q.op": "AND",
                 "boost": 'product(if(termfreq(category,"biolink:PhenotypicFeature"),1.1,1),if(termfreq(category,"biolink:Disease"),1.3,1),if(and(termfreq(in_taxon,"NCBITaxon:9606"),termfreq(category,"biolink:Gene")),1.1,1),if(termfreq(deprecated,"true"),0.1,1))',
@@ -21,7 +21,7 @@ def search_response():
             },
         },
         "response": {
-            "num_found": 101,
+            "num_found": 117,
             "start": 0,
             "docs": [
                 {

diff --git a/frontend/fixtures/node.json b/frontend/fixtures/node.json
@@ -48,7 +48,7 @@
     },
     {
       "id": "Orphanet:98473",
-      "url": null
+      "url": "https://www.orpha.net/en/disease/detail/98473"
     },
     {
       "id": "SCTID:73297009",
@@ -70,7 +70,7 @@
     },
     {
       "id": "GARD:7922",
-      "url": null
+      "url": "https://rarediseases.info.nih.gov/diseases/7922/index"
     },
     {
       "id": "ICD10CM:G71.0",
@@ -94,7 +94,7 @@
     },
     {
       "id": "Orphanet:98473",
-      "url": null
+      "url": "https://www.orpha.net/en/disease/detail/98473"
     },
     {
       "id": "SCTID:73297009",