add mesh example

p2m2 · Sep 20, 2024 · ab8157f · ab8157f
1 parent 6ed909a
commit ab8157f
Show file tree

Hide file tree

Showing 11 changed files with 97,557 additions and 97,633 deletions.
diff --git a/config/1-article.json b/config/1-article.json
@@ -11,6 +11,21 @@
                     "url": "http://purl.obolibrary.org/obo/peco.owl",
                     "prefix": "http://purl.obolibrary.org/obo/PECO_",
                     "format": "xml",
+                    "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
+                    "properties": ["<http://purl.obolibrary.org/obo/IAO_0000115>"]
+                },
+                "po": {
+                    "url": "http://purl.obolibrary.org/obo/po.owl",
+                    "prefix": "http://purl.obolibrary.org/obo/PO_",
+                    "format": "xml",
+                    "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
+                    "properties": ["<http://purl.obolibrary.org/obo/IAO_0000115>"]
+                },
+                "pso": {
+                    "url": "http://purl.obolibrary.org/obo/pso.owl",
+                    "prefix": "http://purl.obolibrary.org/obo/PSO_",
+                    "format": "xml",
+                    "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
                     "properties": ["<http://purl.obolibrary.org/obo/IAO_0000115>"]
                 },
                 "to": {
@@ -28,14 +43,18 @@
                     "format": "xml",
                     "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
                     "properties": ["<http://purl.obolibrary.org/obo/IAO_0000115>"]
-                },
+                }
+            },
+            "mesh_link" : {
                 "mesh" : {
                     "url": "https://nlmpubs.nlm.nih.gov/projects/mesh/rdf/mesh.nt",
                     "prefix": "http://id.nlm.nih.gov/mesh/",
                     "format": "nt",
                     "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
                     "properties": ["<http://id.nlm.nih.gov/mesh/vocab#scopeNote>"]
-                },
+                }
+            },
+            "chemical_link" : {
                 "chmo" : {
                     "url": "http://purl.obolibrary.org/obo/chmo.owl",
                     "prefix": "http://purl.obolibrary.org/obo/CHMO_",

diff --git a/config/human_against_plant.json b/config/human_against_plant.json
diff --git a/config/igepp-20240905.json b/config/igepp-20240905.json
diff --git a/config/igepp-sub.json b/config/igepp-sub.json
diff --git a/config/mesh_example.json b/config/mesh_example.json
@@ -0,0 +1,37 @@
+{
+    "encodeur" : "sentence-transformers/all-MiniLM-L6-v2",
+    "threshold_similarity_tag_chunk" : 0.65,
+    "threshold_similarity_tag" : 0.80,
+    "batch_size" : 32,
+
+    "populate_owl_tag_embeddings" : {
+        "ontologies": {
+            "mesh_link" : {
+                "mesh": {
+                    "filepath" : "data/mesh/2024.ttl",
+                    "prefix": "http://id.nlm.nih.gov/mesh/",
+                    "format": "turtle",
+                    "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
+                    "properties": ["<http://www.w3.org/2000/01/rdf-schema#comment>"]
+                }
+            }
+        },
+        "debug_nb_terms_by_ontology" : -1
+    },
+    "populate_ncbi_taxon_tag_embeddings" : {
+        "regex" : "uman" ,
+        "tags_per_file" : 2000
+    },
+    "populate_abstract_embeddings" : {
+        "abstracts_per_file" : 50,
+
+        "from_ncbi_api" : {
+            "ncbi_api_chunk_size" : 200,
+            "debug_nb_ncbi_request" : -1,
+            "retmax" : 20,
+            "selected_term" : [
+                "Crops%2C+Agricultural%2Fmetabolism%5BMeSH%5D"
+            ]
+        }
+    }
+}
diff --git a/config/prp.json b/config/prp.json
diff --git a/config/test.json b/config/test.json
@@ -1,15 +1,6 @@
 {
     "populate_owl_tag_embeddings" : {
         "ontologies": {
-            "technology_link" : {
-                "ms": {
-                    "url": "http://purl.obolibrary.org/obo/ms.owl",
-                    "prefix": "http://purl.obolibrary.org/obo/MS_",
-                    "format": "xml",
-                    "label" : "<http://www.w3.org/2000/01/rdf-schema#label>",
-                    "properties": ["<http://purl.obolibrary.org/obo/IAO_0000115>"]
-                }
-            },
             "animal_link" : {
                 "zfa": {
                     "url": "http://purl.obolibrary.org/obo/zfa.owl",