Merge pull request #571 from hubmapconsortium/yuanzhou/yaml-to-ubkg

Yuanzhou/yaml to ubkg

src/app.py

@@ -2794,17 +2794,12 @@ def get_prov_info():
         for item in dataset['data_types']:
             try:
                 assay_description_list.append(assay_types_dict[item]['description'])
-            # Some data types aren't given by their code in the assay types yaml and are instead given as an alt name.
-            # In these cases, we have to search each assay type and see if the given code matches any alternate names.
             except KeyError:
-                valid_key = False
-                for each in assay_types_dict:
-                    if valid_key is False:
-                        if item in assay_types_dict[each]['alt-names']:
-                            assay_description_list.append(assay_types_dict[each]['description'])
-                            valid_key = True
-                if valid_key is False:
-                    assay_description_list.append(item)
+                logger.exception(f"Data type {item} not found in resulting assay types via ontology-api")
+
+                # Just use the data type value
+                assay_description_list.append(item)
+
         dataset['data_types'] = assay_description_list
         internal_dict[HEADER_DATASET_DATA_TYPES] = dataset['data_types']
 
@@ -3116,17 +3111,12 @@ def get_prov_info_for_dataset(id):
     for item in dataset['data_types']:
         try:
             assay_description_list.append(assay_types_dict[item]['description'])
-        # Some data types aren't given by their code in the assay types yaml and are instead given as an alt name.
-        # In these cases, we have to search each assay type and see if the given code matches any alternate names.
         except KeyError:
-            valid_key = False
-            for each in assay_types_dict:
-                if valid_key is False:
-                    if item in assay_types_dict[each]['alt-names']:
-                        assay_description_list.append(assay_types_dict[each]['description'])
-                        valid_key = True
-            if valid_key is False:
-                assay_description_list.append(item)
+            logger.exception(f"Data type {item} not found in resulting assay types via ontology-api")
+
+            # Just use the data type value
+            assay_description_list.append(item)
+
     dataset['data_types'] = assay_description_list
     internal_dict[HEADER_DATASET_DATA_TYPES] = dataset['data_types']
     if return_json is False:
@@ -3342,17 +3332,12 @@ def sankey_data():
             assay_description = ""
             try:
                 assay_description = assay_types_dict[dataset[HEADER_DATASET_DATA_TYPES]]['description']
-            # Some data types aren't given by their code in the assay types yaml and are instead given as an alt name.
-            # In these cases, we have to search each assay type and see if the given code matches any alternate names.
             except KeyError:
-                valid_key = False
-                for each in assay_types_dict:
-                    if valid_key is False:
-                        if dataset[HEADER_DATASET_DATA_TYPES] in assay_types_dict[each]['alt-names']:
-                            assay_description = assay_types_dict[each]['description']
-                            valid_key = True
-                if valid_key is False:
-                    assay_description = dataset[HEADER_DATASET_DATA_TYPES]
+                logger.exception(f"Data type {dataset[HEADER_DATASET_DATA_TYPES]} not found in resulting assay types via ontology-api")
+
+                # Just use the data type value
+                assay_description = dataset[HEADER_DATASET_DATA_TYPES]
+
             internal_dict[HEADER_DATASET_DATA_TYPES] = assay_description
 
             # Replace applicable Group Name and Data type with the value needed for the sankey via the mapping_dict

src/schema/schema_constants.py

@@ -14,8 +14,8 @@ class SchemaConstants(object):
     UUID_API_ID_ENDPOINT = '/uuid'
     INGEST_API_FILE_COMMIT_ENDPOINT = '/file-commit'
     INGEST_API_FILE_REMOVE_ENDPOINT = '/file-remove'
-    ONTOLOGY_API_ASSAY_TYPES_ENDPOINT = '/assaytype?application_context=HuBMAP'
-    ONTOLOGY_API_ORGAN_TYPES_ENDPOINT = '/organs?application_context=HuBMAP'
+    ONTOLOGY_API_ASSAY_TYPES_ENDPOINT = '/assaytype?application_context=HUBMAP'
+    ONTOLOGY_API_ORGAN_TYPES_ENDPOINT = '/organs/by-code?application_context=HUBMAP'
 
     DOI_BASE_URL = 'https://doi.org/'
 

src/schema/schema_manager.py

@@ -1778,12 +1778,22 @@ def delete_memcached_cache(uuids_list):
 Returns
 -------
 dict
-    The available organ types
+    The available organ types in the following format:
+
+    {
+        "AO": "Aorta",
+        "BD": "Blood",
+        "BL": "Bladder",
+        "BM": "Bone Marrow",
+        "BR": "Brain",
+        "HT": "Heart",
+        ...
+    }
 """
 def get_organ_types():
     global _ontology_api_url
 
-    target_url = _ontology_api_url + '/organs?application_context=HuBMAP'
+    target_url = _ontology_api_url + SchemaConstants.ONTOLOGY_API_ORGAN_TYPES_ENDPOINT
 
     # Use Memcached to improve performance
     response = make_request_get(target_url, internal_token_used = True)
@@ -1792,14 +1802,10 @@ def get_organ_types():
     response.raise_for_status()
 
     if response.status_code == 200:
-        ids_dict = response.json()
-        return ids_dict
+        return response.json()
     else:
-        # uuid-api will also return 400 if the given id is invalid
-        # We'll just hanle that and all other cases all together here
-        msg = f"Unable to make a request to query the id via uuid-api: {id}"
         # Log the full stack trace, prepend a line with our message
-        logger.exception(msg)
+        logger.exception("Unable to make a request to query the organ types via ontology-api")
 
         logger.debug("======get_organ_types() status code from ontology-api======")
         logger.debug(response.status_code)
@@ -1817,12 +1823,32 @@ def get_organ_types():
 Returns
 -------
 dict
-    The available assay types
+    The available assay types by name in the following format:
+
+    {
+        "10x-multiome": {
+            "contains_pii": true,
+            "description": "10x Multiome",
+            "name": "10x-multiome",
+            "primary": true,
+            "vis_only": false,
+            "vitessce_hints": []
+        },
+        "AF": {
+            "contains_pii": false,
+            "description": "Autofluorescence Microscopy",
+            "name": "AF",
+            "primary": true,
+            "vis_only": false,
+            "vitessce_hints": []
+        },
+        ...
+    }
 """
 def get_assay_types():
     global _ontology_api_url
 
-    target_url = _ontology_api_url + '/assaytype?application_context=HuBMAP'
+    target_url = _ontology_api_url + SchemaConstants.ONTOLOGY_API_ASSAY_TYPES_ENDPOINT
 
     # Use Memcached to improve performance
     response = make_request_get(target_url, internal_token_used = True)
@@ -1831,14 +1857,18 @@ def get_assay_types():
     response.raise_for_status()
 
     if response.status_code == 200:
-        ids_dict = response.json()
-        return ids_dict
+        assay_types_by_name = {}
+        result_dict = response.json()
+
+        # Due to the json envelop being used int the json result
+        assay_types_list = result_dict['result']
+        for assay_type_dict in assay_types_list:
+            assay_types_by_name[assay_type_dict['name']] = assay_type_dict
+
+        return assay_types_by_name
     else:
-        # uuid-api will also return 400 if the given id is invalid
-        # We'll just hanle that and all other cases all together here
-        msg = f"Unable to make a request to query the id via uuid-api: {id}"
         # Log the full stack trace, prepend a line with our message
-        logger.exception(msg)
+        logger.exception("Unable to make a request to query the assay types via ontology-api")
 
         logger.debug("======get_assay_types() status code from ontology-api======")
         logger.debug(response.status_code)
@@ -1876,4 +1906,4 @@ def _create_request_headers(user_token):
         auth_header_name: auth_scheme + ' ' + user_token
     }
 
-    return headers_dict
+    return headers_dict

src/schema/schema_triggers.py

@@ -1024,9 +1024,10 @@ def get_dataset_title(property_key, normalized_type, user_token, existing_data_d
     # Parse the organ description
     if organ_name is not None:
         try: 
-            # The organ_name is the two-letter code only set if specimen_type == 'organ'
+            # The organ_name is the two-letter code only set for 'organ'
             # Convert the two-letter code to a description
-            organ_desc = _get_organ_description(organ_name)
+            organ_types_dict = schema_manager.get_organ_types()
+            organ_desc = organ_types_dict[organ_name].lower()
         except (yaml.YAMLError, requests.exceptions.RequestException) as e:
             raise Exception(e)
 
@@ -2098,21 +2099,3 @@ def _get_combined_assay_type_description(data_types):
 
     return assay_type_desc
 
-
-"""
-Get the organ description based on the given organ code
-
-Parameters
-----------
-organ_code : str
-    The two-letter organ code
-
-Returns
--------
-str: The organ code description
-"""
-def _get_organ_description(organ_code):
-    organ_types_dict = schema_manager.get_organ_types()
-    return organ_types_dict[organ_code]['description'].lower()
-
-