Merge pull request #15 from CanDIG/daisieh/biosamples

DIG-2209: Changing /experiments to /biosamples

Author: daisieh

drs_server/drs_database.py

@@ -370,7 +370,7 @@ def delete_program(program_id, tries=1):
     return None
 
 
-def list_experiments(program_ids=None, submitter_sample_ids=None):
+def list_biosamples(program_ids=None, submitter_sample_ids=None):
     experiment_obj = aliased(DrsObject)
     analysis_obj = aliased(DrsObject)
     stmt = select(experiment_obj.name, experiment_obj.id, experiment_obj.program_id, experiment_obj.description, analysis_obj.name, analysis_obj.description, ContentsObject)
@@ -388,24 +388,22 @@ def list_experiments(program_ids=None, submitter_sample_ids=None):
             result = row._mapping
             if result["name"] not in results_dict:
                 results_dict[result["name"]] = {
-                    "experiment_id": result["name"],
+                    "biosample_id": result["name"],
                     "program": result["program_id"],
-                    "genomes": [],
-                    "transcriptomes": [],
-                    "variants": [],
-                    "reads": [],
-                    "expressions": []
+                    "experiments": {
+                        "wgs": [],
+                        "wts": []
+                    },
+                    "runs": [],
+                    "analyses": {}
                 }
-            this_result = results_dict[result["name"]]
-            if result["description"] == "wgs" and result["id"] not in this_result["genomes"]:
-                results_dict[result["name"]]["genomes"].append(result["id"])
-            elif result["description"] == "wts" and result["id"] not in this_result["transcriptomes"]:
-                results_dict[result["name"]]["transcriptomes"].append(result["id"])
-            if result["description_1"] == "sequence_variation" and result["name_1"] not in this_result["variants"]:
-                results_dict[result["name"]]["variants"].append(result["name_1"])
-            elif result["description_1"] == "reference_alignment" and result["name_1"] not in this_result["reads"]:
-                results_dict[result["name"]]["reads"].append(result["name_1"])
-            elif result["description_1"] == "sequence_annotation" and result["name_1"] not in this_result["expressions"]:
-                results_dict[result["name"]]["expressions"].append(result["name_1"])
+            if result["description"] in ["wgs", "wts"] and result["id"] not in results_dict[result["name"]]["experiments"][result["description"]]:
+                results_dict[result["name"]]["experiments"][result["description"]].append(result["id"])
+            if result["description_1"] == "raw_reads":
+                results_dict[result["name"]]["runs"].append(result["name_1"])
+            else:
+                if result["description_1"] not in results_dict[result["name"]]["analyses"]:
+                    results_dict[result["name"]]["analyses"][result["description_1"]] = []
+                results_dict[result["name"]]["analyses"][result["description_1"]].append(result["name_1"])
         return list(results_dict.values())
     return None

drs_server/drs_openapi.yaml

@@ -139,14 +139,14 @@ paths:
                     description: array of strings that need to be passed with the URL to get bytes
       tags:
         - Objects
-  /experiments:
+  /biosamples:
     parameters:
      - $ref: '#/components/parameters/BearerAuth'
     post:
       summary: List all DRS Experiment objects matching an array of submitter_sample_ids. Only authorized for site admins and authzed services.
-      operationId: drs_operations.list_experiments
+      operationId: drs_operations.list_biosamples
       requestBody:
-        $ref: '#/components/requestBodies/ExperimentsRequest'
+        $ref: '#/components/requestBodies/BiosamplesRequest'
       responses:
         200:
           description: The `DrsObjects` were found successfully
@@ -156,7 +156,7 @@ paths:
                 type: array
                 items:
                   anyOf:
-                    - $ref: '#/components/schemas/ExperimentTicket'
+                    - $ref: '#/components/schemas/BiosampleTicket'
       tags:
         - Objects
   /programs:
@@ -314,7 +314,7 @@ components:
         'application/json':
           schema:
             $ref: '#/components/schemas/Program'
-    ExperimentsRequest:
+    BiosamplesRequest:
       description: list of submitter sample ids requested
       required: false
       content:
@@ -372,35 +372,48 @@ components:
     DrsUri:
       type: string
       description: a DRS object's self_uri
-    ExperimentTicket:
+    BiosampleTicket:
       type: object
       properties:
-          experiment_id:
-              type: string
-              description: MoH submitter_sample_id from a Sample Registration associated with a specimen.
-          program:
-              type: string
-              description: MoH program ID
-          genomes:
+        biosample_id:
+          type: string
+          description: MoH submitter_sample_id from a Sample Registration associated with a specimen.
+        program:
+          type: string
+          description: MoH program ID
+        experiments:
+          type: object
+          description: Arrays of associated experiment IDs derived from the biosample, keyed by type.
+          properties:
+            wgs:
               type: array
-              description: An array of associated AnalysisDrsObjects that describe reads, e.g. fastqs.
               items:
-                  type: string
-          transcriptomes:
+                type: string
+            wts:
               type: array
-              description: An array of associated AnalysisDrsObjects that describe transcriptome data.
               items:
-                  type: string
-          variants:
+                type: string
+        runs:
+          type: array
+          description: An array of associated run IDs that describe raw reads derived from biosample.
+          items:
+            type: string
+        analyses:
+          type: object
+          description: Arrays of associated analysis IDs based on the biosample, keyed by analysis type.
+          properties:
+            reference_alignment:
               type: array
-              description: An array of associated DataDrsObjects that describe variant files.
               items:
-                  type: string
-          reads:
+                type: string
+            sequence_variation:
               type: array
-              description: An array of associated DataDrsObjects that describe read files, e.g. bams.
               items:
-                  type: string
+                type: string
+            sequence_annotation:
+              type: array
+              items:
+                type: string
     ExperimentDrsObject:
       type: object
       description: A DrsObject that describes the sequencing experiment.
@@ -413,10 +426,10 @@ components:
       properties:
         name:
           type: string
-          description: The specimen identifier for the experiment, as defined in sample_registration in the MOHCCN data model.
+          description: The biosample identifier for the experiment.
         id:
           type: string
-          description: The alias for the experiment, as defined by the sequencing center
+          description: A unique ID for the experiment. Often the alias for the experiment, as defined by the sequencing center
         contents:
           type: array
           description: The specific analysis contents objects that were generated from this experiment.
@@ -479,7 +492,7 @@ components:
             $ref: '#/components/schemas/Checksum'
     AnalysisDrsObject:
       type: object
-      description: A DrsObject that describes a sequencing analysis. It usually will consist of a data file, e.g. a variant or read file, and its associated index file. Its contents should also include any associated Experiments (as ExperimentContentObjects), ordered in order of appearance in the associated variant/read files.
+      description: A DrsObject that describes a sequencing analysis. It usually will consist of a data file, e.g. a variant or read file, and its associated index file. Its contents should also include any associated Experiments (as ContentObjects), ordered in order of appearance in the associated variant/read files.
       required:
         - name
         - id

drs_server/drs_operations.py

@@ -84,7 +84,7 @@ def list_objects(program_id=None, submitter_sample_id=None):
     return drs_database.list_drs_objects(program_id=program_id, submitter_sample_id=submitter_sample_id), 200
 
 
-async def list_experiments():
+async def list_biosamples():
     if not authz.has_full_authz(connexion.request):
         return {"message": f"Not authorized to list all objects"}, 403
     req = await connexion.request.json()
@@ -95,8 +95,8 @@ async def list_experiments():
         submitter_sample_ids = req["submitter_sample_ids"]
     if "program_ids" in req:
         program_ids = req["program_ids"]
-    experiments = drs_database.list_experiments(program_ids=program_ids, submitter_sample_ids=submitter_sample_ids)
-    return experiments, 200
+    biosamples = drs_database.list_biosamples(program_ids=program_ids, submitter_sample_ids=submitter_sample_ids)
+    return biosamples, 200
 
 
 @app.route('/ga4gh/drs/v1/objects/<object_id>/access_url/<path:access_id>')