Merge pull request #392 from monarch-initiative/issue391-taxon-mapper

Taxon Label to ID Mapper Endpoint

biolink/api/ontol/endpoints/identifier.py

@@ -0,0 +1,36 @@
+import logging
+
+from flask import request
+from flask_restplus import Resource
+from biolink.datamodel.serializers import association
+from biolink.api.restplus import api
+from ontobio.sparql.sparql_ontol_utils import batch_fetch_ids
+import pysolr
+
+log = logging.getLogger(__name__)
+
+parser = api.parser()
+parser.add_argument('label', action='append', help='List of labels', required=True)
+
+class OntolIdentifierResource(Resource):
+
+    @api.expect(parser)
+    def get(self):
+        """
+        Fetches a map from CURIEs/IDs to labels
+        """
+        args = parser.parse_args()
+
+        return batch_fetch_ids(args.label)
+
+    @api.expect(parser)
+    def post(self):
+        """
+        Fetches a map from CURIEs/IDs to labels.
+        
+        Takes 'label' list argument either as a querystring argument or as a key
+        in the POST body when content-type is application/json.
+        """
+        args = parser.parse_args()
+
+        return batch_fetch_ids(args.label)

biolink/api/ontol/endpoints/labeler.py

@@ -10,10 +10,9 @@
 log = logging.getLogger(__name__)
 
 parser = api.parser()
-parser.add_argument('id', action='append', help='List of ids')
+parser.add_argument('id', action='append', help='List of ids', required=True)
 
 class OntolLabelerResource(Resource):
-
     @api.expect(parser)
     def get(self):
         """
@@ -22,8 +21,3 @@ def get(self):
         args = parser.parse_args()
 
         return batch_fetch_labels(args.id)
-
-
-
-
-

conf/routes.yaml

@@ -309,6 +309,11 @@ route_mapping:
       routes:
         - route: /
           resource: biolink.api.ontol.endpoints.labeler.OntolLabelerResource
+    - name: ontol/identifier
+      description: Retrieve IDs for labels
+      routes:
+        - route: /
+          resource: biolink.api.ontol.endpoints.identifier.OntolIdentifierResource
 #    - name: ontol/enrichment
 #      description: To be implemented
 #      routes:

requirements.txt

@@ -1,6 +1,6 @@
 prefixcommons>=0.0
 # ontobio>=2.7.14
-git+https://github.com/monarch-initiative/ontobio.git@baf3b5a7344fe6dda9c3980c8d472655431cb5df#egg=ontobio
+git+https://github.com/monarch-initiative/ontobio.git@91222c8b442196d6eeeafeb6073946494e8a3a10#egg=ontobio
 pip>=9.0.1
 wheel>0.25.0
 markupsafe==2.0.1

tests/unit/test_ontol_identifier.py

@@ -0,0 +1,89 @@
+from pprint import pprint
+
+import pytest
+from biolink.app import app
+from biolink.api.sim.endpoints.owlsim import get_owlsim_api
+
+
+class TestOntolIdentifier():
+    """
+    Integration tests for ontol identifier endpoint, using labeler as a comparison.
+
+    Specifically, does a sanity check against identifier, then determines if we
+    can (somewhat) losslessly map a set of labels to IDs and back
+    """
+
+    @classmethod
+    def setup_class(self):
+        app.testing = True
+        self.test_client = app.test_client()
+
+    @classmethod
+    def teardown_class(self):
+        self.test_client = None
+
+    def test_sample_identifier(self):
+        response = self.test_client.get('/api/ontol/identifier?label=Mus%20musculus')
+        assert response.status_code == 200
+
+        # todo: assert we have results, and that the response matches a known entry
+        assert len(response.json.items()) > 0
+        assert (
+            response.json == {
+                "Mus musculus": [
+                    "NCBITaxon:10090",
+                    "OBO:FBsp_00000276",
+                    "http://ncicb.nci.nih.gov/xml/owl/EVS/Thesaurus.owl#C45247",
+                    "VTO:0014661"
+                ]
+            }
+        )
+
+    @pytest.mark.xfail(reason="/ontol/labeler returns just one result, so it can't be onto with /ontol/identifier")
+    def test_label_id_nonlossy(self):
+        labels = [
+            "Sus scrofa",
+            "Drosophila melanogaster",
+            "Homo sapiens",
+            "Mus musculus",
+            "Bos taurus",
+            "Saccharomyces cerevisiae S288C",
+            "Xenopus tropicalis",
+            "Danio rerio",
+            "Gallus gallus",
+            "Anolis carolinensis",
+            "Canis lupus familiaris",
+            "Felis catus",
+            "Macaca mulatta",
+            "Monodelphis domestica",
+            "Ornithorhynchus anatinus",
+            "Pan troglodytes",
+            "Rattus norvegicus",
+            "Takifugu rubripes",
+            "Equus caballus"
+        ]
+
+        response = self.test_client.post('/api/ontol/identifier', data={'label': labels})
+        pprint(response.json)
+
+        assert response.status_code == 200
+
+        # todo: assert we have results
+        assert len(response.json.items()) > 0
+
+        # extract the first ID and its label from each result
+        id_set = {ids[0]: label for label, ids in response.json.items()}
+
+        # produce a query of IDs to labels using the /ontol/labels/
+        response = self.test_client.get(
+            '/api/ontol/labeler/?%s' % "&".join("id=%s" % x for x in id_set.keys())
+        )
+        pprint(response.json)
+
+        # gather the results and compare to the initial set of labels
+        orignal_set = set(labels)
+        response_set = set(response.json.values()) 
+
+        assert orignal_set == response_set, (
+            "Response labels and originals don't match! difference: %s" % orignal_set.difference(response_set)
+        )