adding OHD-Carolina

Common/biolink_constants.py

@@ -55,6 +55,9 @@
 KNOWLEDGE_LEVEL = 'knowledge_level'
 MAX_RESEARCH_PHASE = 'max_research_phase'
 HAS_SUPPORTING_STUDY_RESULT = 'has_supporting_study_result'
+LOG_ODDS_RATIO = 'log_odds_ratio'
+LOG_ODDS_RATIO_95_CI = 'log_odds_ratio_95_ci'
+TOTAL_SAMPLE_SIZE = 'total_sample_size'
 
 # enums for knowledge level
 KNOWLEDGE_ASSERTION = 'knowledge_assertion'
@@ -153,6 +156,9 @@
     MECHANISM_OF_ACTION,
     MAX_RESEARCH_PHASE,
     HAS_SUPPORTING_STUDY_RESULT,
+    LOG_ODDS_RATIO,
+    LOG_ODDS_RATIO_95_CI,
+    TOTAL_SAMPLE_SIZE,
     # qualifiers
     ANATOMICAL_CONTEXT_QUALIFIER,
     CAUSAL_MECHANISM_QUALIFIER,

Common/data_sources.py

@@ -30,6 +30,7 @@
 MOLEPRO = 'MolePro'
 MONARCH_KG = 'MonarchKG'
 MONDO_PROPS = 'MONDOProps'
+OHD_CAROLINA = 'OHD-Carolina'
 ONTOLOGICAL_HIERARCHY = 'OntologicalHierarchy'
 PANTHER = 'PANTHER'
 PHAROS = 'PHAROS'
@@ -82,6 +83,7 @@
     MOLEPRO: ("parsers.molepro.src.loadMolePro", "MoleProLoader"),
     MONARCH_KG: ("parsers.monarchkg.src.loadMonarchKG", "MonarchKGLoader"),
     MONDO_PROPS: ("parsers.MONDOProperties.src.loadMP", "MPLoader"),
+    OHD_CAROLINA: ("parsers.ohd_carolina.src.loadOHD", "OHDLoader"),
     ONTOLOGICAL_HIERARCHY: ("parsers.UberGraph.src.loadUG", "OHLoader"),
     PANTHER: ("parsers.panther.src.loadPanther", "PLoader"),
     PHAROS: ("parsers.PHAROS.src.loadPHAROS", "PHAROSLoader"),

parsers/cohd/src/loadCOHD.py

@@ -4,7 +4,6 @@
 import yaml
 
 from Common.loader_interface import SourceDataLoader
-from Common.biolink_constants import PRIMARY_KNOWLEDGE_SOURCE
 from Common.utils import GetData, quick_jsonl_file_iterator
 
 

parsers/ohd_carolina/src/loadOHD.py

@@ -0,0 +1,119 @@
+
+import os
+import requests
+import yaml
+import enum
+import orjson
+
+from io import TextIOWrapper
+from zipfile import ZipFile
+from Common.extractor import Extractor
+from Common.loader_interface import SourceDataLoader
+from Common.biolink_constants import *
+from Common.utils import GetData
+
+
+class EDGESDATACOLS(enum.IntEnum):
+    SUBJECT_ID = 0
+    SUBJECT_NAME = 1
+    OBJECT_ID = 2
+    OBJECT_NAME = 3
+    PREDICATE = 4
+    CHI_SQUARED_P_VALUE = 5
+    LOG_ODDS_RATIO = 6
+    LOG_ODDS_RATIO_95_CI = 7
+    SCORE = 8
+    TOTAL_SAMPLE_SIZE = 9
+    PRIMARY_KS = 10
+
+
+##############
+# Class: OHD source loader
+#
+# Desc: Class that loads/parses the Open Health Data @ Carolina data.
+##############
+class OHDLoader(SourceDataLoader):
+
+    source_id: str = 'OHD-Carolina'
+    provenance_id: str = 'infores:openhealthdata-carolina'
+    parsing_version: str = '1.0'
+
+    def __init__(self, test_mode: bool = False, source_data_dir: str = None):
+        """
+        :param test_mode - sets the run into test mode
+        :param source_data_dir - the specific storage directory to save files in
+        """
+        super().__init__(test_mode=test_mode, source_data_dir=source_data_dir)
+
+        self.data_url = 'https://stars.renci.org/var/data_services/ohd/'
+        self.version_file = 'ohd.yaml'
+        self.ohd_archive_file = 'unc_omop_2018_2022_kg.zip'
+        self.ohd_edges_file = 'unc_omop_2018_2022_kg.csv'
+        self.data_files = [self.ohd_archive_file]
+
+    def get_latest_source_version(self) -> str:
+        version_file_url = f"{self.data_url}{self.version_file}"
+        r = requests.get(version_file_url)
+        if not r.ok:
+            r.raise_for_status()
+        version_yaml = yaml.full_load(r.text)
+        build_version = str(version_yaml['build'])
+        return build_version
+
+    def get_data(self) -> bool:
+        for data_file in self.data_files:
+            source_data_url = f'{self.data_url}{data_file}'
+            data_puller = GetData()
+            data_puller.pull_via_http(source_data_url, self.data_path)
+        return True
+
+    def parse_data(self) -> dict:
+        """
+        Parses the data file for graph nodes/edges
+
+        :return: ret_val: load_metadata
+        """
+        extractor = Extractor(file_writer=self.output_file_writer)
+
+        ohd_archive_file_path: str = os.path.join(self.data_path, self.ohd_archive_file)
+        with ZipFile(ohd_archive_file_path) as ohd_archive:
+            with ohd_archive.open(self.ohd_edges_file, "r") as fp:
+                extractor.csv_extract(TextIOWrapper(fp),
+                                      lambda line: line[EDGESDATACOLS.SUBJECT_ID.value],  # subject id
+                                      lambda line: line[EDGESDATACOLS.OBJECT_ID.value],  # object id
+                                      lambda line: line[EDGESDATACOLS.PREDICATE],  # predicate extractor
+                                      lambda line: {NAME: line[EDGESDATACOLS.SUBJECT_NAME.value]},  # subject props
+                                      lambda line: {NAME: line[EDGESDATACOLS.OBJECT_NAME.value]},  # object props
+                                      lambda line: self.get_edge_properties(line),  # edgeprops
+                                      comment_character=None,
+                                      delim=',',
+                                      has_header_row=True)
+
+        return extractor.load_metadata
+
+    @staticmethod
+    def get_edge_properties(line):
+        return {
+            AGENT_TYPE: DATA_PIPELINE,
+            KNOWLEDGE_LEVEL: STATISTICAL_ASSOCIATION,
+            'score': line[EDGESDATACOLS.SCORE.value],
+            PRIMARY_KNOWLEDGE_SOURCE: line[EDGESDATACOLS.PRIMARY_KS.value],
+            P_VALUE: float(line[EDGESDATACOLS.CHI_SQUARED_P_VALUE.value]),
+            LOG_ODDS_RATIO: float(line[EDGESDATACOLS.LOG_ODDS_RATIO.value]),
+            LOG_ODDS_RATIO_95_CI: orjson.loads(line[EDGESDATACOLS.LOG_ODDS_RATIO_95_CI.value]),
+            TOTAL_SAMPLE_SIZE: int(line[EDGESDATACOLS.TOTAL_SAMPLE_SIZE.value])
+        }
+
+    """
+    # this should probably be something like this instead to match COHD, 
+    because merged edges wont be able to handle conflicting attributes across multiple supporting studies
+    'attributes': [orjson.dumps({
+                HAS_SUPPORTING_STUDY_RESULT: [{
+                    P_VALUE: float(line[EDGESDATACOLS.CHI_SQUARED_P_VALUE.value]),
+                    LOG_ODDS_RATIO: float(line[EDGESDATACOLS.LOG_ODDS_RATIO.value]),
+                    LOG_ODDS_RATIO_95_CI: orjson.loads(line[EDGESDATACOLS.LOG_ODDS_RATIO_95_CI.value]),
+                    TOTAL_SAMPLE_SIZE: int(line[EDGESDATACOLS.TOTAL_SAMPLE_SIZE.value])
+                }]
+            }).decode('utf-8')]
+    """
+