Update data acquisition

Probably need to rewrite since this was for drugbank 4

.flake8

@@ -20,8 +20,8 @@ max-complexity = 10
 import-order-style = pycharm
 application-import-names =
     bio2bel_drugbank
-    bio2bel_uniprot
     tests
     bio2bel
     pybel
+    pyobo
 format = ${cyan}%(path)s${reset}:${yellow_bold}%(row)d${reset}:${green_bold}%(col)d${reset}: ${red_bold}%(code)s${reset} %(text)s

setup.py

@@ -24,16 +24,15 @@
     'License :: OSI Approved :: MIT License',
 ]
 INSTALL_REQUIRES = [
-    'pybel>=0.13.0,<0.14.0',
-    'bio2bel>=0.2.0,<0.3.0',
-    'bio2bel_hgnc>=0.1.0',
+    'pybel>=0.15.0',
+    'bio2bel',
     'click',
     'pandas',
     'sqlalchemy',
     'tqdm',
     'beautifulsoup4',
     'requests',
-    'bio2bel_uniprot',
+    'drugbank_downloader',
 ]
 EXTRAS_REQUIRE = {
     'web': [

src/bio2bel_drugbank/constants.py

@@ -10,6 +10,3 @@
 
 MODULE_NAME = 'drugbank'
 DATA_DIR = get_data_dir(MODULE_NAME)
-
-DRUGBANK_URL = 'https://www.drugbank.ca/releases/5-1-4/downloads/all-full-database'
-DRUGBANK_PATH = os.path.join(DATA_DIR, 'drugbank_all_full_database.xml.zip')

src/bio2bel_drugbank/manager.py

@@ -9,13 +9,12 @@
 from collections import Counter, defaultdict
 from typing import Dict, Iterable, List, Optional, TextIO, Tuple
 
-
 import click
 import networkx as nx
+import pyobo
 from sqlalchemy import func
 from tqdm import tqdm
 
-import bio2bel_hgnc
 from bio2bel import AbstractManager
 from bio2bel.manager.bel_manager import BELManagerMixin
 from bio2bel.manager.flask_manager import FlaskMixin
@@ -31,6 +30,7 @@
 )
 from .parser import extract_drug_info, get_xml_root
 
+
 __all__ = ['Manager']
 
 log = logging.getLogger(__name__)
@@ -579,14 +579,10 @@ def to_bel(self, drug_namespace: Optional[str] = None, target_namespace: Optiona
         """Export DrugBank as BEL."""
         graph = BELGraph(
             name='DrugBank',
-            version='5.1.4',
         )
 
         self.add_namespace_to_graph(graph)
 
-        hgnc_manager = bio2bel_hgnc.Manager(engine=self.engine, session=self.session)
-        hgnc_manager.add_namespace_to_graph(graph)
-
         dpis = self.list_drug_protein_interactions()
         dpis: Iterable[DrugProteinInteraction] = tqdm(
             dpis,
@@ -651,18 +647,12 @@ def get_drug_to_hgnc_symbols(self, cache=True, recalculate=False) -> Dict[str, L
             with open(_dti_symbols_cache_path) as file:
                 return json.load(file)
 
-        hgnc_manager = bio2bel_hgnc.Manager(engine=self.engine, session=self.session)
-        if not hgnc_manager.is_populated():
-            hgnc_manager.populate()
-
-        hgnc_id_symbol_mapping = hgnc_manager.build_hgnc_id_symbol_mapping()
         drug_to_hgnc_ids = self.get_drug_to_hgnc_ids()
-
         rv = defaultdict(list)
 
         for drug, hgnc_ids in drug_to_hgnc_ids.items():
             for hgnc_id in hgnc_ids:
-                hgnc_symbol = hgnc_id_symbol_mapping.get(hgnc_id)
+                hgnc_symbol = pyobo.get_name('hgnc', hgnc_id)
 
                 if hgnc_symbol is None:
                     log.warning('could not map HGNC identifier: %s', hgnc_id)

src/bio2bel_drugbank/parser.py

@@ -4,18 +4,16 @@
 
 import itertools as itt
 import logging
-import os
 import re
-import sys
-import time
-import zipfile
 from datetime import datetime
 from typing import Mapping, Optional
 from xml.etree import ElementTree
 
+from drugbank_downloader import parse_drugbank
 from tqdm import tqdm
 
-from bio2bel_drugbank.constants import DRUGBANK_PATH
+import pyobo.config
+
 
 log = logging.getLogger(__name__)
 
@@ -27,52 +25,19 @@
 pubmed_re = re.compile('pubmed/([0-9]+)')
 
 
-def get_path(path: Optional[str] = None) -> str:
-    """Get the path to the DrugBank data."""
-    if path is not None:
-        return path
-
-    if os.path.exists(DRUGBANK_PATH):
-        return DRUGBANK_PATH
-
-    log.critical("""DrugBank data not found.
-
-Unfortunately, the data for DrugBank is not available via an open HTTP(S) or FTP endpoint, so please follow these 
-steps to get it yourself:
-
-1. Make an account at: https://www.drugbank.ca
-2. Navigate to: https://www.drugbank.ca/releases/5-1-1/downloads/all-full-database
-3. Unzip the file that gets downloaded. There should be an XML called "full database.xml"
-4. Run the following command on the command line: `mkdir -p ~/.bio2bel/drugbank`. If you're specifying the Bio2BEL
-   directory through a config file or an environment variable, just make a "drugbank" folder there.
-5. Copy "full database.xml" into this folder.
-6. Resume business as usual (try `bio2bel_drugbank populate` now)""")
-    sys.exit(0)
-
-
-def get_xml_root(path: Optional[str] = None) -> ElementTree.Element:
+def get_xml_root(path: Optional[str] = None):
     """Get the XML parser root.
 
     Takes between 35-60 seconds.
 
     :param path: A custom URL for DrugBank XML file
     """
-    path = get_path(path=path)
-    log.info('parsing drugbank at %s', path)
-    t = time.time()
-
-    if path.endswith('.xml'):
-        tree = ElementTree.parse(path)
-    elif path.endswith('.zip'):
-        with zipfile.ZipFile(path) as z:
-            with z.open('full database.xml') as f:
-                tree = ElementTree.parse(f)
-    else:
-        raise ValueError
-
-    log.info('parsed drugbank in %.2f seconds', time.time() - t)
-
-    return tree.getroot()
+    if path:
+        return ElementTree.parse(path).getroot()
+    return parse_drugbank(
+        username=pyobo.config.get_config('drugbank_username'),
+        password=pyobo.config.get_config('drugbank_password'),
+    )
 
 
 def extract_drug_info(drug_xml: ElementTree.Element):
@@ -232,11 +197,12 @@ def get_pubchem_to_drugbank(path=None) -> Mapping[str, str]:
 
 
 def main():
-    x = get_pubchem_to_drugbank('/Users/cthoyt/.bio2bel/drugbank/test.xml')
     import json
+    import os
 
-    with open('/Users/cthoyt/Desktop/pubchem_to_drugbank.json', 'w') as f:
-        json.dump(x, f)
+    x = get_pubchem_to_drugbank()
+    with open(os.path.expanduser('~/Desktop/pubchem_to_drugbank.json'), 'w') as f:
+        json.dump(x, f, indent=2)
 
 
 if __name__ == '__main__':