Replaces deprecated pkg_resources dependency for finding DATA_PATH

ot_tractability_pipeline_v2/bin/run_pipeline.py

@@ -10,7 +10,6 @@
 
 import time
 import re
-#import zipfile
 import zlib
 import sys
 import argparse
@@ -19,7 +18,6 @@
 import numpy as np
 import pandas as pd
 import mygene
-import pkg_resources
 from sqlalchemy import create_engine
 import json
 import requests
@@ -33,13 +31,11 @@
 import psutil
 import gc
 
-PY3 = sys.version > '3'
+PY3 = float(sys.version) > 3.0
 if PY3:
-    import urllib.request as urllib2
-    import urllib.parse as urllib
+    pass
 else:
-    import urllib2
-    import urllib
+    import urllib2 as urllib_request
 
 from ot_tractability_pipeline_v2.queries_sm import *
 from ot_tractability_pipeline_v2.queries_ab import *
@@ -51,8 +47,7 @@
 from ot_tractability_pipeline_v2.buckets_protac import *
 from ot_tractability_pipeline_v2.buckets_othercl import *
 
-
-DATA_PATH = pkg_resources.resource_filename('ot_tractability_pipeline_v2', 'data/')
+from settings import DATA_PATH
 
 POLLING_INTERVAL = 3
 API_URL = "https://rest.uniprot.org"

ot_tractability_pipeline_v2/buckets_ab.py

@@ -16,23 +16,20 @@
 import sys
 import zipfile
 import os
-
 # import mygene
 import numpy as np
 import pandas as pd
-import pkg_resources
 from sqlalchemy import text
 
-PY3 = sys.version > '3'
+PY3 = float(sys.version) >= 3.0
 if PY3:
-    import urllib.request as urllib2
+    import urllib.request as urllib_request
 else:
-    import urllib2
-
+    import urllib2  as urllib_request
 
-from ot_tractability_pipeline_v2.queries_ab import *
+from ot_tractability_pipeline_v2 import queries_ab, settings
+CHEMBL_VERSION = settings.chembl_version_str()
 
-DATA_PATH = pkg_resources.resource_filename('ot_tractability_pipeline_v2', 'data/')
 
 class Antibody_buckets(object):
     '''
@@ -167,14 +164,14 @@ def _assign_buckets_1_to_3(self):
 
         print("\t- Assessing clinical buckets 1-3...")
 
-        self.all_chembl_targets = pd.read_sql_query(text(chembl_clinical_ab_targets), self.engine)
+        self.all_chembl_targets = pd.read_sql_query(text(queries_ab.chembl_clinical_ab_targets), self.engine)
         self.all_chembl_targets.loc[self.all_chembl_targets['ref_type'] == 'Expert', ['ref_id', 'ref_url']] = 'NA'
 
         #
 
         self._process_protein_complexes()
 
-        ab_info = pd.read_sql_query(text(chembl_clinical_ab), self.engine)
+        ab_info = pd.read_sql_query(text(queries_ab.chembl_clinical_ab), self.engine)
         self.all_chembl_targets = self.all_chembl_targets.merge(ab_info, how='left', on='parent_molregno')
 
         if self.store_fetched: 
@@ -260,11 +257,11 @@ def set_strings2(x):
     ##############################################################################################################
 
     # def make_request(self, url, data):
-    #     request = urllib2.Request(url)
+    #     request = urllib_request.Request(url)
 
     #     try:
-    #         url_file = urllib2.urlopen(request)
-    #     except urllib2.HTTPError as e:
+    #         url_file = urllib_request.urlopen(request)
+    #     except urllib_request.HTTPError as e:
     #         if e.code == 404:
     #             print("[NOTFOUND %d] %s" % (e.code, url))
     #         else:
@@ -277,10 +274,10 @@ def set_strings2(x):
     # Method is used in several workflows
     @staticmethod
     def make_request(url, data):
-        request = urllib2.Request(url)
+        request = urllib_request.Request(url)
         try:
-            url_file = urllib2.urlopen(request, data)
-        except urllib2.HTTPError as e:
+            url_file = urllib_request.urlopen(request, data)
+        except urllib_request.HTTPError as e:
             if e.code == 404:
                 print("[NOTFOUND %d] %s" % (e.code, url))
             else:
@@ -292,7 +289,7 @@ def make_request(url, data):
     def post_request_uniprot(url, data):
         #base = 'http://legacy.uniprot.org'
         #base = 'https://www.uniprot.org'
-        base = API_URL
+        base = settings.API_URL
         full_url = "%s/%s" % (base, url)
 
         if isinstance(data, (list, tuple)):
@@ -593,8 +590,8 @@ def _assign_bucket_9(self):
         print("\t- Assessing Human Protein Atlas main location bucket 9...")
 
         # Download latest file
-        #zip_file = urllib2.urlopen('https://www.proteinatlas.org/download/subcellular_location.tsv.zip')
-        zip_file = urllib2.urlopen('https://www.proteinatlas.org/download/tsv/subcellular_location.tsv.zip')
+        #zip_file = urllib_request.urlopen('https://www.proteinatlas.org/download/subcellular_location.tsv.zip')
+        zip_file = urllib_request.urlopen('https://www.proteinatlas.org/download/tsv/subcellular_location.tsv.zip')
         with zipfile.ZipFile(io.BytesIO(zip_file.read()), 'r') as pa_file:
             with pa_file.open('subcellular_location.tsv') as subcell_loc:
                 df = pd.read_csv(subcell_loc, sep='\t', header=0)

ot_tractability_pipeline_v2/buckets_othercl.py

@@ -20,7 +20,6 @@
 # import mygene
 import numpy as np
 import pandas as pd
-import pkg_resources
 from sqlalchemy import create_engine, text
 
 PY3 = sys.version > '3'
@@ -32,7 +31,8 @@
 
 from ot_tractability_pipeline_v2.queries_othercl import *
 
-DATA_PATH = pkg_resources.resource_filename('ot_tractability_pipeline_v2', 'data/')
+# Not used in this file.
+from settings import DATA_PATH
 
 class Othercl_buckets(object):
     '''

ot_tractability_pipeline_v2/buckets_protac.py

@@ -20,24 +20,23 @@
 # import mygene
 import numpy as np
 import pandas as pd
-import pkg_resources
 import ast
 # from sqlalchemy import create_engine
 
 import requests
 import psutil
 import gc
 
-PY3 = sys.version > '3'
+PY3 = float(sys.version) >= 3.0
 if PY3:
-    import urllib.request as urllib2
+    import urllib.request as urllib_request
 else:
-    import urllib2
+    import urllib2 as urllib_request
 
+from ot_tractability_pipeline_v2 import queries_protac, settings
 
-from ot_tractability_pipeline_v2.queries_protac import *
+CHEMBL_VERSION = settings.chembl_version_str()
 
-DATA_PATH = pkg_resources.resource_filename('ot_tractability_pipeline_v2', 'data/')
 
 class Protac_buckets(object):
     '''
@@ -470,7 +469,7 @@ def _search_papers():
         #url = "https://www.ebi.ac.uk/europepmc/webservices/rest/search?resultType=core&query=ABSTRACT%3A%28%28ABSTRACT%3A%22proteolysis%20targeting%20chimera%22%20OR%20ABSTRACT%3A%22proteolysis%20targeting%20chimeric%22%29%20OR%20%28PROTAC%20AND%20%28degradation%20OR%20degrade%20OR%20ubiquitin%20OR%20proteolysis%29%29%29&resultType=lite&cursorMark=*&pageSize=1000&format=json"
 
         # urllib not working returning 404 page error -> replaced by requests
-        #url = urllib2.urlopen("https://www.ebi.ac.uk/europepmc/webservices/rest/search?resultType=core&query=ABSTRACT%3A%28%28ABSTRACT%3A%22proteolysis%20targeting%20chimera%22%20OR%20ABSTRACT%3A%22proteolysis%20targeting%20chimeric%22%29%20OR%20%28%28PROTAC%20OR%20SNIPER%29%20AND%20%28degradation%20OR%20degrade%20OR%20proteolysis%29%29%20OR%20%28degrader%20AND%20%28proteasome%20OR%20ubiquitin%29%29%29&resultType=lite&cursorMark=*&pageSize=1000&format=json")
+        #url = urllib_request.urlopen("https://www.ebi.ac.uk/europepmc/webservices/rest/search?resultType=core&query=ABSTRACT%3A%28%28ABSTRACT%3A%22proteolysis%20targeting%20chimera%22%20OR%20ABSTRACT%3A%22proteolysis%20targeting%20chimeric%22%29%20OR%20%28%28PROTAC%20OR%20SNIPER%29%20AND%20%28degradation%20OR%20degrade%20OR%20proteolysis%29%29%20OR%20%28degrader%20AND%20%28proteasome%20OR%20ubiquitin%29%29%29&resultType=lite&cursorMark=*&pageSize=1000&format=json")
         #data = url.read()
         #try: data = json.loads(data.decode())
         #except UnicodeDecodeError: data = json.loads(data)
@@ -509,7 +508,7 @@ def _get_tagged_targets(self):
         tags_list = []
         for chunk in chunks:
             url_s = 'https://www.ebi.ac.uk/europepmc/annotations_api/annotationsByArticleIds?{}&type=Gene_Proteins&format=JSON'.format(chunk)
-            url = urllib2.urlopen(url_s)
+            url = urllib_request.urlopen(url_s)
             data = url.read()
             try: data = json.loads(data.decode())
             except UnicodeDecodeError: data = json.loads(data)
@@ -1050,7 +1049,7 @@ def _search_chembl_smallmol_active(self):
         self.engine = create_engine(database_url)
 
         # small_mol_info = pd.read_sql_query(text(chembl_small_mol), self.engine)
-        self.all_chembl_targets = pd.read_sql_query(text(chembl_small_mol_active_targets), self.engine)
+        self.all_chembl_targets = pd.read_sql_query(text(queries_protac.chembl_small_mol_active_targets), self.engine)
         # self.all_chembl_targets = self.all_chembl_targets.merge(small_mol_info, on='parent_molregno')
 
         if self.store_fetched: 

ot_tractability_pipeline_v2/buckets_sm.py

@@ -20,20 +20,19 @@
 # import mygene
 import numpy as np
 import pandas as pd
-import pkg_resources
 import itertools
 from sqlalchemy import text
 
-PY3 = sys.version > '3'
+PY3 = float(sys.version) >= 3.0
 if PY3:
-    import urllib.request as urllib2
+    import urllib.request as urllib_request
 else:
-    import urllib2
+    import urllib2 as urllib_request
 
 
-from ot_tractability_pipeline_v2.queries_sm import *
+from ot_tractability_pipeline_v2 import queries_sm, settings
 
-DATA_PATH = pkg_resources.resource_filename('ot_tractability_pipeline_v2', 'data/')
+CHEMBL_VERSION = settings.chembl_version_str()
 
 
 class Small_molecule_buckets(object):
@@ -128,10 +127,10 @@ def _search_chembl_clinical(self):
         '''
 
         # print("\t- Querying ChEMBL...")
-        self.all_chembl_targets = pd.read_sql_query(text(chembl_clinical_targets), self.engine)
+        self.all_chembl_targets = pd.read_sql_query(text(queries_sm.chembl_clinical_targets), self.engine)
         if self.store_fetched: 
             self.all_chembl_targets.to_csv("{}/sm_all_chembl_clinical_targets.csv".format(self.store_fetched))
-        small_mol_info = pd.read_sql_query(text(chembl_clinical_small_mol), self.engine)
+        small_mol_info = pd.read_sql_query(text(queries_sm.chembl_clinical_small_mol), self.engine)
         if self.store_fetched: 
             small_mol_info.to_csv("{}/sm_all_chembl_clinical_small_mol.csv".format(self.store_fetched))
         self.all_chembl_targets = self.all_chembl_targets.merge(small_mol_info, how='left', on='parent_molregno')
@@ -294,11 +293,11 @@ def set_strings2(x):
     ##############################################################################################################
 
     # def make_request(self, url, data):
-    #     request = urllib2.Request(url)
+    #     request = urllib_request.Request(url)
 
     #     try:
-    #         url_file = urllib2.urlopen(request, data)
-    #     except urllib2.HTTPError as e:
+    #         url_file = urllib_request.urlopen(request, data)
+    #     except urllib_request.HTTPError as e:
     #         if e.code == 404:
     #             print("[NOTFOUND %d] %s" % (e.code, url))
     #         else:
@@ -311,11 +310,11 @@ def set_strings2(x):
     # Method is used in several workflows
     @staticmethod
     def make_request(url, data):
-        request = urllib2.Request(url)
+        request = urllib_request.Request(url)
 
         try:
-            url_file = urllib2.urlopen(request, data)
-        except urllib2.HTTPError as e:
+            url_file = urllib_request.urlopen(request, data)
+        except urllib_request.HTTPError as e:
             if e.code == 404:
                 print("[NOTFOUND %d] %s" % (e.code, url))
             else:
@@ -342,8 +341,10 @@ def _pdb_list(self, s):
         if not isinstance(pdb, list): pdb = [pdb]
 
         # Python 2/3 compatability
-        try: pdb = [p.lower() for p in pdb if isinstance(p,(str,unicode))] #Python 2
-        except: pdb = [p.lower() for p in pdb if isinstance(p,str)] #Python 3
+        if not PY3:
+            pdb = [p.lower() for p in pdb if isinstance(p,(str,unicode))] #Python 2
+        else:
+            pdb = [p.lower() for p in pdb if isinstance(p,str)] #Python 3
 
         self.pdb_list += pdb
         for p in pdb:

ot_tractability_pipeline_v2/queries_sm.py

@@ -3,15 +3,8 @@
 
 import os
 
-try:
-    chembl_number = int(os.getenv('CHEMBL_VERSION'))
-    CHEMBL_VERSION = "chembl_{}".format(chembl_number)
-    print("\t  ChEMBL version in use is "+CHEMBL_VERSION)
-except TypeError:
-    print('''
-    Please set the CHEMBL version '''
-          )
-    raise
+CHEMBL_VERSION = chembl_version_str()
+global CHEMBL_VERSION
 
 
 chembl_clinical_small_mol = """

ot_tractability_pipeline_v2/settings.py

@@ -0,0 +1,5 @@
+import os
+import inspect
+
+DATA_PATH = os.path.join(os.path.dirname(inspect.getfile(inspect.currentframe())), "data{}".format(os.sep))
+# print(DATA_PATH)