Refactor biogrid and coronavirus_biogrid scripts

DaddyWesker · May 18, 2020 · 74fc6b7 · 74fc6b7
1 parent b07b920
commit 74fc6b7
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 36 deletions.
diff --git a/biogrid.py b/biogrid.py
@@ -15,18 +15,12 @@
 from zipfile import ZipFile
 from io import BytesIO
 import os
-import sys
+import collections
 import metadata
 from datetime import date
 from atomwrappers import *
 
 
-def checkdisc(diction, key, value):
-    try:
-        diction.setdefault(key, []).append(value)
-    except KeyError:
-        return "key error"
-
 
 def import_data_from_web(version, form='tab2'):
     if form not in ('tab2', 'tab3'):
@@ -87,7 +81,7 @@ def import_data(data, source, version, gene_level=False, form='tab2'):
 
     biogrid_path = os.path.join(dataset_path, 'biogrid_gene_gene_' + version + '_' + str(date.today()) + '.scm')
     with open(biogrid_path, 'w') as f:
-        pairs = {}
+        pairs = collections.defaultdict(list)
         entrez = []
         for i in range(len(data)):
             if not (pd.isnull(data.iloc[i]['Official Symbol Interactor A']) or pd.isnull(
@@ -103,7 +97,7 @@ def import_data(data, source, version, gene_level=False, form='tab2'):
                     interactors = node2 + ':' + node1
 
                 pubmed_link = 'https://www.ncbi.nlm.nih.gov/pubmed/?term=' + str(pubmed)
-                checkdisc(pairs, interactors, CConceptNode(pubmed_link))
+                pairs[interactors].append(CConceptNode(pubmed_link))
 
                 if not node1 in entrez:
                     entrez_1 = CConceptNode("entrez:" + str(data.iloc[i]['Entrez Gene Interactor A']))

diff --git a/coronavirus_biogrid.py b/coronavirus_biogrid.py
@@ -1,5 +1,5 @@
-__author__ = "Hedra"
-__email__ = "[email protected]"
+__author__ = "Hedra & Abdulrahman"
+__email__ = "[email protected] & [email protected]"
 
 # The following script imports SARS-CoV-2 (COVID-19) and Coronavirus-Related Interactions from thebiogrid.com
 
@@ -10,15 +10,15 @@
 # The version 183 is first release (March 25 2020), It can also be any of the latest versions with the same format
 
 import argparse
+import os
+import zipfile
+from datetime import date
+
 import pandas as pd
 import wget
-import os
-import sys
+
 import metadata
-from datetime import date
-import zipfile
 from atomwrappers import *
-from biogrid import checkdisc
 
 
 def evaLink(node1, node1_type, node2, node2_type, predicate, prefix1="", prefix2="", symmetric=False, stv=""):
@@ -50,31 +50,25 @@ def add_taxonomy(taxonomy_id, node, fp):
     fp.write(member_ln.recursive_print() + "\n")
 
 
-def add_protein_interaction(proteins_lst, prot_node_1, gene_node_1, prot_node_2, gene_node_2, bio_id_1, bio_id_2, fp):
-    if not prot_node_1.name in proteins_lst:
-        express_ln = CEvaluationLink(CPredicateNode("expresses"), CListLink(gene_node_1, prot_node_1))
-        gene_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
-                                   CListLink(gene_node_1, CConceptNode("Bio:" + bio_id_1)))
-        prot_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
-                                   CListLink(prot_node_1, CConceptNode("Bio:" + bio_id_1)))
+def define_protein_lns(gene_node, prot_node, bio_id, fp):
+    express_ln = CEvaluationLink(CPredicateNode("expresses"), CListLink(gene_node, prot_node))
+    gene_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
+                               CListLink(gene_node, CConceptNode("Bio:" + bio_id)))
+    prot_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
+                               CListLink(prot_node, CConceptNode("Bio:" + bio_id)))
 
-        fp.write(express_ln.recursive_print() + "\n")
-        fp.write(gene_bio.recursive_print() + "\n")
-        fp.write(prot_bio.recursive_print() + "\n")
+    fp.write(express_ln.recursive_print() + "\n")
+    fp.write(gene_bio.recursive_print() + "\n")
+    fp.write(prot_bio.recursive_print() + "\n")
 
+
+def add_protein_interaction(proteins_lst, prot_node_1, gene_node_1, prot_node_2, gene_node_2, bio_id_1, bio_id_2, fp):
+    if not prot_node_1.name in proteins_lst:
+        define_protein_lns(gene_node_1, prot_node_1, bio_id_1, fp)
         proteins_lst.append(prot_node_1.name)
 
     if not prot_node_2.name in proteins_lst:
-        express_ln = CEvaluationLink(CPredicateNode("expresses"), CListLink(gene_node_2, prot_node_2))
-        gene_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
-                                   CListLink(gene_node_2, CConceptNode("Bio:" + bio_id_2)))
-        prot_bio = CEvaluationLink(CPredicateNode("has_biogridID"),
-                                   CListLink(prot_node_2, CConceptNode("Bio:" + bio_id_2)))
-
-        fp.write(express_ln.recursive_print() + "\n")
-        fp.write(gene_bio.recursive_print() + "\n")
-        fp.write(prot_bio.recursive_print() + "\n")
-
+        define_protein_lns(gene_node_2, prot_node_2, bio_id_2, fp)
         proteins_lst.append(prot_node_2.name)