Correcting None node error in networkX

package/scripts/preprocess/B1.001/run.py

@@ -239,7 +239,8 @@ def create_class_prot():
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/B2.001/run.py

@@ -174,7 +174,8 @@ def create_class_prot():
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/B4.001/run.py

@@ -197,7 +197,8 @@ def create_class_prot():
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     kinase_idx = set([x for w in kinase_idx for k, v in nx.dfs_successors(
         G, w).items() for x in v] + kinase_idx)
@@ -241,7 +242,8 @@ def create_class_prot():
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/B5.001/run.py

@@ -76,7 +76,8 @@ def parse_chembl():
     G = nx.DiGraph()
 
     for s in S:
-        G.add_edge(s[1], s[0])  # The tree
+        if s[1] is not None:
+            G.add_edge(s[1], s[0])  # The tree
 
     S = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/C2.001/run.py

@@ -42,7 +42,8 @@ def prepare_hotnet_input(outdir, all_binary_sif):
             if l[0] in chebi_inchikey and l[2] in chebi_inchikey:
                 ik1 = chebi_inchikey[l[0]]
                 ik2 = chebi_inchikey[l[2]]
-                G.add_edge(ik1, ik2)
+                if ik1 is not None:
+                    G.add_edge(ik1, ik2)
 
     with open(os.path.join(outdir, pcomms_file), "w") as f:
         for e in G.edges():

package/scripts/preprocess/C3.001/run.py

@@ -183,7 +183,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     kinase_idx = set([x for w in kinase_idx for k, v in nx.dfs_successors(
         G, w).items() for x in v] + kinase_idx)
@@ -237,7 +238,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/C4.001/prot2allgos.py

@@ -17,7 +17,8 @@ G = nx.DiGraph()
 f = open(myfold + "/bp.tsv", "r")
 for l in f:
     l = l.rstrip("\n").split("\t")
-    G.add_edge(l[0], l[1])
+    if l[0] is not None:
+        G.add_edge(l[0], l[1])
 f.close()
 
 # Read Uniprot GOA

package/scripts/preprocess/C4.001/run.py

@@ -184,7 +184,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     kinase_idx = set([x for w in kinase_idx for k, v in nx.dfs_successors(
         G, w).items() for x in v] + kinase_idx)
@@ -238,7 +239,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 
@@ -449,7 +451,8 @@ def main(args):
                     parent = l.split("is_a: ")[1].split(" !")[0]
                     if parent == "regulates":
                         continue
-                    G.add_edge(parent, child)
+                    if parent is not None:
+                        G.add_edge(parent, child)
 
         f = open(os.path.join(args.models_path, "bp.tsv"), "w")
         for e in G.edges():

package/scripts/preprocess/C5.001/run.py

@@ -217,7 +217,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     kinase_idx = set([x for w in kinase_idx for k, v in nx.dfs_successors(
         G, w).items() for x in v] + kinase_idx)
@@ -271,7 +272,8 @@ def process_activity_according_to_pharos(ACTS):
     G = nx.DiGraph()
 
     for r in R:
-        G.add_edge(r[1], r[0])  # The tree
+        if r[1] is not None:
+            G.add_edge(r[1], r[0])  # The tree
 
     R = psql.qstring("SELECT cs.accession, cc.protein_class_id FROM component_sequences cs, component_class cc WHERE cs.component_id = cc.component_id AND cs.accession IS NOT NULL", chembl_dbname)
 

package/scripts/preprocess/D5.001/run.py

@@ -91,7 +91,8 @@ def parse_cellosaurus(R, cellosaurus_obo):
                 G.add_node(child)
             if "relationship: derived_from " in l:
                 parent = l.split("derived_from ")[1].split(" !")[0]
-                G.add_edge(parent, child)
+                if parent is not None:
+                    G.add_edge(parent, child)
             # if "relationship: originate_from_same_individual_as" in l:
             #    parent = l.split("originate_from_same_individual_as ")[1].split(" !")[0]
             #    G.add_edge(parent, child)
@@ -100,7 +101,8 @@ def parse_cellosaurus(R, cellosaurus_obo):
 
     for n in list(G.nodes()):
         if not nx.ancestors(G, n):
-            G.add_edge(n, "*")
+            if n is not None:
+                G.add_edge(n, "*")
 
     # Cell hierarchy
 

package/scripts/preprocess/E2.001/run.py

@@ -130,6 +130,7 @@ def include_mesh(ctd_diseases, IND):
             if pardisids == [""]:
                 pardisids = ["ROOT"]
             for pardisid in pardisids:
+                if pardisid is not None:
                 G.add_edge(pardisid, disid)
 
     classIND = collections.defaultdict(list)