Add files via upload

create_genome_maps.py

@@ -0,0 +1,22 @@
+import pandas as pd 
+import requests 
+import sys
+import pickle
+from get_data import get_data_genome
+
+arg=sys.argv
+arg=arg[1:]
+
+if len(arg)!=3:
+    print("No. of arguments more or less. Please check")
+    sys.exit(1)
+
+dir_g="data"
+cmap,cimap,ld,ldg,a,d=get_data_genome(arg,dir_g)
+
+data=dict(cmap=cmap,cimap=cimap,ld=ld,ldg=ldg,a=a,d=d)
+
+with open("genome_maps","wb") as file:
+    pickle.dump(data,file)
+
+print("Genome Maps Created Successfully.")

get_data.py

@@ -10,16 +10,16 @@ def get_data_genome(arg,dir):
     ld=[]
     ldg=[]
     if arg[0]=='-d':
-        if arg[4]=="-r":
+        if arg[2]=="-r":
             c=0
         else:
-            return a,d,ld,ldg
+            return a,d,ld,ldg,dict(),dict()
     elif arg[0]=='-f':
         get_data_file(arg[1],dir)
     elif arg[0]=="-nd":
-        return ld,ldg,a,d
+        return ld,ldg,a,d,dict(),dict()
 
-    if arg[4]=="-r":
+    if arg[2]=="-r":
         a,d=read_data_genome(dir,a,d)
         assert(len(a)==len(d))
         print("Creating Maps:")

process_data.py

@@ -120,7 +120,7 @@ def get_nearest_neighbors(g,gs,n,a,d,ld,ldg,cmap,cimap):
         end=int(sldf.loc[itemp].end)
     return ne,nr
 
-def create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break,update):
+def create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break):
     lsy={} #dictionary which stores +/- n genes of the given gene by id. Each key is a gene id which corresponds to the one in center.
     t=0
     if os.path.exists("processed/neighbor_genes.json"):  
@@ -161,15 +161,8 @@ def create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_br
                 except:
                     continue
             t+=1
-            if t>=save_after and update==0:
-                t=0
-                c+=1
-                write_dict_json(name+str(c),"processed",lsytemp)
-                lsytemp={}
+            if t>=save_after:                
                 if enable_break==1:
                     break
-    if not update:
-        c+=1
-        write_dict_json(name+str(c),"processed",lsytemp)
     write_dict_json(name,"processed",lsy)
     return lsy

update_neighbor_genes.py

@@ -4,6 +4,7 @@ import pandas as pd
 import json
 import os
 import gc
+import pickle
 from get_data import get_data_homology,get_data_genome
 from process_data import create_data_homology_ls
 
@@ -20,13 +21,21 @@ if arg[-1]=="-test":
 
 arg=arg[:-1]
 
-if len(arg)!=5:
+if len(arg)!=3:
     print("No. of arguments more or less. Please check")
     sys.exit(1)
 
-dir_g="data"
-cmap,cimap,ld,ldg,a,d=get_data_genome(arg,dir_g)
+data={}
+with open("genome_maps","rb") as file:
+    data=pickle.load(file)
+cmap=data["cmap"]
+cimap=data["cimap"]
+ld=data["ld"]
+ldg=data["ldg"]
+a=data["a"]
+d=data["d"]
 
+print("Genome_Maps Loaded")
 dir_hom="data_homology"
 a_h,d_h=get_data_homology(arg,dir_hom)
 
@@ -38,10 +47,10 @@ print("Data Read")
 n=3 #no. of numbers neighbors
 save_after=5 #to save data after n steps
 
-lsy=create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break,1)
+lsy=create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break)
 print(len(lsy))
 
-print("Neighbor Genes Updated Successfully")
+print("Neighbor Genes Updated Successfully.")
 
 
 

update_neighbor_genes_ndf.py

@@ -4,6 +4,7 @@ import pandas as pd
 import json
 import os
 import gc
+import pickle
 from get_data import get_data_homology,get_data_genome
 from process_data import create_data_homology_ls
 
@@ -20,12 +21,19 @@ if arg[-1]=="-test":
 
 arg=arg[:-1]
 
-if len(arg)!=5:
+if len(arg)!=3:
     print("No. of arguments more or less. Please check")
     sys.exit(1)
 
-dir_g="data"
-cmap,cimap,ld,ldg,a,d=get_data_genome(arg,dir_g)
+data={}
+with open("genome_maps","rb") as file:
+    data=pickle.load(file)
+cmap=data["cmap"]
+cimap=data["cimap"]
+ld=data["ld"]
+ldg=data["ldg"]
+a=data["a"]
+d=data["d"]
 
 df=pd.read_hdf("negative_dataset.h5",key="ndf")
 a_h=[]
@@ -37,6 +45,6 @@ print("Data Read")
 n=3 #no. of numbers neighbors
 save_after=5 #to save data after n steps
 
-lsy=create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break,1)
+lsy=create_data_homology_ls(a_h,d_h,n,a,d,ld,ldg,cmap,cimap,save_after,enable_break)
 print(len(lsy))
 print("Neighbor Genes Updated Successfully")