Add files via upload

This commit is contained in:
HarshitGupta11 2019-06-22 20:43:37 +05:30 committed by GitHub
parent 1168972b00
commit 81c967b5d4
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 54 additions and 22 deletions

22
create_genome_maps.py Normal file
View file

@ -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.")

View file

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

View file

@ -120,7 +120,7 @@ def get_nearest_neighbors(g,gs,n,a,d,ld,ldg,cmap,cimap):
end=int(sldf.loc[itemp].end) end=int(sldf.loc[itemp].end)
return ne,nr 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. 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 t=0
if os.path.exists("processed/neighbor_genes.json"): 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: except:
continue continue
t+=1 t+=1
if t>=save_after and update==0: if t>=save_after:
t=0
c+=1
write_dict_json(name+str(c),"processed",lsytemp)
lsytemp={}
if enable_break==1: if enable_break==1:
break break
if not update:
c+=1
write_dict_json(name+str(c),"processed",lsytemp)
write_dict_json(name,"processed",lsy) write_dict_json(name,"processed",lsy)
return lsy return lsy

View file

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

View file

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