db="TESTING")
cursor = conn_obj.cursor()
cursor.execute("SELECT NAME FROM CSA_catalyticCofactors")
print cursor.fetchall()
cursor.execute("SHOW TABLES FROM TESTING")
for table in cursor.fetchall():
print list(table)
cursor.execute("SHOW FIELDS FROM CSA_catalyticCofactors")
print cursor.fetchall()
cursor.execute(
"SELECT column_name FROM information_schema.columns WHERE table_name='CSA_catalyticCofactors'"
)
print cursor.fetchall()
cursor.execute(
"SELECT PDBID, CHAIN,NUMBER, UPNUMBER , TYPE FROM CSA_CATALYTICRESIDUES")
RESULTS = [list(X) for X in cursor.fetchall()]
print len(RESULTS)
RESULTS = unique(RESULTS)
print RESULTS
print len(RESULTS)
RESULTS_NULL = [X for X in RESULTS if X[3] == None]
print len(RESULTS_NULL), RESULTS_NULL
#CANCER_NETWORK_MAPPING_2 - DONE IN NETWORK WITH PRE-JULY DATA
#Check how many genes are actual present in network
import pickle
from FUNCTIONS import UNIPROT_TO_GENES_CLASS, unique
#Load CANCER_to_Gene dict
FILE_IN1=open("DATABASES/OBJECTS/DICT_CANCER_TO_GENES.pi")
CANCER_DICT=pickle.load(FILE_IN1)
FILE_IN1.close()
#Get all cancer genes
ALL_CANCER_GENES=[]
for genes in CANCER_DICT.itervalues():
ALL_CANCER_GENES=ALL_CANCER_GENES+genes
ALL_CANCER_GENES=unique(ALL_CANCER_GENES)
print len(ALL_CANCER_GENES)
#Load NETWORK UNIPROTS
FILE_IN2=open("NETWORK/ENDOGENOUS_UNIPROT")
UNIPROTS=FILE_IN2.read().splitlines()
#Run class
UNIPROT_CLASS=UNIPROT_TO_GENES_CLASS(UNIPROTS)
NETWORK_GENES=UNIPROT_CLASS.get_genes()
NETWORK_UNIPROT_GENES=UNIPROT_CLASS.uniprot_gene()
print "uniprot in network has match", len(UNIPROT_CLASS.has_genes_uniprots())
print "total cancer genes in network", len([X for X in ALL_CANCER_GENES if X in NETWORK_GENES])
#Check per cancer and make UNIPROT_DICT
CANCER_TO_UNIPROT_DICT=dict((X, []) for X in CANCER_DICT.iterkeys())
METABOLITES_ZERO=[]
for metabolite in METABO_EN.nodes():
if METABO_EN.in_degree(metabolite)==0 and METABO_EN.out_degree(metabolite)>0:
METABOLITES.append(metabolite)
elif METABO_EN.in_degree(metabolite)==0 and METABO_EN.out_degree(metabolite)==0:
METABOLITES_ZERO.append(metabolite)
print "METABOLITES:", len (METABOLITES)
print len (METABOLITES) + len (METABOLITES_ZERO)
print METABOLITES[:20]
#Get proteins only
PROTEINS=[]
for protein in METABO_EN.nodes():
if METABO_EN.in_degree(protein)>0:
PROTEINS.append(protein)
print "PROTEINS:",len (PROTEINS)
print "UNIQUE PROTEINS:", len (unique(PROTEINS))
print PROTEINS[:5]
#DRAW NETWORK
plt.figure(figsize=(25,12))
pos=NX.spectral_layout(METABO_EN)
NX.draw_networkx_nodes(METABO_EN, pos, nodelist=METABOLITES, node_color="r", node_size=100)
#NX.draw_networkx_nodes(METABO_EN, pos, nodelist=METABOLITES_ZERO, node_color="b", node_size=100) #EXCLUDE ZEROES ALTOGETHER
#FOR NOW
NX.draw_networkx_nodes(METABO_EN, pos, nodelist=PROTEINS, node_color="g", node_size=10)
NX.draw_networkx_edges(METABO_EN, pos, width=0.2)
plt.show()
#Separate cancer genes by cancer type
import subprocess, pickle
from FUNCTIONS import unique
#Get genes in stats
GENES_STATS=subprocess.check_output("ls", cwd="DATABASES/CANCER_DATA/geneStats").splitlines()
#CANCER TYPES
CANCER_DICT={"LUSC":[], "READ":[], "GBM":[], "KIRC":[], "UCEC":[], "OV":[], "BRCA":[], "COAD":[]}
#Assign to CANCER TYPES
for gene in GENES_STATS:
FILE_IN=open("DATABASES/CANCER_DATA/geneStats/%s"%gene)
RECORDS=FILE_IN.read().splitlines()
FILE_IN.close()
for record in RECORDS:
CANCER_DICT[record.split()[4].strip()]=CANCER_DICT[record.split()[4].strip()]+[gene.split(".")[0]]
for cancer in CANCER_DICT.iterkeys():
CANCER_DICT[cancer]=unique(CANCER_DICT[cancer])
print cancer, CANCER_DICT[cancer]
#Store as object
PICKLE_OUT=open("DATABASES/OBJECTS/DICT_CANCER_TO_GENES.pi", "w")
pickle.dump(CANCER_DICT, PICKLE_OUT)
for cancer in CANCER_DICT.iterkeys():
print cancer, len(CANCER_DICT[cancer])
metabolite) == 0 and METABO_EN.out_degree(metabolite) > 0:
METABOLITES.append(metabolite)
elif METABO_EN.in_degree(metabolite) == 0 and METABO_EN.out_degree(
metabolite) == 0:
METABOLITES_ZERO.append(metabolite)
print "METABOLITES:", len(METABOLITES)
print len(METABOLITES) + len(METABOLITES_ZERO)
print METABOLITES[:20]
#Get proteins only
PROTEINS = []
for protein in METABO_EN.nodes():
if METABO_EN.in_degree(protein) > 0:
PROTEINS.append(protein)
print "PROTEINS:", len(PROTEINS)
print "UNIQUE PROTEINS:", len(unique(PROTEINS))
print PROTEINS[:5]
#DRAW NETWORK
plt.figure(figsize=(25, 12))
pos = NX.spectral_layout(METABO_EN)
NX.draw_networkx_nodes(METABO_EN,
pos,
nodelist=METABOLITES,
node_color="r",
node_size=100)
#NX.draw_networkx_nodes(METABO_EN, pos, nodelist=METABOLITES_ZERO, node_color="b", node_size=100) #EXCLUDE ZEROES ALTOGETHER
#FOR NOW
NX.draw_networkx_nodes(METABO_EN,
pos,
UNIPROT_GENE_DICT = {}
for cancer in CANCER_DICT.itervalues():
for record in cancer:
UNIPROT_GENE_DICT[record.split("|")[0]] = record.split("|")[1]
#BREAK DOWN NODES BY CANCER
OV_PROTEINS = [
X for X in PROTEINS
if X.split("|")[1] in [Y.split("|")[0] for Y in CANCER_DICT["OV"]]
]
print "OV_PROTEINS", len(OV_PROTEINS)
OV_METABOLITES = []
for protein in OV_PROTEINS:
OV_METABOLITES = OV_METABOLITES + METABO_EN.neighbors(protein)
OV_METABOLITES = unique(OV_METABOLITES)
print "OV_METABOLITES", len(OV_METABOLITES)
OV_EDGES = METABO_EN.edges(OV_PROTEINS)
OV_EDGES = unique(OV_EDGES)
OV_LABELS_PROTEINS = {}
for protein in OV_PROTEINS:
OV_LABELS_PROTEINS[protein] = UNIPROT_GENE_DICT[protein.split("|")[1]]
OV_LABELS_METABOLITES = dict(
(X[0], X[1]) for X in zip(OV_METABOLITES, OV_METABOLITES))
#DRAW NETWORK
plt.figure(figsize=(25, 12))
pos = NX.spring_layout(METABO_EN)
#GET ALL ENDOGENOUS UNIPROTS FROM NETWORK
from FUNCTIONS import unique
FILE_IN=open("NETWORK/ENDOGENOUS.ADJ").read().splitlines()
ENDOGENOUS_UNIPROT=[]
for record in FILE_IN:
if len(record.split())>1:
ENDOGENOUS_UNIPROT=ENDOGENOUS_UNIPROT+[X.split("|")[1].strip() for X in record.split()[1:]]
print len(ENDOGENOUS_UNIPROT), ENDOGENOUS_UNIPROT
ENDOGENOUS_UNIPROT=unique(ENDOGENOUS_UNIPROT)
print len(ENDOGENOUS_UNIPROT)
FILE_OUT=open("NETWORK/ENDOGENOUS_UNIPROT", "w")
for record in ENDOGENOUS_UNIPROT:
FILE_OUT.write(record+"\n")
FILE_OUT.close()
#Create connection object that represents database
conn_obj=MQ.connect(host="localhost",user="******", passwd="mysql", db="TESTING")
cursor=conn_obj.cursor()
cursor.execute("SELECT NAME FROM CSA_catalyticCofactors")
print cursor.fetchall()
cursor.execute("SHOW TABLES FROM TESTING")
for table in cursor.fetchall():
print list(table)
cursor.execute("SHOW FIELDS FROM CSA_catalyticCofactors")
print cursor.fetchall()
cursor.execute("SELECT column_name FROM information_schema.columns WHERE table_name='CSA_catalyticCofactors'")
print cursor.fetchall()
cursor.execute("SELECT PDBID, CHAIN,NUMBER, UPNUMBER , TYPE FROM CSA_CATALYTICRESIDUES")
RESULTS= [list(X) for X in cursor.fetchall()]
print len(RESULTS)
RESULTS=unique(RESULTS)
print RESULTS
print len(RESULTS)
RESULTS_NULL=[X for X in RESULTS if X[3]==None]
print len(RESULTS_NULL), RESULTS_NULL
#GET ALL ENDOGENOUS UNIPROTS FROM NETWORK
from FUNCTIONS import unique
FILE_IN = open("NETWORK/ENDOGENOUS.ADJ").read().splitlines()
ENDOGENOUS_UNIPROT = []
for record in FILE_IN:
if len(record.split()) > 1:
ENDOGENOUS_UNIPROT = ENDOGENOUS_UNIPROT + [
X.split("|")[1].strip() for X in record.split()[1:]
]
print len(ENDOGENOUS_UNIPROT), ENDOGENOUS_UNIPROT
ENDOGENOUS_UNIPROT = unique(ENDOGENOUS_UNIPROT)
print len(ENDOGENOUS_UNIPROT)
FILE_OUT = open("NETWORK/ENDOGENOUS_UNIPROT", "w")
for record in ENDOGENOUS_UNIPROT:
FILE_OUT.write(record + "\n")
FILE_OUT.close()
TO_R2=open("/Users/jzamalloa/Desktop/FOLDER/LAB/ECLIPSE_TO_R/HUBS_90_DEGREE", "w")
HUBS_90_DEGREE=[]
HUBS_90_PROTEINS=[]
for metabolite in CLEAN_TC_LIST:
HUBS_90_DEGREE.append(METABO_NT.out_degree(metabolite))
TO_R2.write(str(METABO_NT.out_degree(metabolite))+"\n")
PROTEINS=METABO_NT.neighbors(metabolite)
for prot in PROTEINS:
HUBS_90_PROTEINS.append(prot.split("|")[1])
TO_R2.close()
print len(HUBS_90_DEGREE)
print len(HUBS_90_PROTEINS)
print HUBS_90_PROTEINS[:5]
HUBS_90_PROTEINS=unique(HUBS_90_PROTEINS)
print len(HUBS_90_PROTEINS)
#Plot degree distribution
plt.hist(HUBS_90_DEGREE)
plt.show()
#Get gene names out of these proteins, use uniprot_sprot.dat
HUBS_90_GENES=[]
for protein in HUBS_90_PROTEINS:
PROT_GENE=UNIPROT_GENE_PLUS(protein)
for gene in PROT_GENE:
HUBS_90_GENES.append(protein+"$"+gene)
print HUBS_90_GENES[0:5]
print len(HUBS_90_GENES)
CANCER_DICT=pickle.load(FILE_IN1)
#Make {UNIPROT:GENE} dictionary
UNIPROT_GENE_DICT={}
for cancer in CANCER_DICT.itervalues():
for record in cancer:
UNIPROT_GENE_DICT[record.split("|")[0]]=record.split("|")[1]
#BREAK DOWN NODES BY CANCER
OV_PROTEINS=[X for X in PROTEINS if X.split("|")[1] in [Y.split("|")[0] for Y in CANCER_DICT["OV"]]]
print "OV_PROTEINS", len(OV_PROTEINS)
OV_METABOLITES=[]
for protein in OV_PROTEINS:
OV_METABOLITES=OV_METABOLITES+METABO_EN.neighbors(protein)
OV_METABOLITES=unique(OV_METABOLITES)
print "OV_METABOLITES", len(OV_METABOLITES)
OV_EDGES=METABO_EN.edges(OV_PROTEINS)
OV_EDGES=unique(OV_EDGES)
OV_LABELS_PROTEINS={}
for protein in OV_PROTEINS:
OV_LABELS_PROTEINS[protein]=UNIPROT_GENE_DICT[protein.split("|")[1]]
OV_LABELS_METABOLITES=dict((X[0],X[1]) for X in zip(OV_METABOLITES, OV_METABOLITES))
#DRAW NETWORK
plt.figure(figsize=(25,12))
pos=NX.spring_layout(METABO_EN)
#CANCER_NETWORK_MAPPING_2 - DONE IN NETWORK WITH PRE-JULY DATA
#Check how many genes are actual present in network
import pickle
from FUNCTIONS import UNIPROT_TO_GENES_CLASS, unique
#Load CANCER_to_Gene dict
FILE_IN1 = open("DATABASES/OBJECTS/DICT_CANCER_TO_GENES.pi")
CANCER_DICT = pickle.load(FILE_IN1)
FILE_IN1.close()
#Get all cancer genes
ALL_CANCER_GENES = []
for genes in CANCER_DICT.itervalues():
ALL_CANCER_GENES = ALL_CANCER_GENES + genes
ALL_CANCER_GENES = unique(ALL_CANCER_GENES)
print len(ALL_CANCER_GENES)
#Load NETWORK UNIPROTS
FILE_IN2 = open("NETWORK/ENDOGENOUS_UNIPROT")
UNIPROTS = FILE_IN2.read().splitlines()
#Run class
UNIPROT_CLASS = UNIPROT_TO_GENES_CLASS(UNIPROTS)
NETWORK_GENES = UNIPROT_CLASS.get_genes()
NETWORK_UNIPROT_GENES = UNIPROT_CLASS.uniprot_gene()
print "uniprot in network has match", len(UNIPROT_CLASS.has_genes_uniprots())
print "total cancer genes in network", len(
[X for X in ALL_CANCER_GENES if X in NETWORK_GENES])
#Check per cancer and make UNIPROT_DICT
"LUSC": [],
"READ": [],
"GBM": [],
"KIRC": [],
"UCEC": [],
"OV": [],
"BRCA": [],
"COAD": []
}
#Assign to CANCER TYPES
for gene in GENES_STATS:
FILE_IN = open("DATABASES/CANCER_DATA/geneStats/%s" % gene)
RECORDS = FILE_IN.read().splitlines()
FILE_IN.close()
for record in RECORDS:
CANCER_DICT[record.split()[4].strip(
)] = CANCER_DICT[record.split()[4].strip()] + [gene.split(".")[0]]
for cancer in CANCER_DICT.iterkeys():
CANCER_DICT[cancer] = unique(CANCER_DICT[cancer])
print cancer, CANCER_DICT[cancer]
#Store as object
PICKLE_OUT = open("DATABASES/OBJECTS/DICT_CANCER_TO_GENES.pi", "w")
pickle.dump(CANCER_DICT, PICKLE_OUT)
for cancer in CANCER_DICT.iterkeys():
print cancer, len(CANCER_DICT[cancer])