def fromCancerList(n, value):
cutoff = value
casedict = {
'ACC': 92,
'BRCA': 1044,
'ESCA': 184,
'HNSC': 510,
'LAML': 149,
'MESO': 83,
'SKCM': 470,
'THCA': 496,
'COAD': 433
}
diseaseList = casedict.keys()
listCandidateProteins = [
] #list of all proteins in all files with DA > cutoff (not filtered for idiosync)
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if float(readfile[i][2]) > cutoff and readfile[i][
0] not in listCandidateProteins:
# scorelist.append(round(float(readfile[i][2]),5))
# listCandidateProteins.append((readfile[i][0],readfile[i][2])) #[(protein, DAscore), (protein, DAscore),(protein, DAscore)...]
listCandidateProteins.append(readfile[i][0])
fromCancerList = []
nProteins = numlist.cancerlist(n)
for protein in listCandidateProteins:
if protein in nProteins[1]:
fromCancerList.append(protein)
return fromCancerList
def idiosync(num):
cancerdictlist = countproteins()
genelistnum = cancerlist(num)
returndict = {}
for genel in genelistnum[1]:
tupl = ()
for dict in cancerdictlist:
for key in dict.keys():
for gene in dict[key].keys():
if genel == gene:
tupl += (key, )
if tupl not in returndict.keys():
returndict[tupl] = [genel]
else:
returndict[tupl].append(genel)
# print(returndict)
return returndict
def numCompareTcga(nin, value):
n = nin
cutoff = value
casedict = {
'ACC': 92,
'BRCA': 1044,
'ESCA': 184,
'HNSC': 510,
'LAML': 149,
'MESO': 83,
'SKCM': 470,
'THCA': 496,
'COAD': 433
}
diseaseList = casedict.keys()
listCandidateProteins = [
] #list of all proteins in all files with DA > cutoff (not filtered for idiosync)
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if float(readfile[i][2]) > cutoff and readfile[i][
0] not in listCandidateProteins:
# scorelist.append(round(float(readfile[i][2]),5))
# listCandidateProteins.append((readfile[i][0],readfile[i][2])) #[(protein, DAscore), (protein, DAscore),(protein, DAscore)...]
listCandidateProteins.append(readfile[i][0])
fromCancerList = []
nProteins = numlist.cancerlist(n)
for protein in listCandidateProteins:
if protein in nProteins[1]:
fromCancerList.append(protein)
'''
2) For each protein, calculate the percent it is DA in all TCGA files
represented as: (%, # DA, # normal)
'''
returnDict = {}
tcgaProteins = []
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if readfile[i][0] in fromCancerList and float(
readfile[i][2]) > cutoff:
if readfile[i][0] not in returnDict:
returnDict[readfile[i][0]] = 1
else:
returnDict[readfile[i][0]] += 1
tcgaProteins.append(readfile[i][0])
tcgaProteinCount = Counter(tcgaProteins)
for protein in returnDict:
returnDict[protein] = (round(
((returnDict[protein] / tcgaProteinCount[protein]) * 100),
1), returnDict[protein], tcgaProteinCount[protein])
# print(returnDict)
return returnDict
listCandidateProteins = [
] #list of all proteins in all files with DA > cutoff (not filtered for idiosync)
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if float(readfile[i][2]) > cutoff and readfile[i][
0] not in listCandidateProteins:
# scorelist.append(round(float(readfile[i][2]),5))
# listCandidateProteins.append((readfile[i][0],readfile[i][2])) #[(protein, DAscore), (protein, DAscore),(protein, DAscore)...]
listCandidateProteins.append(readfile[i][0])
fromCancerList = []
nProteins = numlist.cancerlist(n)
for protein in listCandidateProteins:
if protein in nProteins[1]:
fromCancerList.append(protein)
freq_dict = {}
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if readfile[i][0] in fromCancerList:
if readfile[i][0] not in freq_dict:
freq_dict[readfile[i][0]] = 1
else:
def relFreqDict(num, value):
n = num
cutoff = value
casedict = {
'ACC': 92,
'BRCA': 1044,
'ESCA': 184,
'HNSC': 510,
'LAML': 149,
'MESO': 83,
'SKCM': 470,
'THCA': 496,
'COAD': 433
}
diseaseList = casedict.keys()
listCandidateProteins = [
] #list of all proteins in all files with DA > cutoff (not filtered for idiosync)
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if float(readfile[i][2]) > cutoff and readfile[i][
0] not in listCandidateProteins:
# scorelist.append(round(float(readfile[i][2]),5))
# listCandidateProteins.append((readfile[i][0],readfile[i][2])) #[(protein, DAscore), (protein, DAscore),(protein, DAscore)...]
listCandidateProteins.append(readfile[i][0])
fromCancerList = []
nProteins = numlist.cancerlist(n)
for protein in listCandidateProteins:
if protein in nProteins[1]:
fromCancerList.append(protein)
freq_dict = {}
for disease in diseaseList:
filename = "epex_outputs\epexchosen_{}.txt".format(disease)
fin = open(filename)
readfile = fin.readlines()
for i in range(len(readfile)):
readfile[i] = readfile[i].split()
if readfile[i][0] in fromCancerList:
if readfile[i][0] not in freq_dict:
freq_dict[readfile[i][0]] = 1
else:
freq_dict[readfile[i][0]] += 1
# print(freq_dict)
'''
2. Make list of all of the frequencies. Compute mean, std, and z-score for each protein.
'''
''' Make list '''
freq_list = []
for protein in freq_dict:
freq_list.append(freq_dict[protein])
''' Compute mean '''
def Average(lst):
return sum(lst) / len(lst)
mean = Average(freq_list)
''' Compute standard deviation '''
import statistics
standardDev = statistics.stdev(freq_list)
''' Compute z-score for each protein. Create dictionary like the following: {protein: z-score, protein : z-score ...} '''
zscore_dict = {}
for protein in freq_dict:
zscore = round((freq_dict[protein] - mean) / standardDev, 2)
zscore_dict[protein] = zscore
# print(zscore_dict)
return zscore_dict