def main():
parser = argparse.ArgumentParser()
parser.add_argument('--cancerType', dest='type',\
help='Cancer type to be collected')
parser.add_argument('--getData',dest='get', action='store_true',\
default=False,help='Set flag to get all data')
opts = parser.parse_args()
if opts.get:
for ds in ['brca', 'ccrcc', 'colon', 'ovarian', 'endometrial', 'luad']:
cptac.download(dataset=ds)
if opts.type.lower() == 'brca':
dat = cptac.Brca()
elif opts.type.lower() == 'ccrcc':
dat = cptac.Ccrcc()
elif opts.type.lower() == 'coad':
dat = cptac.Colon()
elif opts.type.lower() == 'ovca':
dat = cptac.Ovarian()
elif opts.type.lower() == 'luad':
dat = cptac.Luad()
elif opts.type.lower() == 'endometrial':
dat = cptac.Endometrial()
else:
exit()
df = dat.get_phosphoproteomics()
pdf = dat.get_proteomics()
# df.columns = [' '.join(col).strip() for col in df.columns.values]
df.to_csv(path_or_buf="phos_file.tsv", sep='\t')
pdf.to_csv(path_or_buf='prot_file.tsv', sep='\t')
def getDataForCancer(ctype):
if ctype.lower() == 'brca':
dat = cptac.Brca()
elif ctype.lower() == 'ccrcc':
dat = cptac.Ccrcc()
elif ctype.lower() == 'coad':
dat = cptac.Colon()
elif ctype.lower() == 'ovca':
dat = cptac.Ovarian()
elif ctype.lower() == 'luad':
dat = cptac.Luad()
elif ctype.lower() == 'endometrial':
dat = cptac.Endometrial()
else:
exit()
return dat
def __init__(self):
cptac.download(dataset="endometrial", version='latest')
# cptac.download(dataset="brca", version='latest')
# cptac.download(dataset="gbm", version='latest')
# cptac.download(dataset="hsncc", version='latest')
# cptac.download(dataset="luad", version='latest')
cptac.download(dataset="ovarian", version='latest')
cptac.download(dataset="ccrcc", version='latest')
cptac.download(dataset="colon", version='latest')
self.en = cptac.Endometrial()
# self.brca = cptac.Brca()
# self.gbm = cptac.Gbm()
# self.hsncc = cptac.Hnscc()
# self.luad= cptac.Luad()
self.ovarian = cptac.Ovarian()
self.ccrcc = cptac.Ccrcc()
self.colon = cptac.Colon()
# self.datasets = list(self.en,self.brca,self.gbm,self.hsncc,self.luad,self.ovarian,self.ccrcc)
self.datasets = list([self.en, self.ovarian, self.ccrcc, self.colon])
def cptacData():
'''
We need to collect and load CPTAC data
'''
print("Loading cptac datasets")
#we need to make sure all datasets are downloaded
##here are the cancers that are available without login information
allcans = ['brca', 'ccrcc', 'colon', 'ovarian', 'luad',\
#'hnscc','gbm','lscc',\
'endometrial']
print("Downloading cptac data")
for ct in allcans:
cptac.download(dataset=ct)
#then we load them into a dictionary
fdict = {'brca':cptac.Brca(), 'ccrcc':cptac.Ccrcc(),\
'colon':cptac.Colon(), 'ovarian':cptac.Ovarian(),\
#'hnscc':cptac.Hnscc(),'gbm':cptac.Gbm(), 'lscc':cptac.Lscc(),\
'endometrial':cptac.Endometrial(), 'luad':cptac.Luad()}
return fdict
def test_get_frequently_mutated_ov_05_cutoff():
ov = cptac.Ovarian()
print('Running get_frequently_mutated...')
df = ut.get_frequently_mutated(ov, 0.05)
dimensions = (142, 4)
headers = ['Gene', 'Unique_Samples_Mut', 'Missense_Mut', 'Truncation_Mut']
# test genes names
test_coord_names = ((133, 0), (127, 0), (141, 0))
test_vals_names = ('WDFY4', 'TP53', 'ZNF865')
total_tumors = 83
#test missense and trucation not equal to unique_samples_mutated
#(miss and trunc in same sample)
test_coord_WDFY4 = ((133, 1), (133, 2), (133, 3))
test_vals_WDFY4 = (10 / total_tumors, 8 / total_tumors, 3 / total_tumors)
# test miss and trunc almost equal
test_coord_CDK12 = ((11, 1), (11, 2), (11, 3))
test_vals_CDK12 = (6 / total_tumors, 4 / total_tumors, 3 / total_tumors)
# test no truncation mutations
test_coord_ZNF865 = ((141, 1), (141, 2), (141, 3))
test_vals_ZNF865 = (5 / total_tumors, 5 / total_tumors, 0 / total_tumors)
# test close to cutoff
test_coord_SYNE1 = ((122, 1), (122, 2), (122, 3))
test_vals_SYNE1 = (5 / total_tumors, 5 / total_tumors, 1 / total_tumors)
# common test and highest count
test_coord_TP53 = ((127, 1), (127, 2), (127, 3))
test_vals_TP53 = (77 / total_tumors, 50 / total_tumors, 27 / total_tumors)
#CHECK silent mut not counted
test_coord_vals = [(test_coord_names, test_vals_names),
(test_coord_WDFY4, test_vals_WDFY4),
(test_coord_CDK12, test_vals_CDK12),
(test_coord_ZNF865, test_vals_ZNF865),
(test_coord_SYNE1, test_vals_SYNE1),
(test_coord_TP53, test_vals_TP53)]
for coord, vals in test_coord_vals:
PASS = check_getter(df, dimensions, headers, coord, vals)
print_test_result(PASS)
def test_get_frequently_mutated_ov_default_cutoff():
ov = cptac.Ovarian()
print('Running get_frequently_mutated...')
df = ut.get_frequently_mutated(ov)
dimensions = (16, 4)
headers = ['Gene', 'Unique_Samples_Mut', 'Missense_Mut', 'Truncation_Mut']
# test genes names
test_coord_names = ((15, 0), (13, 0), (2, 0))
test_vals_names = ('WDFY4', 'TP53', 'MT-CO1')
total_tumors = 83
#test missense and trucation not equal to unique_samples_mutated
#(miss and trunc in same sample)
test_coord_WDFY4 = ((15, 1), (15, 2), (15, 3))
test_vals_WDFY4 = (10 / total_tumors, 8 / total_tumors, 3 / total_tumors)
# test miss and trunc equal to unique_samples_mutated
test_coord_MUC4 = ((8, 1), (8, 2), (8, 3))
test_vals_MUC4 = (27 / total_tumors, 26 / total_tumors, 1 / total_tumors)
# test no truncation mutations
test_coord_MTCO1 = ((2, 1), (2, 2), (2, 3))
test_vals_MTCO1 = (10 / total_tumors, 10 / total_tumors, 0 / total_tumors)
# test close to cutoff
test_coord_FSIP2 = ((1, 1), (1, 2), (1, 3))
test_vals_FSIP2 = (9 / total_tumors, 8 / total_tumors, 2 / total_tumors)
# common test and highest count
test_coord_TP53 = ((13, 1), (13, 2), (13, 3))
test_vals_TP53 = (77 / total_tumors, 50 / total_tumors, 27 / total_tumors)
test_coord_vals = [(test_coord_names, test_vals_names),
(test_coord_WDFY4, test_vals_WDFY4),
(test_coord_MUC4, test_vals_MUC4),
(test_coord_MTCO1, test_vals_MTCO1),
(test_coord_FSIP2, test_vals_FSIP2),
(test_coord_TP53, test_vals_TP53)]
for coord, vals in test_coord_vals:
PASS = check_getter(df, dimensions, headers, coord, vals)
print_test_result(PASS)
def downloadCptac():
# To view available datasets, enter 'cptac.list_data()'.
cptac.list_datasets()
cptac.download(dataset = "endometrial")
cptac.download(dataset = 'colon')
cptac.download(dataset = 'ovarian')
cptac.download(dataset = 'RenalCcrcc')
#cptac.download(dataset ='luad')
#cptac.download(dataset ='brca')
downloadCptac()
endometrialData = cptac.Endometrial()
colorectalData = cptac.Colon()
ovarianData = cptac.Ovarian()
renalData = cptac.RenalCcrcc()
lungData = cptac.Luad()
breastData = cptac.Brca()
def listDataForEachCancer():
print("endometrial")
endometrialData.list_data()
print("\n\ncolorectal")
colorectalData.list_data()
print("\n\novarian")
ovarianData.list_data()
print("\n\nrenal")
renalData.list_data()
listDataForEachCancer()