def import_file(name, time_row=0, censor_row=1, features_rows=None):
cohort = np.genfromtxt(name, delimiter=',', dtype=None, filling_values='', comments="!")
survival_time, survival_censor = get_time_and_censor(cohort, time_row, censor_row)
row_headers = list(cohort[:,0])
if 'patient' in row_headers:
patient_row_idx = row_headers.index('patient')
elif 'ID_REF' in row_headers:
patient_row_idx = row_headers.index('ID_REF')
else:
print 'Error: \'patient\' or \'ID_REF\' header not found'
help_message.usage()
# Note: these return *all* the metadata rows if feature_rows is None,
# or the selected ones if feature_rows is set.
features, feature_names = get_features(features_rows, patient_row_idx, cohort)
all_metadata_row_names = get_feature_row_names(patient_row_idx, cohort)
gene_names = row_headers[patient_row_idx+1:]
patient_value = cohort[patient_row_idx+1:, 1:]
patient_value = clear_blanks(patient_value)
patient_value = patient_value.astype(np.float)
input_data = {
'patient_values': patient_value,
'survival_time': survival_time,
'survival_censor': survival_censor,
'time_row_num': time_row,
'censor_row_num': censor_row,
'metadata_row_names': all_metadata_row_names,
'gene_names': gene_names,
'metadata_feature_names': feature_names,
'metadata_features': features,
}
return input_data
def get_row_titles(name):
with open(name, 'rU') as f:
row_titles = []
for i,line in enumerate(f.readlines()):
row_title = line.split(',')[0]
if row_title[0] == '!':
continue
if row_title == 'patient' or row_title == 'ID_REF':
break
elif i > 50:
print "'patient' nor 'ID_REF' found. Aborting."
help_message.usage()
else:
row_titles.append(row_title)
return row_titles
def get_options(argv):
try:
opts, args = getopt.getopt(argv[1:], 'ho:i:vm:',
['help', 'input=', 'output-directory=', 'multivariates=', 'interactive'])
except getopt.error, msg:
help_message.usage()
def get_options(argv):
try:
opts, args = getopt.getopt(argv[1:], 'ho:i:vm:',
['help', 'input=', 'output-directory=', 'multivariates=', 'interactive'])
except getopt.error, msg:
help_message.usage()
infile = None
outdir = '.'
multivariates = []
for option, value in opts:
if option == '-v':
verbose = True
if option in ('-h', '--help'):
help_message.usage()
if option in ('-i', '--input'):
infile = value
if option in ('-o', '--output-directory'):
outdir = value
#TODO(joans): process this by looking up row numbers so that
# all the weird complexity of metadata_features sometimes having
# all the features can go away
if option in ('-m', '--mutlivariates'):
multivariates = value.split(',')
if not infile:
help_message.usage()
interactive = ('--interactive', '') in opts
return infile, outdir, multivariates, interactive
def get_row_number_from_title(title, row_titles):
