def split_tables(sdrf_path):
sdrf_df = isatab.read_tfile(sdrf_path)
sdrf_df_isatab_header = sdrf_df.isatab_header
sample_name_index = list(sdrf_df.columns).index("Sample Name")
study_df = sdrf_df[sdrf_df.columns[0:sample_name_index+1]].drop_duplicates()
study_df.isatab_header = sdrf_df_isatab_header[0:sample_name_index+1]
assay_df = sdrf_df[sdrf_df.columns[sample_name_index:]]
assay_df.isatab_header = sdrf_df_isatab_header[sample_name_index:]
return study_df, assay_df
def replace_factor_with_protocol_parameter_value(self, factor_name,
protocol_ref):
"""Fixes a factor if it's supposed to be a Parameter Value
:param factor_name: The factor that's incorrect
:param protocol_ref: Protocol REF for the new Parameter Value
:return: None
"""
table_file_df = isatab.read_tfile(self.path)
field_names = list(table_file_df.columns)
clean_field_names = self.clean_isatab_field_names(field_names)
factor_index = clean_field_names.index(
'Factor Value[{factor_name}]'.format(factor_name=factor_name))
with open(self.path) as tfile_fp:
next(tfile_fp)
line1 = next(tfile_fp)
protocol_ref_index = list(
map(lambda x: x[1:-1] if x[0] == '"' and x[-1] == '"' else x,
line1.split('\t'))).index(protocol_ref)
if protocol_ref_index < 0:
raise IOError(
'Could not find protocol ref matching {protocol_ref}'.format(
protocol_ref=protocol_ref))
if factor_index < len(field_names) and \
'Term Source REF' in field_names[factor_index + 1] and \
'Term Accession' in field_names[factor_index + 2]:
log.debug('Moving Factor Value[{}] with term columns'.format(
factor_name))
# move Factor Value and Term Source REF and Term Accession columns
field_names.insert(protocol_ref_index + 1,
field_names[factor_index])
field_names.insert(protocol_ref_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(protocol_ref_index + 3,
field_names[factor_index + 2 + 2])
del field_names[factor_index + 3] # del Factor Value[{}]
del field_names[factor_index + 1 + 2] # del Term Source REF
del field_names[factor_index + 2 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1] and \
'Term Source REF' in field_names[factor_index + 2] and \
'Term Accession' in field_names[factor_index + 3]:
log.debug(
'Moving Factor Value[{factor_name}] with unit term columns'.
format(factor_name=factor_name))
# move Factor Value and Unit as ontology annotation
field_names.insert(protocol_ref_index + 1,
field_names[factor_index])
field_names.insert(protocol_ref_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(protocol_ref_index + 3,
field_names[factor_index + 2 + 2])
field_names.insert(protocol_ref_index + 4,
field_names[factor_index + 3 + 3])
del field_names[factor_index + 4] # del Factor Value[{}]
del field_names[factor_index + 1 + 3] # del Unit
del field_names[factor_index + 2 + 2] # del Term Source REF
del field_names[factor_index + 3 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1]:
log.debug(
'Moving Factor Value[{factor_name}] with unit column'.format(
factor_name=factor_name))
# move Factor Value and Unit columns
field_names.insert(protocol_ref_index + 1,
field_names[factor_index])
field_names.insert(protocol_ref_index + 2,
field_names[factor_index + 1 + 1])
del field_names[factor_index + 2] # del Factor Value[{}]
del field_names[factor_index + 1 + 1] # del Unit
else: # move only the Factor Value column
log.debug('Moving Factor Value[{factor_name}]'.format(
factor_name=factor_name))
field_names.insert(protocol_ref_index + 1,
field_names[factor_index])
del field_names[factor_index] # del Factor Value[{}]
table_file_df.columns = self.clean_isatab_field_names(field_names)
# Rename Factor Value column to Parameter Value column
field_names_modified = list(table_file_df.columns)
field_names_modified[protocol_ref_index + 1] = \
field_names_modified[protocol_ref_index + 1].replace(
'Factor Value', 'Parameter Value')
table_file_df.columns = self.clean_isatab_field_names(
field_names_modified)
investigation = isatab.load(os.path.dirname(self.path),
skip_load_tables=True)
study = investigation.studies[-1]
protocol = study.get_prot(protocol_ref)
if protocol is None:
raise ISAModelAttributeError(
'No protocol with name {protocol_ref} was found'.format(
protocol_ref=protocol_ref))
protocol.add_param(factor_name)
factor = study.get_factor(factor_name)
if factor is None:
raise ISAModelAttributeError(
'No factor with name {factor_name} was found'.format(
factor_name=factor_name))
else:
study.del_factor(name=factor_name, are_you_sure=True)
study.filename = '{study_filename}.fix'.format(
study_filename=study.filename)
isatab.dump(investigation,
output_path=os.path.dirname(self.path),
i_file_name='i_Investigation.txt.fix',
skip_dump_tables=True)
with open(
os.path.join(
os.path.dirname(self.path), '{s_filename}.fix'.format(
s_filename=os.path.basename(self.path))),
'w') as out_fp:
table_file_df.to_csv(path_or_buf=out_fp,
index=False,
sep='\t',
encoding='utf-8')
def replace_factor_with_source_characteristic(self, factor_name):
"""Fixes a factor if it's supposed to be a source characteristic
attached
:param factor_name: The factor that's incorrect
:return: None
"""
table_file_df = isatab.read_tfile(self.path)
field_names = list(table_file_df.columns)
clean_field_names = self.clean_isatab_field_names(field_names)
factor_index = clean_field_names.index(
'Factor Value[{}]'.format(factor_name))
source_name_index = clean_field_names.index('Source Name')
if factor_index < len(field_names) and \
'Term Source REF' in field_names[factor_index + 1] and \
'Term Accession' in field_names[factor_index + 2]:
log.debug('Moving Factor Value[{}] with term columns'.format(
factor_name))
# move Factor Value and Term Source REF and Term Accession columns
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(source_name_index + 3,
field_names[factor_index + 2 + 2])
del field_names[factor_index + 3] # del Factor Value[{}]
del field_names[factor_index + 1 + 2] # del Term Source REF
del field_names[factor_index + 2 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1] and \
'Term Source REF' in field_names[factor_index + 2] and \
'Term Accession' in field_names[factor_index + 3]:
log.debug('Moving Factor Value[{}] with unit term columns'.format(
factor_name))
# move Factor Value and Unit as ontology annotation
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(source_name_index + 3,
field_names[factor_index + 2 + 2])
field_names.insert(source_name_index + 4,
field_names[factor_index + 3 + 3])
del field_names[factor_index + 4] # del Factor Value[{}]
del field_names[factor_index + 1 + 3] # del Unit
del field_names[factor_index + 2 + 2] # del Term Source REF
del field_names[factor_index + 3 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1]:
log.debug(
'Moving Factor Value[{}] with unit column'.format(factor_name))
# move Factor Value and Unit columns
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
del field_names[factor_index + 2] # del Factor Value[{}]
del field_names[factor_index + 1 + 1] # del Unit
else: # move only the Factor Value column
log.debug('Moving Factor Value[{}]'.format(factor_name))
field_names.insert(source_name_index + 1,
field_names[factor_index])
del field_names[factor_index] # del Factor Value[{}]
table_file_df.columns = self.clean_isatab_field_names(field_names)
# Rename Factor Value column to Characteristics column
field_names_modified = list(table_file_df.columns)
field_names_modified[source_name_index + 1] = \
field_names_modified[source_name_index + 1].replace(
'Factor Value', 'Characteristics')
table_file_df.columns = self.clean_isatab_field_names(
field_names_modified)
with open(self.path, 'w') as out_fp:
table_file_df.to_csv(path_or_buf=out_fp,
index=False,
sep='\t',
encoding='utf-8')
def load_df(path):
df = ISATAB.read_tfile(path)
df.replace(to_replace = '', value = numpy.nan, inplace = True)
return df
def get_study_df(input_dir, study):
return ISATAB.read_tfile(os.path.join(input_dir, study.filename))
def get_measures_df(input_dir, assay):
data_filename = get_data_file(assay)
array = ISATAB.read_tfile(os.path.join(input_dir, data_filename))
return array
def get_assay_df(input_dir, assay):
return ISATAB.read_tfile(os.path.join(input_dir, assay.filename))
def replace_factor_with_source_characteristic(self, factor_name):
table_file_df = isatab.read_tfile(self.path)
field_names = list(table_file_df.columns)
clean_field_names = self.clean_isatab_field_names(field_names)
factor_index = clean_field_names.index(
'Factor Value[{}]'.format(factor_name))
source_name_index = clean_field_names.index('Source Name')
if factor_index < len(field_names) and \
'Term Source REF' in field_names[factor_index + 1] and \
'Term Accession' in field_names[factor_index + 2]:
log.debug('Moving Factor Value[{}] with term columns'.format(
factor_name))
# move Factor Value and Term Source REF and Term Accession columns
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(source_name_index + 3,
field_names[factor_index + 2 + 2])
del field_names[factor_index + 3] # del Factor Value[{}]
del field_names[factor_index + 1 + 2] # del Term Source REF
del field_names[factor_index + 2 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1] and \
'Term Source REF' in field_names[factor_index + 2] and \
'Term Accession' in field_names[factor_index + 3]:
log.debug('Moving Factor Value[{}] with unit term columns'.format(
factor_name))
# move Factor Value and Unit as ontology annotation
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
field_names.insert(source_name_index + 3,
field_names[factor_index + 2 + 2])
field_names.insert(source_name_index + 4,
field_names[factor_index + 3 + 3])
del field_names[factor_index + 4] # del Factor Value[{}]
del field_names[factor_index + 1 + 3] # del Unit
del field_names[factor_index + 2 + 2] # del Term Source REF
del field_names[factor_index + 3 + 1] # del Term Accession
elif factor_index < len(field_names) and \
'Unit' in field_names[factor_index + 1]:
log.debug(
'Moving Factor Value[{}] with unit column'.format(factor_name))
# move Factor Value and Unit columns
field_names.insert(source_name_index + 1,
field_names[factor_index])
field_names.insert(source_name_index + 2,
field_names[factor_index + 1 + 1])
del field_names[factor_index + 2] # del Factor Value[{}]
del field_names[factor_index + 1 + 1] # del Unit
else: # move only the Factor Value column
log.debug('Moving Factor Value[{}]'.format(factor_name))
field_names.insert(source_name_index + 1,
field_names[factor_index])
del field_names[factor_index] # del Factor Value[{}]
table_file_df.columns = self.clean_isatab_field_names(field_names)
with open(self.path, 'w') as out_fp:
table_file_df.to_csv(path_or_buf=out_fp,
index=False,
sep='\t',
encoding='utf-8')