def merge_rearrangement(out_filename, in_filenames, drop=False, debug=False):
"""
Merge one or more AIRR rearrangements files
Arguments:
out_filename (str): output file path.
in_filenames (list): list of input files to merge.
drop (bool): drop flag. If True then drop fields that do not exist in all input
files, otherwise combine fields from all input files.
debug (bool): debug flag. If True print debugging information to standard error.
Returns:
bool: True if files were successfully merged, otherwise False.
"""
try:
# gather fields from input files
readers = [
RearrangementReader(open(f, 'r'), debug=debug)
for f in in_filenames
]
field_list = [x.fields for x in readers]
if drop:
field_set = set.intersection(*map(set, field_list))
else:
field_set = set.union(*map(set, field_list))
field_order = OrderedDict([(f, None) for f in chain(*field_list)])
out_fields = [f for f in field_order if f in field_set]
# write input files to output file sequentially
with open(out_filename, 'w+') as handle:
writer = RearrangementWriter(handle,
fields=out_fields,
debug=debug)
for reader in readers:
for r in reader:
writer.write(r)
reader.close()
except:
sys.stderr.write(
'Error occurred while merging AIRR rearrangement files.\n')
return False
return True
def dump_rearrangement(dataframe, filename, debug=False):
"""
Write the contents of a data frame to an AIRR rearrangements file
Arguments:
dataframe (pandas.DataFrame): data frame of rearrangement data.
filename (str): output file path.
debug (bool): debug flag. If True print debugging information to standard error.
Returns:
bool: True if the file is written without error.
"""
# TODO: test pandas.DataFrame.to_csv with converters argument as an alterative
# dataframe.to_csv(handle, sep='\t', header=True, index=False, encoding='utf-8')
fields = dataframe.columns.tolist()
with open(filename, 'w+') as handle:
writer = RearrangementWriter(handle, fields=fields, debug=debug)
for __, row in dataframe.iterrows():
writer.write(row.to_dict())
return True