def write(self, data_df, version_line=True):
'''Write the given data to a GFF3 file, using the converter if given.
Generates an empty file if given an empty DataFrame.
Args:
version_line (bool): If True, write the GFF3 version line at the.
Note that this will cause an existing file to be overwritten, but
will only be added in the first call to `write`.
'''
if self.filename is None:
raise ValueError('Trying to write to filename None! Give GFF3Writer'
' a filename.')
if len(data_df) == 0:
warn_empty('Writing out an empty GFF3 file to {0}'.format(self.filename))
touch(self.filename)
return
if not self.created and version_line is True:
with open(self.filename, 'w') as fp:
fp.write(self.version_line + '\n')
with open(self.filename, 'a') as fp:
self.created = True
if self.converter is not None:
data_df = self.convert(data_df)
else:
self.mangle_coordinates(data_df)
data_df.to_csv(fp, sep='\t', na_rep='.', columns=[k for k, v in GFF3Parser.columns],
index=False, header=False, quoting=csv.QUOTE_NONE,
float_format='%.6e')
def cmd():
writer = GFF3Writer(output_filename,
converter=cmscan_to_gff3,
database=database)
try:
for group in InfernalParser(input_filename):
writer.write(group)
except EmptyFile as e:
touch(output_filename)
def cmd():
writer = GFF3Writer(output_filename,
converter=hmmscan_to_gff3,
database=database)
try:
for group in pd.read_csv(input_filename, chunksize=10000):
writer.write(group)
except EmptyFile as e:
touch(output_filename)
def cmd():
it = pd.read_csv(input_filename, chunksize=10000)
writer = GFF3Writer(output_filename,
converter=shmlast_to_gff3,
database=database)
try:
for group in it:
writer.write(group)
except EmptyFile:
touch(output_filename)
def cmd():
if input_filename.endswith('.csv') or input_filename.endswith('.tsv'):
it = pd.read_csv(input_filename, chunksize=10000)
else:
it = MafParser(input_filename)
writer = GFF3Writer(output_filename,
converter=maf_to_gff3,
database=database)
try:
for group in it:
writer.write(group)
except EmptyFile:
touch(output_filename)