def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
biom_table_fp = opts.biom_table_fp
mapping_fp = opts.mapping_fp
fields = opts.fields.split(',')
output_dir = opts.output_dir
suppress_mf = opts.suppress_mapping_file_output
# column_rename_ids = opts.column_rename_ids
# include_repeat_cols = opts.include_repeat_cols
bt = load_table(biom_table_fp)
mdata, mheaders, mcomments = parse_mapping_file(mapping_fp)
mdata = array(mdata)
# check that biom file and mapping file have matching sample names. discard
# those samples that do not appear in both.
shared_samples = list(set(mdata[:, 0]).intersection(bt.ids(axis='sample')))
if len(shared_samples) == 0:
raise ValueError('Mapping file and biom table share no samples.')
elif len(shared_samples) == len(mdata[:, 0]):
mdata = array(mdata)
else:
# we want to preserve the order of the samples in the biom table
ss_bt_order = [s for s in bt.ids(axis='sample') if s in
shared_samples]
bt = bt.filter(ss_bt_order, axis='sample', inplace=True)
mdata = subset_mapping_data(mdata, shared_samples)
# check that headers in mapping data
if not all([i in mheaders for i in fields]):
raise ValueError('One or more of the specified fields was not found ' +\
'in the mapping file.')
# create output directory and create base names
create_dir(output_dir)
mf_base_name = join(output_dir, splitext(split(mapping_fp)[1])[0])
bt_base_name = join(output_dir, splitext(split(biom_table_fp)[1])[0])
# run code and append output
sample_groups, value_groups = make_non_empty_sample_lists(fields, mheaders,
mdata)
for sg, vg in zip(sample_groups, value_groups):
name_base = '__' + '%s_%s_' * len(vg) + '_'
name_tmp = []
for f, v in zip(fields, vg):
name_tmp.extend([f, v])
nb = name_base % tuple(name_tmp)
tmp_mf_data = subset_mapping_data(mdata, sg)
tmp_mf_str = format_mapping_file(mheaders, tmp_mf_data, mcomments)
write_biom_table(bt.filter(sg, axis='sample', inplace=False),
bt_base_name + nb + '.biom')
if not suppress_mf:
o = open(mf_base_name + nb + '.txt', 'w')
o.writelines(tmp_mf_str)
o.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
biom_table_fp = opts.biom_table_fp
mapping_fp = opts.mapping_fp
fields = opts.fields.split(',')
output_dir = opts.output_dir
suppress_mf = opts.suppress_mapping_file_output
# column_rename_ids = opts.column_rename_ids
# include_repeat_cols = opts.include_repeat_cols
bt = load_table(biom_table_fp)
mdata, mheaders, mcomments = parse_mapping_file(mapping_fp)
mdata = array(mdata)
# check that biom file and mapping file have matching sample names. discard
# those samples that do not appear in both.
shared_samples = list(set(mdata[:, 0]).intersection(bt.ids(axis='sample')))
if len(shared_samples) == 0:
raise ValueError('Mapping file and biom table share no samples.')
elif len(shared_samples) == len(mdata[:, 0]):
mdata = array(mdata)
else:
# we want to preserve the order of the samples in the biom table
ss_bt_order = [s for s in bt.ids(axis='sample') if s in
shared_samples]
bt = bt.filter(ss_bt_order, axis='sample', inplace=True)
mdata = subset_mapping_data(mdata, shared_samples)
# check that headers in mapping data
if not all([i in mheaders for i in fields]):
raise ValueError('One or more of the specified fields was not found ' +\
'in the mapping file.')
# create output directory and create base names
create_dir(output_dir)
mf_base_name = join(output_dir, splitext(split(mapping_fp)[1])[0])
bt_base_name = join(output_dir, splitext(split(biom_table_fp)[1])[0])
# run code and append output
sample_groups, value_groups = make_non_empty_sample_lists(fields, mheaders,
mdata)
for sg, vg in zip(sample_groups, value_groups):
name_base = '__' + '%s_%s_' * len(vg) + '_'
name_tmp = []
for f, v in zip(fields, vg):
name_tmp.extend([f, v])
nb = name_base % tuple(name_tmp)
tmp_mf_data = subset_mapping_data(mdata, sg)
tmp_mf_str = format_mapping_file(mheaders, tmp_mf_data, mcomments)
write_biom_table(bt.filter(sg, axis='sample', inplace=False),
bt_base_name + nb + '.biom')
if not suppress_mf:
o = open(mf_base_name + nb + '.txt', 'w')
o.writelines(tmp_mf_str)
o.close()
def test_subset_mapping_data(self):
"""Test that mapping data is subset correctly."""
samples_of_interest = ['s0', 's1', 's2']
obs = subset_mapping_data(self.mdata, samples_of_interest)
exp = array([['s0', 'blue', 'hot', '13'], ['s1', 'blue', 'cold', '1'],
['s2', 'green', 'cold', '12']],
dtype='|S5')
assert_array_equal(obs, exp)
samples_of_interest = ['s1', 's2',
's0'] #order change won't affect out
obs = subset_mapping_data(self.mdata, samples_of_interest)
assert_array_equal(obs, exp)
samples_of_interest = ['s4', 's0'] #order change won't affect out
obs = subset_mapping_data(self.mdata, samples_of_interest)
exp = array([['s0', 'blue', 'hot', '13'], ['s4', 'blue', '0', '0']],
dtype='|S5')
assert_array_equal(obs, exp)
def test_subset_mapping_data(self):
"""Test that mapping data is subset correctly."""
samples_of_interest = ['s0', 's1', 's2']
obs = subset_mapping_data(self.mdata, samples_of_interest)
exp = array([['s0', 'blue', 'hot', '13'],
['s1', 'blue', 'cold', '1'],
['s2', 'green', 'cold', '12']],
dtype='|S5')
assert_array_equal(obs, exp)
samples_of_interest = ['s1', 's2', 's0'] #order change won't affect out
obs = subset_mapping_data(self.mdata, samples_of_interest)
assert_array_equal(obs, exp)
samples_of_interest = ['s4', 's0'] #order change won't affect out
obs = subset_mapping_data(self.mdata, samples_of_interest)
exp = array([['s0', 'blue', 'hot', '13'],
['s4', 'blue', '0', '0']],
dtype='|S5')
assert_array_equal(obs, exp)
