def test_extract_per_individual_state_metadata_from_sample_metadata_and_biom(
self):
"""extract_per_individual_state_metadata_from_sample_metadata_and_biom functions as expected
"""
# single observations
o1_expected = {'o1': {'001': [22, 10],
'006': [25, 4],
'007': [33, 26],
'008': [99, 75]}}
actual = extract_per_individual_state_metadata_from_sample_metadata_and_biom(
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre", "Post"],
individual_identifier_category="PersonalID",
observation_ids=['o1'])
self.assertEqual(actual, o1_expected)
# all observations
all_expected = {'o1': {'001': [22, 10],
'006': [25, 4],
'007': [33, 26],
'008': [99, 75]},
'o2': {'001': [10, 44],
'006': [3, 99],
'007': [8, 18],
'008': [64, 164]},
'o3': {'001': [10, 50],
'006': [50, 10],
'007': [10, 50],
'008': [50, 10]}}
actual = extract_per_individual_state_metadata_from_sample_metadata_and_biom(
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre", "Post"],
individual_identifier_category="PersonalID",
observation_ids=None)
self.assertEqual(actual, all_expected)
# invalid observation id
self.assertRaises(
UnknownIDError,
extract_per_individual_state_metadata_from_sample_metadata_and_biom,
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre", "Post"],
individual_identifier_category="PersonalID",
observation_ids=['o1', 'bad.obs.id'])
def test_extract_per_individual_state_metadata_from_sample_metadata_and_biom(self):
"""extract_per_individual_state_metadata_from_sample_metadata_and_biom functions as expected
"""
# single observations
o1_expected = {'o1':{'001':[22,10],
'006':[25,4],
'007':[33,26],
'008':[99,75]}}
actual = extract_per_individual_state_metadata_from_sample_metadata_and_biom(
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre","Post"],
individual_identifier_category="PersonalID",
observation_ids=['o1'])
self.assertEqual(actual,o1_expected)
# all observations
all_expected = {'o1':{'001':[22,10],
'006':[25,4],
'007':[33,26],
'008':[99,75]},
'o2':{'001':[10,44],
'006':[3,99],
'007':[8,18],
'008':[64,164]},
'o3':{'001':[10,50],
'006':[50,10],
'007':[10,50],
'008':[50,10]}}
actual = extract_per_individual_state_metadata_from_sample_metadata_and_biom(
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre","Post"],
individual_identifier_category="PersonalID",
observation_ids=None)
self.assertEqual(actual,all_expected)
# invalid observation id
self.assertRaises(
UnknownID,
extract_per_individual_state_metadata_from_sample_metadata_and_biom,
self.individual_states_and_responses_map_f1,
self.paired_difference_biom1,
state_category="TreatmentState",
state_values=["Pre","Post"],
individual_identifier_category="PersonalID",
observation_ids=['o1','bad.obs.id'])
def main():
option_parser, opts, args =\
parse_command_line_parameters(**script_info)
mapping_fp = opts.mapping_fp
state_values = opts.state_values.split(',')
metadata_categories = opts.metadata_categories
state_category = opts.state_category
individual_id_category = opts.individual_id_category
output_dir = opts.output_dir
biom_table_fp = opts.biom_table_fp
observation_ids = opts.observation_ids
if not observation_ids is None:
observation_ids = observation_ids.split(',')
valid_states = opts.valid_states
ymin = opts.ymin
ymax = opts.ymax
line_color = opts.line_color
# validate the input - currently only supports either biom data
# or mapping file data. if useful in the future it shouldn't be too
# hard to allow the user to provide both.
if metadata_categories and biom_table_fp:
option_parser.error(
"Can only pass --metadata_categories or --biom_table_fp, not both.")
elif not (metadata_categories or biom_table_fp):
option_parser.error(
"Must pass either --metadata_categories or --biom_table_fp.")
else:
pass
# parse the mapping file to a dict
mapping_data = parse_mapping_file_to_dict(open(mapping_fp, 'U'))[0]
# currently only support for pre/post (ie, two-state) tests
if len(state_values) != 2:
option_parser.error(
"Exactly two state_values must be passed separated by a comma.")
# filter mapping_data, if requested
if valid_states:
sample_ids_to_keep = sample_ids_from_metadata_description(
open(mapping_fp, 'U'), valid_states)
for sid in mapping_data.keys():
if sid not in sample_ids_to_keep:
del mapping_data[sid]
if biom_table_fp:
biom_table = parse_biom_table(open(biom_table_fp, 'U'))
analysis_categories = observation_ids or biom_table.ObservationIds
personal_ids_to_state_values = \
extract_per_individual_state_metadata_from_sample_metadata_and_biom(
mapping_data,
biom_table,
state_category,
state_values,
individual_id_category,
observation_ids=analysis_categories)
else:
analysis_categories = metadata_categories.split(',')
personal_ids_to_state_values = \
extract_per_individual_state_metadata_from_sample_metadata(
mapping_data,
state_category,
state_values,
individual_id_category,
analysis_categories)
paired_difference_analyses(personal_ids_to_state_values,
analysis_categories,
state_values,
output_dir,
line_color=line_color,
ymin=ymin,
ymax=ymax)
def main():
option_parser, opts, args =\
parse_command_line_parameters(**script_info)
mapping_fp = opts.mapping_fp
state_values = opts.state_values.split(',')
metadata_categories = opts.metadata_categories
state_category = opts.state_category
individual_id_category = opts.individual_id_category
output_dir = opts.output_dir
biom_table_fp = opts.biom_table_fp
observation_ids = opts.observation_ids
if not observation_ids is None:
observation_ids = observation_ids.split(',')
valid_states = opts.valid_states
ymin = opts.ymin
ymax = opts.ymax
line_color = opts.line_color
# validate the input - currently only supports either biom data
# or mapping file data. if useful in the future it shouldn't be too
# hard to allow the user to provide both.
if metadata_categories and biom_table_fp:
option_parser.error(
"Can only pass --metadata_categories or --biom_table_fp, not both."
)
elif not (metadata_categories or biom_table_fp):
option_parser.error(
"Must pass either --metadata_categories or --biom_table_fp.")
else:
pass
# parse the mapping file to a dict
mapping_data = parse_mapping_file_to_dict(open(mapping_fp, 'U'))[0]
# currently only support for pre/post (ie, two-state) tests
if len(state_values) != 2:
option_parser.error(
"Exactly two state_values must be passed separated by a comma.")
# filter mapping_data, if requested
if valid_states:
sample_ids_to_keep = sample_ids_from_metadata_description(
open(mapping_fp, 'U'), valid_states)
for sid in mapping_data.keys():
if sid not in sample_ids_to_keep:
del mapping_data[sid]
if biom_table_fp:
biom_table = parse_biom_table(open(biom_table_fp, 'U'))
analysis_categories = observation_ids or biom_table.ObservationIds
personal_ids_to_state_values = \
extract_per_individual_state_metadata_from_sample_metadata_and_biom(
mapping_data,
biom_table,
state_category,
state_values,
individual_id_category,
observation_ids=analysis_categories)
else:
analysis_categories = metadata_categories.split(',')
personal_ids_to_state_values = \
extract_per_individual_state_metadata_from_sample_metadata(
mapping_data,
state_category,
state_values,
individual_id_category,
analysis_categories)
paired_difference_analyses(personal_ids_to_state_values,
analysis_categories,
state_values,
output_dir,
line_color=line_color,
ymin=ymin,
ymax=ymax)
