def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
alpha_fps = opts.alpha_fps
mapping_fp = opts.mapping_fp
output_mapping_fp = opts.output_mapping_fp
binning_method = opts.binning_method
missing_value_name = opts.missing_value_name
depth = opts.depth
number_of_bins = opts.number_of_bins
collated_input = opts.collated_input
# if using collated data, make sure they specify a depth
if collated_input:
alpha_dict = {}
# build up a dictionary with the filenames as keys and lines as values
for single_alpha_fp in alpha_fps:
alpha_dict[splitext(basename(single_alpha_fp))[0]] = open(single_alpha_fp, "U").readlines()
# format the collated data
try:
metrics, alpha_sample_ids, alpha_data = mean_alpha(alpha_dict, depth)
except ValueError, e: # see mean_alpha for the possible exceptions
option_parser.error(e.message)
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
alpha_fps = opts.alpha_fps
mapping_fp = opts.mapping_fp
output_mapping_fp = opts.output_mapping_fp
binning_method = opts.binning_method
missing_value_name = opts.missing_value_name
depth = opts.depth
# make sure the number of bins is an integer
try:
number_of_bins = int(opts.number_of_bins)
except ValueError:
raise ValueError, 'The number of bins must be an integer, not %s'\
% opts.number_of_bins
# if using collated data, make sure they specify a depth
if depth is not None:
alpha_dict = {}
# build up a dictionary with the filenames as keys and lines as values
for single_alpha_fp in alpha_fps:
alpha_dict[splitext(basename(single_alpha_fp))[0]] = open(
single_alpha_fp, 'U').readlines()
# format the collated data
metrics, alpha_sample_ids, alpha_data = mean_alpha(alpha_dict,
depth)
# when not using collated data, the user can only specify one input file
else:
if len(alpha_fps) > 1:
option_parser.error('A comma-separated list of files should only be'
' passed with the --alpha_fps option when using collated alpha '
'diversity data and also selecting a rarefaction depth with the'
' --depth option.')
else:
metrics, alpha_sample_ids, alpha_data = parse_matrix(open(
alpha_fps[0], 'U'))
# parse the data from the files
mapping_file_data, mapping_file_headers, comments = parse_mapping_file(
open(mapping_fp, 'U'))
# add the alpha diversity data to the mapping file
out_mapping_file_data, out_mapping_file_headers = \
add_alpha_diversity_values_to_mapping_file(metrics, alpha_sample_ids,
alpha_data, mapping_file_headers, mapping_file_data, number_of_bins,
binning_method, missing_value_name)
# format the new data and write it down
lines = format_mapping_file(out_mapping_file_headers, out_mapping_file_data)
fd_out = open(output_mapping_fp, 'w')
fd_out.writelines(lines)
fd_out.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
alpha_fps = opts.alpha_fps
mapping_fp = opts.mapping_fp
output_mapping_fp = opts.output_mapping_fp
binning_method = opts.binning_method
missing_value_name = opts.missing_value_name
depth = opts.depth
number_of_bins = opts.number_of_bins
collated_input = opts.collated_input
# if using collated data, make sure they specify a depth
if collated_input:
alpha_dict = {}
# build up a dictionary with the filenames as keys and lines as values
for single_alpha_fp in alpha_fps:
alpha_dict[splitext(basename(single_alpha_fp))[0]] = open(
single_alpha_fp, 'U').readlines()
# format the collated data
try:
metrics, alpha_sample_ids, alpha_data = mean_alpha(
alpha_dict, depth)
except ValueError as e: # see mean_alpha for the possible exceptions
option_parser.error(e.message)
# when not using collated data, the user can only specify one input file
else:
if len(alpha_fps) > 1:
option_parser.error(
'A comma-separated list of files should only be'
' passed with the --alpha_fps option when using collated alpha '
'diversity data and also selecting a rarefaction depth with the'
' --depth option.')
else:
metrics, alpha_sample_ids, alpha_data = parse_matrix(
open(alpha_fps[0], 'U'))
# parse the data from the files
mapping_file_data, mapping_file_headers, comments = parse_mapping_file(
open(mapping_fp, 'U'))
# add the alpha diversity data to the mapping file
out_mapping_file_data, out_mapping_file_headers = \
add_alpha_diversity_values_to_mapping_file(metrics, alpha_sample_ids,
alpha_data, mapping_file_headers, mapping_file_data, number_of_bins,
binning_method, missing_value_name)
# format the new data and write it down
lines = format_mapping_file(out_mapping_file_headers,
out_mapping_file_data)
fd_out = open(output_mapping_fp, 'w')
fd_out.writelines(lines)
fd_out.close()
def test_mean_alpha(self):
"""checks data is being correctly averaged"""
# regular use-cases for this function
expected_data = [[9.441785, 82.93],
[0.42877, 5.2006], [9.625995, 8.18]]
expected_metrics = ['PD_whole_tree_even_310', 'chao1_even_310']
expected_sample_ids = ['s1', 's2', 's3']
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a, 310)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
expected_data = [[12.508435, 11.6105],
[0.42877, 8.42], [11.58785, 1.0]]
expected_metrics = ['PD_whole_tree_even_610', 'chao1_even_610']
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a, 610)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
# should default to the highest depth
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a,
None)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
# non-existant depth
with self.assertRaises(ValueError):
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_b, 111111)
# files with non-matching sample ids should raise an exception
with self.assertRaises(ValueError):
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_b, 310)
# input types that should not be processed
with self.assertRaises(AssertionError):
output = mean_alpha([1, 2, 3], 5)
with self.assertRaises(AssertionError):
output = mean_alpha({'a': 'b'}, -1.4)
def test_mean_alpha(self):
"""checks data is being correctly averaged"""
# regular use-cases for this function
expected_data = [[9.441785, 82.93],
[0.42877, 5.2006], [9.625995, 8.18]]
expected_metrics = ['PD_whole_tree_even_310', 'chao1_even_310']
expected_sample_ids = ['s1', 's2', 's3']
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a, 310)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
expected_data = [[12.508435, 11.6105],
[0.42877, 8.42], [11.58785, 1.0]]
expected_metrics = ['PD_whole_tree_even_610', 'chao1_even_610']
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a, 610)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
# should default to the highest depth
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_a,
None)
self.assertEquals(o_metrics, expected_metrics)
self.assertEquals(o_sample_ids, expected_sample_ids)
self.assertEquals(o_data, expected_data)
# non-existant depth
with self.assertRaises(ValueError):
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_b, 111111)
# files with non-matching sample ids should raise an exception
with self.assertRaises(ValueError):
o_metrics, o_sample_ids, o_data = mean_alpha(
self.collated_alpha_dict_b, 310)
# input types that should not be processed
with self.assertRaises(AssertionError):
output = mean_alpha([1, 2, 3], 5)
with self.assertRaises(AssertionError):
output = mean_alpha({'a': 'b'}, -1.4)
