def main(environment_file,
sample_mapping_file=None,
unit_mapping_file=None,
min_abundance=0,
min_sum_normalized_percent=1):
samples_dict = utils.get_samples_dict_from_environment_file(
environment_file)
oligos = utils.get_oligos_sorted_by_abundance(samples_dict,
min_abundance=min_abundance)
unit_counts, unit_percents = utils.get_unit_counts_and_percents(
oligos, samples_dict)
if sample_mapping_file:
sample_mapping = utils.get_sample_mapping_dict(sample_mapping_file)
if unit_mapping_file:
unit_mapping = utils.get_sample_mapping_dict(unit_mapping_file)
output_file = '.'.join(environment_file.split('.')[:-1]) + '.gexf'
utils.generate_gexf_network_file(
oligos,
samples_dict,
unit_percents,
output_file,
sample_mapping_dict=sample_mapping if sample_mapping_file else None,
unit_mapping_dict=unit_mapping if unit_mapping_file else None)
def main(
environment_file, sample_mapping_file=None, unit_mapping_file=None, min_abundance=0, min_sum_normalized_percent=1
):
samples_dict = utils.get_samples_dict_from_environment_file(environment_file)
oligos = utils.get_oligos_sorted_by_abundance(samples_dict, min_abundance=min_abundance)
unit_counts, unit_percents = utils.get_unit_counts_and_percents(oligos, samples_dict)
if sample_mapping_file:
sample_mapping = utils.get_sample_mapping_dict(sample_mapping_file)
if unit_mapping_file:
unit_mapping = utils.get_sample_mapping_dict(unit_mapping_file)
output_file = ".".join(environment_file.split(".")[:-1]) + ".gexf"
utils.generate_gexf_network_file(
oligos,
samples_dict,
unit_percents,
output_file,
sample_mapping_dict=sample_mapping if sample_mapping_file else None,
unit_mapping_dict=unit_mapping if unit_mapping_file else None,
)
def generate_exclusive_figures(_object):
import Oligotyping
scripts_dir_path = os.path.dirname(Oligotyping.__file__)
exclusive_figures_dict = {}
sample_mapping_dict = get_sample_mapping_dict(_object.sample_mapping)
for category in sample_mapping_dict:
exclusive_figures_dict[category] = {}
samples = sample_mapping_dict[category].keys()
# double filter: first makes sure sample was not removed from the analysis due to losing all its reads during the
# refinement, second makes sure that sample was actually mapped to something in the sample mapping file.
samples = filter(lambda s: sample_mapping_dict[category][s],
filter(lambda s: s in _object.samples, samples))
samples.sort()
mapping_file_path = get_temporary_file_name('%s-' % category,
'-mapping.txt',
_object.tmp_directory)
mapping_file = open(mapping_file_path, 'w')
mapping_file.write('samples\t%s\n' % (category))
for sample in samples:
mapping_file.write('%s\t%s\n' %
(sample, sample_mapping_dict[category][sample]))
mapping_file.close()
if samples == _object.samples:
matrix_percent_path = _object.matrix_percent_file_path
matrix_count_path = _object.matrix_count_file_path
else:
matrix_percent_path = get_temporary_file_name(
'%s-' % category, '-matrix-percent.txt', _object.tmp_directory)
matrix_count_path = get_temporary_file_name(
'%s-' % category, '-matrix-count.txt', _object.tmp_directory)
if store_filtered_matrix(_object.matrix_percent_file_path,
matrix_percent_path, samples) < 3:
_object.logger.info("skipping exclusive figs for '%s'; less than 3 samples were left in MP"\
% (category))
continue
if store_filtered_matrix(_object.matrix_count_file_path,
matrix_count_path, samples) < 3:
_object.logger.info("skipping exclusive figs for '%s'; less than 3 samples were left in MC"\
% (category))
continue
# ready to roll.
_object.logger.info("exclusive figs for '%s' with %d samples; mapping: '%s', MP: '%s', MC: '%s'"\
% (category, len(samples), mapping_file_path, matrix_percent_path, matrix_count_path))
for (analysis, script, output_dir) in [
('NMDS Analysis', '../Scripts/R/metaMDS-analysis-with-metadata.R',
'nmds_analysis')
]:
exclusive_figures_dict[category][output_dir] = {}
target_dir = _object.generate_output_destination(
'%s/%s/%s' % (os.path.basename(
_object.figures_directory), category, output_dir),
directory=True)
for (distance_metric,
matrix_file) in [("canberra", matrix_percent_path),
("kulczynski", matrix_percent_path),
("jaccard", matrix_percent_path),
("horn", matrix_percent_path),
("bray", matrix_percent_path)]:
output_prefix = os.path.join(target_dir, distance_metric)
cmd_line = (
'%s "%s" "%s" %s "%s" "%s" "%s" >> "%s" 2>&1' %
(os.path.join(scripts_dir_path, script), matrix_file,
mapping_file_path, distance_metric, category,
_object.project, output_prefix, _object.log_file_path))
_object.progress.update('%s "%s" for "%s" ...' %
(analysis, distance_metric, category))
_object.logger.info('exclusive figure: %s' % (cmd_line))
run_command(cmd_line)
exclusive_figures_dict[category][output_dir][
distance_metric] = output_prefix
# heatmap
for (analysis, script, output_dir) in [
('Heatmap Analysis', '../Scripts/R/heatmap.R', 'heatmap_analysis')
]:
exclusive_figures_dict[category][output_dir] = {}
target_dir = _object.generate_output_destination(
'%s/%s/%s' % (os.path.basename(
_object.figures_directory), category, output_dir),
directory=True)
for (distance_metric,
matrix_file) in [("canberra", matrix_percent_path),
("kulczynski", matrix_percent_path),
("jaccard", matrix_percent_path),
("horn", matrix_percent_path),
("bray", matrix_percent_path)]:
output_prefix = os.path.join(target_dir, distance_metric)
cmd_line = (
'%s "%s" -m "%s" -d %s --title "%s" -o "%s" >> "%s" 2>&1' %
(os.path.join(scripts_dir_path, script), matrix_file,
mapping_file_path, distance_metric, _object.project,
output_prefix, _object.log_file_path))
_object.progress.update('%s "%s" for "%s" ...' %
(analysis, distance_metric, category))
_object.logger.info('exclusive figure: %s' % (cmd_line))
run_command(cmd_line)
exclusive_figures_dict[category][output_dir][
distance_metric] = output_prefix
return exclusive_figures_dict
def generate_exclusive_figures(_object):
import Oligotyping
scripts_dir_path = os.path.dirname(Oligotyping.__file__)
exclusive_figures_dict = {}
sample_mapping_dict = get_sample_mapping_dict(_object.sample_mapping)
for category in sample_mapping_dict:
exclusive_figures_dict[category] = {}
samples = sample_mapping_dict[category].keys()
# double filter: first makes sure sample was not removed from the analysis due to losing all its reads during the
#  refinement, second makes sure that sample was actually mapped to something in the sample mapping file.
samples = filter(lambda s: sample_mapping_dict[category][s], filter(lambda s: s in _object.samples, samples))
samples.sort()
mapping_file_path = get_temporary_file_name("%s-" % category, "-mapping.txt", _object.tmp_directory)
mapping_file = open(mapping_file_path, "w")
mapping_file.write("samples\t%s\n" % (category))
for sample in samples:
mapping_file.write("%s\t%s\n" % (sample, sample_mapping_dict[category][sample]))
mapping_file.close()
if samples == _object.samples:
matrix_percent_path = _object.matrix_percent_file_path
matrix_count_path = _object.matrix_count_file_path
else:
matrix_percent_path = get_temporary_file_name(
"%s-" % category, "-matrix-percent.txt", _object.tmp_directory
)
matrix_count_path = get_temporary_file_name("%s-" % category, "-matrix-count.txt", _object.tmp_directory)
if store_filtered_matrix(_object.matrix_percent_file_path, matrix_percent_path, samples) < 3:
_object.logger.info(
"skipping exclusive figs for '%s'; less than 3 samples were left in MP" % (category)
)
continue
if store_filtered_matrix(_object.matrix_count_file_path, matrix_count_path, samples) < 3:
_object.logger.info(
"skipping exclusive figs for '%s'; less than 3 samples were left in MC" % (category)
)
continue
# ready to roll.
_object.logger.info(
"exclusive figs for '%s' with %d samples; mapping: '%s', MP: '%s', MC: '%s'"
% (category, len(samples), mapping_file_path, matrix_percent_path, matrix_count_path)
)
for (analysis, script, output_dir) in [
("NMDS Analysis", "../Scripts/R/metaMDS-analysis-with-metadata.R", "nmds_analysis")
]:
exclusive_figures_dict[category][output_dir] = {}
target_dir = _object.generate_output_destination(
"%s/%s/%s" % (os.path.basename(_object.figures_directory), category, output_dir), directory=True
)
for (distance_metric, matrix_file) in [
("canberra", matrix_percent_path),
("kulczynski", matrix_percent_path),
("jaccard", matrix_percent_path),
("horn", matrix_percent_path),
("bray", matrix_percent_path),
]:
output_prefix = os.path.join(target_dir, distance_metric)
cmd_line = '%s -o "%s" -d "%s" -m "%s" --title "%s" "%s" "%s" >> "%s" 2>&1' % (
os.path.join(scripts_dir_path, script),
output_prefix,
distance_metric,
category,
_object.project,
matrix_file,
mapping_file_path,
_object.log_file_path,
)
_object.progress.update('%s "%s" for "%s" ...' % (analysis, distance_metric, category))
_object.logger.info("exclusive figure: %s" % (cmd_line))
run_command(cmd_line)
exclusive_figures_dict[category][output_dir][distance_metric] = output_prefix
# heatmap
for (analysis, script, output_dir) in [("Heatmap Analysis", "../Scripts/R/heatmap.R", "heatmap_analysis")]:
exclusive_figures_dict[category][output_dir] = {}
target_dir = _object.generate_output_destination(
"%s/%s/%s" % (os.path.basename(_object.figures_directory), category, output_dir), directory=True
)
for (distance_metric, matrix_file) in [
("canberra", matrix_percent_path),
("kulczynski", matrix_percent_path),
("jaccard", matrix_percent_path),
("horn", matrix_percent_path),
("bray", matrix_percent_path),
]:
output_prefix = os.path.join(target_dir, distance_metric)
cmd_line = '%s "%s" -m "%s" -d %s --title "%s" -o "%s" >> "%s" 2>&1' % (
os.path.join(scripts_dir_path, script),
matrix_file,
mapping_file_path,
distance_metric,
_object.project,
output_prefix,
_object.log_file_path,
)
_object.progress.update('%s "%s" for "%s" ...' % (analysis, distance_metric, category))
_object.logger.info("exclusive figure: %s" % (cmd_line))
run_command(cmd_line)
exclusive_figures_dict[category][output_dir][distance_metric] = output_prefix
return exclusive_figures_dict