def merge_split_summaries(self):
merged_summary_index = {}
merged_summary_index_path = os.path.join(self.output_directory, 'SUMMARY.cp')
summary_dir = filesnpaths.gen_output_directory(os.path.join(self.output_directory, 'SUMMARY'), delete_if_exists = True)
# read all index files per run into a dict here, so the access is easier from within
# the for loop below
run_sum_indices = {}
for runinfo in self.input_runinfo_dicts.values():
run_sum_indices[runinfo['sample_id']] = dictio.read_serialized_object(runinfo['profile_summary_index'])
for i in range(0, len(self.split_names)):
self.progress.update('merging summaries for splits %s of %s' % (i + 1, len(self.split_names)))
split_name = self.split_names[i]
merged_summary = {}
for runinfo in self.input_runinfo_dicts.values():
run_split_summary = dictio.read_serialized_object(os.path.join(runinfo['input_dir'], run_sum_indices[runinfo['sample_id']][split_name]))
merged_summary[runinfo['sample_id']] = run_split_summary[runinfo['sample_id']]
merged_split_summary_path = os.path.join(summary_dir, os.path.basename(run_sum_indices[runinfo['sample_id']][split_name]))
dictio.write_serialized_object(merged_summary, merged_split_summary_path)
merged_summary_index[split_name] = merged_split_summary_path
self.progress.update('Serializing merged split summary index ...')
dictio.write_serialized_object(dictio.strip_prefix_from_dict_values(merged_summary_index, self.output_directory),\
merged_summary_index_path)
return summary_dir, merged_summary_index_path
def store_summarized_profile_for_each_split(self):
summary_index = {}
summary_index_output_path = self.generate_output_destination("SUMMARY.cp")
summary_dir = self.generate_output_destination("SUMMARY", directory=True)
self.progress.new("Storing summary files")
counter = 1
for contig in self.contigs:
self.progress.update("working on contig %s of %s" % (pp(counter), pp(len(self.contigs))))
for split in self.contigs[contig].splits:
split_summary_path = self.generate_output_destination(os.path.join(summary_dir, "%.6d.cp" % counter))
dictio.write_serialized_object(
{
self.sample_id: {
"coverage": split.coverage.c,
"variability": split.auxiliary.v,
"competing_nucleotides": split.auxiliary.competing_nucleotides,
}
},
split_summary_path,
)
summary_index[split.name] = split_summary_path
counter += 1
self.progress.end()
self.run.info("profile_summary_dir", summary_dir)
dictio.write_serialized_object(
dictio.strip_prefix_from_dict_values(summary_index, self.output_directory), summary_index_output_path
)
self.run.info("profile_summary_index", summary_index_output_path)
def store_info_dict(self, destination, strip_prefix=None):
if strip_prefix:
# mostly to get rid of output_dir prefix in output file names.
# surprisingly enough, this is the best place to do it. live
# and learn :/
self.info_dict = dictio.strip_prefix_from_dict_values(self.info_dict, strip_prefix)
dictio.write_serialized_object(self.info_dict, destination)
def store_info_dict(self, destination, strip_prefix = None):
if strip_prefix:
# mostly to get rid of output_dir prefix in output file names.
# surprisingly enough, this is the best place to do it. live
# and learn :/
self.info_dict = dictio.strip_prefix_from_dict_values(self.info_dict, strip_prefix)
dictio.write_serialized_object(self.info_dict, destination)
