def parse_counts(biom_file, tree_file=None):
counts = []
table = biom.load_table(biom_file)
if tree_file:
tree = ParsedTree(
tree_file,
'newick',
taxon_name_re,
lineage_prefixes=['k__', 'p__', 'c__', 'o__', 'f__', 'g__', 's__'])
tree.set_index_clades(contains=['A', 'T', 'G', 'C'])
for obs_id, samp_id in table.nonzero():
lineage = parse_lineage(table, obs_id)
if not lineage and tree:
lineage = parse_lineage(table,
obs_id,
parsed_tree=tree,
index_only=True)
seq_var = parse_sequencing_variant(obs_id)
count = table.get_value_by_ids(obs_id, samp_id)
counts.append(
Count(samp_id=samp_id,
obs_id=obs_id,
count=count,
seq_var=seq_var,
lineage=lineage))
return counts
with session_scope() as session_2:
for exp_id, experiment in experiments.items():
for subject in experiment.subjects:
for sample in subject.samples:
for prep in sample.preparations:
for workflow in prep.workflows:
try:
workflow.count_dict
except AttributeError:
continue
for count in workflow.count_dict[sample.orig_sample_id]:
fact = Count(experiment=experiment,
subject=subject,
sample=sample,
sample_site=sample.sampling_site,
sample_time=sample.sampling_time,
preparation=prep,
workflow=workflow,
lineage=count.lineage,
seq_variant=count.seq_var,
count=count.count)
session_2.add(fact)
end = time.time()
print("Main loop took: ", end-start)
def parser(session):
"""Parse individual prep and BIOM files when parsing a study.
Note: This method will produce duplicates of sample, subject, and
processing data each time a BIOM file is inserted. It also unfortunately
duplicates count data (for each processing in the prep data file). This
