def summarize_analysis(
*args,
**kwargs) -> Tuple[Optional[Sample], Counter[Id], Counter[Id], Scores]:
"""
Summarize for a cross-analysis (to be usually called in parallel!).
"""
# Recover input and parameters
analysis: str = args[0]
ontology: Ontology = kwargs['ontology']
# TODO: Delete the following comment lines in a future release
# including = ontology.including # See comment below for the reason
# excluding = ontology.excluding # in/excluding are not used anymore
counts: Dict[Sample, Counter[Id]] = kwargs['counts']
scores: Dict[Sample, Dict[Id, Score]] = kwargs['scores']
samples: List[Sample] = kwargs['samples']
output: io.StringIO = io.StringIO(newline='')
# Declare/define variables
summary_counts: Counter[Id] = col.Counter()
summary_acc: Counter[Id] = col.Counter()
summary_score: Scores = Scores({})
summary: Optional[Sample] = None
output.write(gray('Summary for ') + analysis + gray('... '))
target_samples: List[Sample] = [
smpl for smpl in samples if smpl.startswith(analysis)
]
assert len(target_samples) >= 1, \
red('ERROR! ') + analysis + gray(' has no samples to summarize!')
for smpl in target_samples:
summary_counts += counts[smpl]
summary_score.update(scores[smpl])
tree = TaxTree()
tree.grow(ontology=ontology, counts=summary_counts, scores=summary_score)
tree.subtract()
tree.shape()
summary_counts.clear()
summary_score.clear()
# Avoid including/excluding here as get_taxa is not as 'clever' as allin1
# and taxa are already included/excluded in the derived samples
tree.get_taxa(counts=summary_counts,
accs=summary_acc,
scores=summary_score)
summary_counts = +summary_counts # remove counts <= 0
if summary_counts: # Avoid returning empty sample (summary would be None)
summary = Sample(f'{analysis}_{STR_SUMMARY}')
output.write(
gray('(') + cyan(f'{len(target_samples)}') + gray(' samples)') +
green(' OK!\n'))
else:
output.write(yellow(' VOID\n'))
# Print output and return
print(output.getvalue(), end='')
sys.stdout.flush()
return summary, summary_counts, summary_acc, summary_score
def summarize_analysis(
*args, **kwargs) -> Tuple[Sample, Counter[Id], Counter[Id], Scores]:
"""
Summarize for a cross-analysis (to be usually called in parallel!).
"""
# Recover input and parameters
analysis: str = args[0]
ontology: Ontology = kwargs['ontology']
including = ontology.including
excluding = ontology.excluding
counts: Dict[Sample, Counter[Id]] = kwargs['counts']
scores: Dict[Sample, Dict[Id, Score]] = kwargs['scores']
samples: List[Sample] = kwargs['samples']
output: io.StringIO = io.StringIO(newline='')
# Declare/define variables
summary_counts: Counter[Id] = Counter()
summary_acc: Counter[Id] = Counter()
summary_score: Scores = Scores({})
summary: Sample = None
output.write(gray('Summary for ') + analysis + gray('... '))
target_samples: List[Sample] = [
smpl for smpl in samples if smpl.startswith(analysis)
]
assert len(target_samples) >= 1, \
red('ERROR! ') + analysis + gray(' has no samples to summarize!')
for smpl in target_samples:
summary_counts += counts[smpl]
summary_score.update(scores[smpl])
tree = TaxTree()
tree.grow(ontology=ontology, counts=summary_counts, scores=summary_score)
tree.subtract()
tree.shape()
summary_counts.clear()
summary_score.clear()
tree.get_taxa(counts=summary_counts,
accs=summary_acc,
scores=summary_score,
include=including,
exclude=excluding)
summary_counts = +summary_counts # remove counts <= 0
if summary_counts: # Avoid returning empty sample (summary would be None)
summary = Sample(f'{analysis}_{STR_SUMMARY}')
output.write(
gray('(') + cyan(f'{len(target_samples)}') + gray(' samples)') +
green(' OK!\n'))
else:
output.write(yellow(' VOID\n'))
# Print output and return
print(output.getvalue(), end='')
sys.stdout.flush()
return summary, summary_counts, summary_acc, summary_score