def test_cmd_all_ok_long_options(self, cmd_all_ok_long_options) -> None:
# Function tests how `parse_arguments` parses correct long options
# Backup argv
buff_argv: List[str] = sys.argv
# Set test argv
sys.argv = cmd_all_ok_long_options
# Parse arguments
params: args.HighlighterParams = args.parse_arguments()
# Restore argv
argv = buff_argv
assert params.target_fasta_fpath == cmd_all_ok_long_options[2]
assert params.bam_fpath == cmd_all_ok_long_options[4]
assert params.outfpath == cmd_all_ok_long_options[6]
assert params.suppress_zero_cov_output == True
assert params.topology == 'circular'
assert params.organism == cmd_all_ok_long_options[12]
obtained_threshold_repr: str = ','.join(
map(lambda covthr: str(covthr.get_coverage()),
params.coverage_thresholds))
expected_threshold_repr: str = cmd_all_ok_long_options[8]
assert obtained_threshold_repr == expected_threshold_repr
def test_cmd_bam_missing(self, cmd_bam_missing) -> None:
# Function tests how `parse_arguments` parses command line
# with positional arguments
# Backup argv
buff_argv: List[str] = sys.argv
# Set test argv
sys.argv = cmd_bam_missing
# Parse arguments
with pytest.raises(SystemExit):
params: args.HighlighterParams = args.parse_arguments()
# end with
# Restore argv
argv = buff_argv
def main():
# Get the arguments for the current execution.
args = parse_arguments()
print(args)
# Read the training manifest tsv file.
df = pd.read_csv(args.manifest, sep='\t')
# Define the scoring function and apply it to the manifest.
scoring_function = ScoringFunction().create(args)
df = scoring_function(df)
# Create the directory for output manifest storage.
os.makedirs(args.out_dir, exist_ok=True)
# Save the sorted manifest.
df.to_csv(os.path.join(args.out_dir, 'sorted_manifest.tsv'),
sep='\t',
index=None)
def main():
# Get the arguments for the current execution.
args = parse_arguments()
print(args)
# Read the training manifest tsv file.
df = pd.read_csv(args.manifest, sep='\t')
# Define the pacing function and apply it to the sorted manifest.
pacing_function = PacingFunction().create(args)
df_list = pacing_function(df)
# Create the directory for output smaller datasets storage.
if not os.path.isdir(args.out_dir):
os.mkdir(args.out_dir)
df_paths_list = []
for index, sub_df in enumerate(df_list):
path = os.path.join(args.out_dir, f'train_{index}.tsv')
sub_df.to_csv(path, sep='\t', index=None)
df_paths_list.append(path)
if args.fairseq:
train_fairseq(args, df_paths_list)
def main(version: str, last_update_date: str) -> None:
# Parse arguments
params: HighlighterParams = parse_arguments()
# This string will be used for annotation of result GenBank file
base_feature_note: str = f'generated by consensus-highlighter v{version}'
# String for storing info about warnings
with_warnings: str = ''
# Read fasta records from input file
print('Importing fasta from `{}`...'.format(params.target_fasta_fpath),
end=' ')
sys.stdout.flush()
fasta_records: Sequence[SeqRecord] = pfr.parse_fasta_reference(
params.target_fasta_fpath)
print('done')
# Create ouput directory
_create_outdir_from_outfile(params.outfpath)
out.create_or_emply_file(params.outfpath)
# Obtain path to coverage file
coverage_fpath: str = out.conf_path_to_depth_file(params.outfpath)
# Count coverages with samtools depth
print('Silently counting coverages with `samtools depth`...', end=' ')
sys.stdout.flush()
cov_fpath: str = oc.count_cov_for_all_refs(params.target_fasta_fpath,
params.bam_fpath,
coverage_fpath)
print('done\n')
# Proceed with annotation
rec: SeqRecord
for rec in fasta_records:
print(f'Processing sequence `{rec.description}`')
# Obtain coverages for current sequence
cov_array: CoverageArray = oc.get_coverage_for_reference(
rec.id, cov_fpath)
# Check length of the coverage array
if len(cov_array) == 0:
print(
f'! Warning: no coverage information found for sequence `{rec.id}`.'
)
print(
f"""! Please, make sure that field `RNAME` (3-rd column) in your BAM file contains
! id of this sequence specified in fasta header (i.e. `{rec.id}`).""")
print('! Omitting this sequence.')
print('=' * 10)
with_warnings = ' with warnings'
continue
# end if
if len(cov_array) != len(rec.seq):
print(
f"""! Warning: length of sequence `{rec.id}` ({len(rec.seq)} bp)
! is not equal to number of coverage positions ({len(cov_array)}) reported by `samtools depth`
! and stored in coverage file `{cov_fpath}`.""")
print(
'! Re-creating the bam file might be the solution of this issue.'
)
print('! Omitting this sequence.')
print('=' * 10)
with_warnings = ' with warnings'
continue
# end if
mean_coverage = round(sts.mean(cov_array.coverages), 2)
print(f'Average coverage: {mean_coverage}')
cov_threshold: CoverageThreshold
coverage_features: MutableSequence[SeqFeature]
# Detect all necessary coverage features
for cov_threshold in params.coverage_thresholds:
print(
f'Screening the sequence for regions with {cov_threshold.get_label()}...',
end=' ')
sys.stdout.flush()
# Get coverage features
coverage_features = hlft.highlight_coverage_features(
cov_array, cov_threshold, base_feature_note)
coverage_features = ddf.dedupl_features(coverage_features,
rec.features)
# Append features to list
rec.features.extend(coverage_features)
print('done')
# end for
print(f'Writing annotated sequence to `{params.outfpath}`...', end=' ')
sys.stdout.flush()
# Write result GanBank record
out.write_genbank_output(rec, params.topology, params.organism,
params.outfpath)
print('done')
print('=' * 10)
# end for
print(f'Completed{with_warnings}!')