def test_spectrum_none_exception(none_spectrum, filename):
""" Test for exception being thrown if the spectrum to be saved. """
with pytest.raises(AttributeError) as exception:
save_as_msp(none_spectrum, filename)
message = exception.value.args[0]
assert message == "'NoneType' object has no attribute 'metadata'"
def write(self, filename: str):
"""Write data to back to 'msp' file
Args:
filename (str): Path to filename under which to store the data.
"""
save_as_msp(self._spectra, filename)
def test_wrong_filename_exception():
""" Test for exception being thrown if output file doesn't end with .msp. """
with tempfile.TemporaryDirectory() as temp_dir:
filename = os.path.join(temp_dir, "test.mzml")
with pytest.raises(AssertionError) as exception:
save_as_msp(None, filename)
message = exception.value.args[0]
assert message == "File extension must be 'msp'."
def save_and_reload_spectra(filename, spectra: List[Spectrum]):
""" Utility function to save spectra to msp and load them again.
Params:
-------
spectra: Spectra objects to store
Returns:
--------
reloaded_spectra: Spectra loaded from saved msp file.
"""
save_as_msp(spectra, filename)
reloaded_spectra = list(load_from_msp(filename))
return reloaded_spectra
def test_num_peaks_last_metadata_field(filename, data):
""" Test to check whether the last line before the peaks is NUM PEAKS: ... """
save_as_msp(data, filename)
with open(filename, mode='r', encoding="utf-8") as file:
content = file.readlines()
for idx, line in enumerate(content):
if line.startswith('NUM PEAKS: '):
num_peaks = int(line.split()[2])
peaks = content[idx + 1:idx + num_peaks + 1]
for peak in peaks:
mz, intensity = peak.split()[:2]
mz = float(mz)
intensity = float(intensity)
assert isinstance(mz, float)
assert isinstance(intensity, float)
def test_file_exists_single_spectrum(spectrum, filename):
""" Test checking if the file is created. """
save_as_msp(spectrum, filename)
assert os.path.isfile(filename)
def main(argv):
parser = argparse.ArgumentParser(
description="Compute MSP similarity scores")
parser.add_argument("--spectra",
type=str,
required=True,
help="Mass spectra file to be filtered.")
parser.add_argument("--spectra_format",
type=str,
required=True,
help="Format of spectra file.")
parser.add_argument("--output",
type=str,
required=True,
help="Filtered mass spectra file.")
parser.add_argument(
"-normalise_intensities",
action='store_true',
help="Normalize intensities of peaks (and losses) to unit height.")
parser.add_argument(
"-default_filters",
action='store_true',
help=
"Collection of filters that are considered default and that do no require any (factory) arguments."
)
parser.add_argument(
"-clean_metadata",
action='store_true',
help=
"Apply all adding and cleaning filters if possible, so that the spectra have canonical metadata."
)
parser.add_argument(
"-relative_intensity",
action='store_true',
help=
"Keep only peaks within set relative intensity range (keep if to_intensity >= intensity >= from_intensity)."
)
parser.add_argument("--from_intensity",
type=float,
help="Lower bound for intensity filter")
parser.add_argument("--to_intensity",
type=float,
help="Upper bound for intensity filter")
parser.add_argument(
"-mz_range",
action='store_true',
help=
"Keep only peaks between set m/z range (keep if to_mz >= m/z >= from_mz)."
)
parser.add_argument("--from_mz",
type=float,
help="Lower bound for m/z filter")
parser.add_argument("--to_mz",
type=float,
help="Upper bound for m/z filter")
args = parser.parse_args()
if not (args.normalise_intensities or args.default_filters or
args.clean_metadata or args.relative_intensity or args.mz_range):
raise ValueError('No filter selected.')
if args.spectra_format == 'msp':
spectra = list(load_from_msp(args.spectra))
elif args.queries_format == 'mgf':
spectra = list(load_from_mgf(args.spectra))
else:
raise ValueError(
f'File format {args.spectra_format} not supported for mass spectra file.'
)
filtered_spectra = []
for spectrum in spectra:
if args.normalise_intensities:
spectrum = normalize_intensities(spectrum)
if args.default_filters:
spectrum = default_filters(spectrum)
if args.clean_metadata:
filters = [
add_compound_name, add_precursor_mz, add_fingerprint,
add_losses, add_parent_mass, add_retention_index,
add_retention_time, clean_compound_name
]
for metadata_filter in filters:
spectrum = metadata_filter(spectrum)
if args.relative_intensity:
spectrum = select_by_relative_intensity(spectrum,
args.from_intensity,
args.to_intensity)
if args.mz_range:
spectrum = select_by_mz(spectrum, args.from_mz, args.to_mz)
filtered_spectra.append(spectrum)
if args.spectra_format == 'msp':
save_as_msp(filtered_spectra, args.output)
else:
save_as_mgf(filtered_spectra, args.output)
return 0