def test_save_as_mgf_single_spectrum():
"""Test saving spectrum to .mgf file"""
spectrum = SpectrumBuilder().with_mz(
numpy.array([100, 200, 300], dtype="float")).with_intensities(
numpy.array([10, 10, 500], dtype="float")).with_metadata(
{
"charge": -1,
"inchi": '"InChI=1S/C6H12"',
"pepmass": (100, 10.0),
"test_field": "test"
},
metadata_harmonization=False).build()
# Write to test file
with tempfile.TemporaryDirectory() as d:
filename = os.path.join(d, "test.mgf")
save_as_mgf(spectrum, filename)
# test if file exists
assert os.path.isfile(filename)
# Test if content of mgf file is correct
with open(filename, "r", encoding="utf-8") as f:
mgf_content = f.readlines()
assert mgf_content[0] == "BEGIN IONS\n"
assert mgf_content[2] == "CHARGE=1-\n"
assert mgf_content[4] == "TEST_FIELD=test\n"
assert mgf_content[7].split(" ")[0] == "300.0"
def test_save_load_mgf_consistency(tmpdir, charge, ionmode, parent_mass):
"""Test saving and loading spectrum to .mgf file"""
mz = numpy.array([100.1, 200.02, 300.003], dtype="float")
intensities = numpy.array([0.01, 0.02, 1.0], dtype="float")
metadata = {
"precursor_mz": 200.5,
"charge": charge,
"ionmode": ionmode,
"parent_mass": parent_mass
}
builder = SpectrumBuilder().with_mz(mz).with_intensities(intensities)
spectrum1 = builder.with_metadata(metadata,
metadata_harmonization=True).build()
spectrum2 = builder.with_metadata(metadata,
metadata_harmonization=True).build()
# Write to test file
filename = os.path.join(tmpdir, "test.mgf")
save_as_mgf([spectrum1, spectrum2], filename)
# Test if file exists
assert os.path.isfile(filename)
# Test importing spectra again
spectrum_imports = list(load_from_mgf(filename))
assert spectrum_imports[0].get("precursor_mz") == 200.5
assert spectrum_imports[0].get("charge") == charge
assert spectrum_imports[0].get("ionmode") == ionmode
assert spectrum_imports[0].get("parent_mass") == str(parent_mass)
def test_save_as_mgf_spectrum_list():
"""Test saving spectrum list to .mgf file"""
mz = numpy.array([100, 200, 300], dtype="float")
intensities = numpy.array([10, 10, 500], dtype="float")
builder = SpectrumBuilder().with_mz(mz).with_intensities(intensities)
spectrum1 = builder.with_metadata({
"test_field": "test1"
},
metadata_harmonization=False).build()
spectrum2 = builder.with_metadata({
"test_field": "test2"
},
metadata_harmonization=False).build()
# Write to test file
with tempfile.TemporaryDirectory() as d:
filename = os.path.join(d, "test.mgf")
save_as_mgf([spectrum1, spectrum2], filename)
# test if file exists
assert os.path.isfile(filename)
# Test if content of mgf file is correct
with open(filename, "r", encoding="utf-8") as f:
mgf_content = f.readlines()
assert mgf_content[5] == mgf_content[12] == "END IONS\n"
assert mgf_content[1].split("=")[1] == "test1\n"
assert mgf_content[8].split("=")[1] == "test2\n"
def test_save_as_mgf_spectrum_list():
"""Test saving spectrum list to .mgf file"""
spectrum1 = Spectrum(mz=numpy.array([100, 200, 300], dtype="float"),
intensities=numpy.array([10, 10, 500], dtype="float"),
metadata={"test_field": "test1"})
spectrum2 = Spectrum(mz=numpy.array([100, 200, 300], dtype="float"),
intensities=numpy.array([10, 10, 500], dtype="float"),
metadata={"test_field": "test2"})
# Write to test file
with tempfile.TemporaryDirectory() as d:
filename = os.path.join(d, "test.mgf")
save_as_mgf([spectrum1, spectrum2], filename)
# test if file exists
assert os.path.isfile(filename)
# Test if content of mgf file is correct
with open(filename, "r") as f:
mgf_content = f.readlines()
assert mgf_content[5] == mgf_content[12] == "END IONS\n"
assert mgf_content[1].split("=")[1] == "test1\n"
assert mgf_content[8].split("=")[1] == "test2\n"
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