def main(args: Union[str, List[str]] = None) -> int:
# Initialize logging.
logging.basicConfig(format='{asctime} [{levelname}/{processName}] '
'{module}.{funcName} : {message}',
style='{',
level=logging.DEBUG)
# Load the configuration.
config.parse(args)
# Perform the search.
spec_lib = spectral_library.SpectralLibrary(
config.spectral_library_filename)
identifications = spec_lib.search(config.query_filename)
writer.write_mztab(identifications, config.out_filename,
spec_lib._library_reader)
spec_lib.shutdown()
logging.shutdown()
return 0
def main():
# initialize logging
logging.basicConfig(
format='%(asctime)s [%(levelname)s/%(processName)s] '
'%(module)s.%(funcName)s : %(message)s',
level=logging.DEBUG)
# load the config
config.parse()
# execute the search
if config.mode == 'bf':
spec_lib = spectral_library.SpectralLibraryBf(
config.spectral_library_filename)
elif config.mode == 'ann':
spec_lib = spectral_library.SpectralLibraryAnnoy(
config.spectral_library_filename)
identifications = spec_lib.search(config.query_filename)
writer.write_mztab(identifications, config.out_filename,
spec_lib._library_reader)
spec_lib.shutdown()
logging.shutdown()
# Search settings.
for key in metadata:
if 'software[1]-setting' in key:
param = metadata[key][:metadata[key].find(' ')]
value = metadata[key][metadata[key].rfind(' ') + 1:]
if value != 'None':
if value != 'False':
settings.append('--{}'.format(param))
if value not in ('False', 'True'):
settings.append(value)
# File names.
settings.append('dummy_spectral_library_filename')
settings.append('dummy_query_filename')
settings.append('dummy_output_filename')
config.parse(' '.join(settings))
# Retrieve information on the requested query.
query_id = args.Filename + "." + args.Scan + "." + args.Scan + "." + args.Charge
query_usi = "mzspec:PXD018117:" + args.Filename + ":scan:" + args.Scan + ":" + args.Sequence #[UNIMOD:121] [UNIMOD:21] [UNIMOD:21#g1] [-28.031300]
txt = 'm/z'
query_spectrum_number = "controllerType=0 controllerNumber=1 scan=" + args.Scan
query_uri = urlparse.urlparse(urlparse.unquote(metadata['ms_run[1]-location']))
query_filename = os.path.abspath(os.path.join(query_uri.netloc,
query_uri.path))
ssm = ssms.loc[query_id]
library_id = ssm['accession']
library_uri = urlparse.urlparse(urlparse.unquote(ssm['database']))
library_filename = os.path.abspath(
os.path.join(library_uri.netloc, library_uri.path))
score = ssm['search_engine_score[1]']
def main():
# Load the cmd arguments.
parser = argparse.ArgumentParser(
description='Visualize spectrum–spectrum matches from your '
'ANN-SoLo identification results')
parser.add_argument('mztab_filename',
help='Identifications in mzTab format')
parser.add_argument('query_id',
help='The identifier of the query to visualize')
args = parser.parse_args()
# Read the mzTab file.
metadata = {}
with open(args.mztab_filename) as f_mztab:
for line in f_mztab:
line_split = line.strip().split('\t')
if line_split[0] == 'MTD':
metadata[line_split[1]] = line_split[2]
else:
break # Metadata lines should be on top.
ssms = reader.read_mztab_ssms(args.mztab_filename)
# make sure the SSM ids are strings.
ssms.index = ssms.index.map(str)
# Recreate the search configuration.
settings = []
# Search settings.
for key in metadata:
if 'software[1]-setting' in key:
param = metadata[key][:metadata[key].find(' ')]
value = metadata[key][metadata[key].rfind(' ') + 1:]
if value != 'None':
if value != 'False':
settings.append('--{}'.format(param))
if value not in ('False', 'True'):
settings.append(value)
# File names.
settings.append('dummy_spectral_library_filename')
settings.append('dummy_query_filename')
settings.append('dummy_output_filename')
config.parse(' '.join(settings))
# Retrieve information on the requested query.
query_id = args.query_id
query_uri = urlparse.urlparse(
urlparse.unquote(metadata['ms_run[1]-location']))
query_filename = os.path.abspath(
os.path.join(query_uri.netloc, query_uri.path))
ssm = ssms.loc[query_id]
library_id = ssm['accession']
library_uri = urlparse.urlparse(urlparse.unquote(ssm['database']))
library_filename = os.path.abspath(
os.path.join(library_uri.netloc, library_uri.path))
score = ssm['search_engine_score[1]']
# Read library and query spectrum.
with reader.SpectralLibraryReader(library_filename) as lib_reader:
library_spectrum = lib_reader.get_spectrum(library_id, True)
query_spectrum = None
for spec in reader.read_mgf(query_filename):
if spec.identifier == query_id:
query_spectrum = process_spectrum(spec, False)
# Make sure that the precursor charge is set for query spectra
# with a undefined precursor charge.
query_spectrum.precursor_charge = library_spectrum.precursor_charge
break
# verify that the query spectrum was found
if query_spectrum is None:
raise ValueError('Could not find the specified query spectrum')
# Set the matching peaks in the query spectrum to correctly color them.
set_matching_peaks(library_spectrum, query_spectrum)
# Modify the colors to differentiate non-matching peaks.
plot.colors[None] = '#757575'
# Plot the match.
fig, ax = plt.subplots(figsize=(20, 10))
# Plot without annotations.
plot.mirror(query_spectrum, library_spectrum, {
'color_ions': True,
'annotate_ions': False
}, ax)
# Add annotations to the library spectrum.
max_intensity = library_spectrum.intensity.max()
for i, annotation in enumerate(library_spectrum.annotation):
if annotation is not None and annotation.ion_type != 'unknown':
x = library_spectrum.mz[i]
y = -library_spectrum.intensity[i] / max_intensity
ax.text(x,
y,
str(annotation),
color=plot.colors[annotation.ion_type],
zorder=5,
horizontalalignment='right',
verticalalignment='center',
rotation=90,
rotation_mode='anchor')
ax.set_ylim(-1.1, 1.05)
ax.text(0.5,
1.06,
f'{library_spectrum.peptide}, Score: {score:.3f}',
horizontalalignment='center',
verticalalignment='bottom',
fontsize='x-large',
fontweight='bold',
transform=ax.transAxes)
ax.text(0.5,
1.02, f'File: {os.path.basename(query_filename)}, '
f'Scan: {query_spectrum.identifier}, '
f'Precursor m/z: {query_spectrum.precursor_mz:.4f}, '
f'Library m/z: {library_spectrum.precursor_mz:.4f}, '
f'Charge: {query_spectrum.precursor_charge}',
horizontalalignment='center',
verticalalignment='bottom',
fontsize='large',
transform=ax.transAxes)
plt.savefig(f'{query_id}.png', dpi=300, bbox_inches='tight')
plt.close()
def main():
# load the cmd arguments
parser = argparse.ArgumentParser(
description='Visualize spectrum-spectrum matches from your '
'ANN-SoLo identification results')
parser.add_argument('mztab_filename',
help='Identifications in mzTab format')
parser.add_argument('query_id',
help='The identifier of the query to visualize')
args = parser.parse_args()
# read the mzTab file
metadata = {}
with open(args.mztab_filename) as f_mztab:
for line in f_mztab:
line_split = line.strip().split('\t')
if line_split[0] == 'MTD':
metadata[line_split[1]] = line_split[2]
else:
break # metadata lines should be on top
psms = reader.read_mztab_psms(args.mztab_filename)
# make sure the PSM id's are strings
psms.index = psms.index.map(str)
# recreate the search configuration
settings = []
# search settings
for key in metadata:
if 'software[1]-setting' in key:
param = metadata[key][:metadata[key].find(' ')]
value = metadata[key][metadata[key].rfind(' ') + 1:]
if value != 'False':
settings.append('--{}'.format(param))
if value != 'False' and value != 'True':
settings.append(value)
# file names
settings.append('dummy_spectral_library_filename')
settings.append('dummy_query_filename')
settings.append('dummy_output_filename')
config.parse(' '.join(settings))
# retrieve information on the requested query
query_id = args.query_id
query_uri = urlparse.urlparse(
urlparse.unquote(metadata['ms_run[1]-location']))
query_filename = os.path.abspath(
os.path.join(query_uri.netloc, query_uri.path))
psm = psms.loc[query_id]
library_id = psm['accession']
library_uri = urlparse.urlparse(urlparse.unquote(psm['database']))
library_filename = os.path.abspath(
os.path.join(library_uri.netloc, library_uri.path))
score = psm['search_engine_score[1]']
# read library and query spectrum
with reader.get_spectral_library_reader(library_filename) as lib_reader:
library_spectrum = lib_reader.get_spectrum(library_id, True)
query_spectrum = None
for spec in reader.read_mgf(query_filename):
if spec.identifier == query_id:
query_spectrum = spec
query_spectrum.process_peaks()
# make sure that the precursor charge is set for query spectra
# with a undefined precursor charge
if query_spectrum.precursor_charge is None:
query_spectrum.precursor_charge =\
library_spectrum.precursor_charge
break
# verify that the query spectrum was found
if query_spectrum is None:
raise ValueError('Could not find the specified query spectrum')
# compute the matching peaks
library_matches, query_matches, _ =\
get_matching_peaks(library_spectrum, query_spectrum)
# plot the match
fix, ax = plt.subplots(figsize=(20, 10))
# query spectrum on top
max_intensity = np.max(query_spectrum.intensities)
for i, (mass, intensity) in enumerate(
zip(query_spectrum.masses, query_spectrum.intensities)):
color = colors[query_matches.get(i)]
zorder = zorders[query_matches.get(i)]
ax.plot([mass, mass], [0, intensity / max_intensity],
color=color,
zorder=zorder)
# library spectrum mirrored underneath
max_intensity = np.max(library_spectrum.intensities)
for i, (mass, intensity, annotation) in enumerate(
zip(library_spectrum.masses, library_spectrum.intensities,
library_spectrum.annotations)):
color = colors[library_matches.get(i)]
zorder = zorders[library_matches.get(i)]
ax.plot([mass, mass], [0, -intensity / max_intensity],
color=color,
zorder=zorder)
if annotation is not None:
ax.text(mass - 5,
-intensity / max_intensity - 0.05,
'{}{}'.format(annotation[0], '+' * annotation[1]),
color=color,
rotation=270)
# horizontal line between the two spectra
ax.axhline(0, color='black')
# consistent axes range and labels
ax.set_xticks(np.arange(0, config.max_mz, 200))
ax.set_xlim(config.min_mz, config.max_mz)
y_ticks = np.arange(-1, 1.05, 0.25)
y_ticklabels = np.arange(-1, 1.05, 0.25)
y_ticklabels[y_ticklabels < 0] = -y_ticklabels[y_ticklabels < 0]
y_ticklabels = ['{:.0%}'.format(l) for l in y_ticklabels]
ax.set_yticks(y_ticks)
ax.set_yticklabels(y_ticklabels)
ax.set_ylim(-1.15, 1.05)
# show major/minor tick lines
ax.xaxis.set_minor_locator(mticker.AutoMinorLocator())
ax.yaxis.set_minor_locator(mticker.AutoMinorLocator())
ax.grid(b=True,
which='major',
color='lightgrey',
linestyle='--',
linewidth=1.0)
ax.grid(b=True,
which='minor',
color='lightgrey',
linestyle='--',
linewidth=0.5)
# small tick labels
ax.tick_params(axis='both', which='both', labelsize='small')
ax.set_xlabel('m/z')
ax.set_ylabel('Intensity')
ax.text(0.5,
1.06,
'{}, Score: {:.3f}'.format(library_spectrum.peptide, score),
horizontalalignment='center',
verticalalignment='bottom',
fontsize='x-large',
fontweight='bold',
transform=plt.gca().transAxes)
ax.text(0.5,
1.02,
'File: {}, Scan: {}, Precursor m/z: {:.4f}, '
'Library m/z: {:.4f}, Charge: {}'.format(
os.path.basename(query_filename), query_spectrum.identifier,
query_spectrum.precursor_mz, library_spectrum.precursor_mz,
query_spectrum.precursor_charge),
horizontalalignment='center',
verticalalignment='bottom',
fontsize='large',
transform=plt.gca().transAxes)
plt.savefig('{}.png'.format(query_id), dpi=300, bbox_inches='tight')
plt.close()