def parse_scan(scan_elem, nsmap):
scan = parse.parse_attrib(scan_elem)
scan['matches'] = []
tag = lambda tag_id: parse.fixtag('', tag_id, nsmap)
for search_elem in scan_elem.findall(parse.fixtag('', "search_result", nsmap)):
search_hit_elem = search_elem[0]
match = parse.parse_attrib(search_hit_elem)
match['modified_sequence'] = match['peptide']
match['other_seqids'] = []
for alt_protein in search_hit_elem.findall(parse.fixtag('', 'alternative_protein', nsmap)):
match['other_seqids'].append(alt_protein.attrib['protein'])
match['modifications'] = []
for modified_elem in search_hit_elem.findall(parse.fixtag('', 'modification_info', nsmap)):
attr = parse.parse_attrib(modified_elem)
match['modified_sequence'] = attr['modified_peptide']
for modification_elem in modified_elem.findall(parse.fixtag('', 'mod_aminoacid_mass', nsmap)):
attr = parse.parse_attrib(modification_elem)
attr['i'] = attr['position'] - 1
del attr['position']
match['modifications'].append(attr)
for score_elem in search_hit_elem.findall(tag('search_score')):
match.update(parse.parse_name_value(score_elem))
for analysis_elem in search_hit_elem.find(parse.fixtag('', 'analysis_result', nsmap)):
if analysis_elem.tag == parse.fixtag('', 'peptideprophet_result', nsmap):
match.update(parse.parse_attrib(analysis_elem))
for param_elem in analysis_elem[0]:
match.update(parse.parse_name_value(param_elem))
scan['matches'].append(match)
return scan
def parse_protein_group(elem, nsmap):
group = parse.parse_attrib(elem)
group['proteins'] = []
for protein_elem in elem.findall(parse.fixtag('', 'protein', nsmap)):
protein = parse.parse_attrib(protein_elem)
protein['group_number'] = group['group_number']
annotation_elem = protein_elem.find(parse.fixtag('', 'annotation', nsmap))
if annotation_elem is not None:
protein['description'] = annotation_elem.attrib['protein_description']
protein['other_seqids'] = []
for alt_protein in protein_elem.findall(parse.fixtag('', 'indistinguishable_protein', nsmap)):
protein['other_seqids'].append(alt_protein.attrib['protein_name'])
protein['other_seqids'] = protein['other_seqids']
protein['protein_name'] = protein['protein_name']
protein['peptides'] = []
n_unique_peptide = 0
for peptide_elem in protein_elem.findall(parse.fixtag('', 'peptide', nsmap)):
peptide = parse.parse_attrib(peptide_elem)
protein['peptides'].append(peptide)
peptide['modifications'] = []
peptide['modified_sequence'] = peptide['peptide_sequence']
for modified_elem in peptide_elem.findall(parse.fixtag('', 'modification_info', nsmap)):
attr = parse.parse_attrib(modified_elem)
peptide['modified_sequence'] = attr['modified_peptide']
for modification_elem in modified_elem.findall(parse.fixtag('', 'mod_aminoacid_mass', nsmap)):
attr = parse.parse_attrib(modification_elem)
peptide['modifications'].append(attr)
group['proteins'].append(protein)
return group
def parse_scan(group, nsmap):
scan = {}
fastas = {}
scan.update(parse.parse_attrib(group))
scan['matches'] = []
for protein in group.findall('protein'):
match = {}
words = protein.attrib['label'].split()
seqid = words[0]
description = ' '.join(words[1:])
peptide_elem = protein.find('peptide')
match['seqid'] = seqid
if seqid not in fastas:
sequence = strip_whitespace(peptide_elem.text)
fastas[seqid] = {
'description': description,
'sequence': sequence,
}
domain_elem = peptide_elem.find('domain')
match.update(parse.parse_attrib(domain_elem))
scan['matches'].append(match)
elem = group.find('group[@label="fragment ion mass spectrum"]')
note = elem.find('note')
if note:
scan['Description'] = elem.find('note').text.strip()
else:
scan['Description'] = ''
masses_elem = elem.find(
'GAML:trace/GAML:Xdata/GAML:values', namespaces=nsmap)
scan['masses'] = masses_elem.text.strip()
intensities_elem = elem.find(
'GAML:trace/GAML:Ydata/GAML:values', namespaces=nsmap)
scan['intensities'] = intensities_elem.text.strip()
charge_elem = elem.find(
'GAML:trace/GAML:attribute[@type="charge"]', namespaces=nsmap)
scan['charge'] = charge_elem.text.strip()
mass_elem = elem.find(
'GAML:trace/GAML:attribute[@type="M+H"]', namespaces=nsmap)
scan['mass'] = mass_elem.text.strip()
return scan, fastas
def parse_scan(group, nsmap):
scan = {}
fastas = {}
scan.update(parse.parse_attrib(group))
scan['matches'] = []
for protein in group.findall('protein'):
match = {}
words = protein.attrib['label'].split()
seqid = words[0]
description = ' '.join(words[1:])
peptide_elem = protein.find('peptide')
match['seqid'] = seqid
if seqid not in fastas:
sequence = strip_whitespace(peptide_elem.text)
fastas[seqid] = {
'description': description,
'sequence': sequence,
}
domain_elem = peptide_elem.find('domain')
match.update(parse.parse_attrib(domain_elem))
scan['matches'].append(match)
elem = group.find('group[@label="fragment ion mass spectrum"]')
note = elem.find('note')
if note:
scan['Description'] = elem.find('note').text.strip()
else:
scan['Description'] = ''
masses_elem = elem.find('GAML:trace/GAML:Xdata/GAML:values',
namespaces=nsmap)
scan['masses'] = masses_elem.text.strip()
intensities_elem = elem.find('GAML:trace/GAML:Ydata/GAML:values',
namespaces=nsmap)
scan['intensities'] = intensities_elem.text.strip()
charge_elem = elem.find('GAML:trace/GAML:attribute[@type="charge"]',
namespaces=nsmap)
scan['charge'] = charge_elem.text.strip()
mass_elem = elem.find('GAML:trace/GAML:attribute[@type="M+H"]',
namespaces=nsmap)
scan['mass'] = mass_elem.text.strip()
return scan, fastas
def parse_scan(top_elem, nsmap):
scan = {}
scan.update(parse.parse_attrib(top_elem))
scan['matches'] = []
for protein in top_elem.findall('protein'):
words = protein.attrib['label'].split()
seqid = words[0]
description = ' '.join(words[1:])
peptide_elem = protein.find('peptide')
sequence = strip_whitespace(peptide_elem.text)
match = {
'seqid': seqid,
'sequence': sequence,
'description': description,
'modifications': []
}
domain_elem = peptide_elem.find('domain')
for mod_elem in domain_elem.findall('aa'):
match['modifications'].append(parse.parse_attrib(mod_elem))
match.update(parse.parse_attrib(domain_elem))
scan['matches'].append(match)
elem = top_elem.find('group[@label="fragment ion mass spectrum"]')
note = elem.find('note')
if note is not None:
scan['Description'] = elem.find('note').text.strip()
else:
scan['Description'] = ''
masses_elem = elem.find('GAML:trace/GAML:Xdata/GAML:values',
namespaces=nsmap)
scan['masses'] = masses_elem.text.strip()
intensities_elem = elem.find('GAML:trace/GAML:Ydata/GAML:values',
namespaces=nsmap)
scan['intensities'] = intensities_elem.text.strip()
charge_elem = elem.find('GAML:trace/GAML:attribute[@type="charge"]',
namespaces=nsmap)
scan['charge'] = charge_elem.text.strip()
mass_elem = elem.find('GAML:trace/GAML:attribute[@type="M+H"]',
namespaces=nsmap)
scan['mass'] = mass_elem.text.strip()
return scan
def parse_protein_group(elem, nsmap):
group = parse.parse_attrib(elem)
group['proteins'] = []
for protein_elem in elem.findall(parse.fixtag('', 'protein', nsmap)):
protein = parse.parse_attrib(protein_elem)
protein['group_number'] = group['group_number']
for parameter_elem in protein_elem.findall(
parse.fixtag('', 'parameter', nsmap)):
key = parameter_elem.attrib['name']
val = parameter_elem.attrib['value']
protein[key] = val
annotation_elem = protein_elem.find(
parse.fixtag('', 'annotation', nsmap))
if annotation_elem is not None:
protein['description'] = annotation_elem.attrib[
'protein_description']
protein['other_seqids'] = []
for alt_protein in protein_elem.findall(
parse.fixtag('', 'indistinguishable_protein', nsmap)):
protein['other_seqids'].append(alt_protein.attrib['protein_name'])
protein['other_seqids'] = protein['other_seqids']
protein['protein_name'] = protein['protein_name']
protein['peptides'] = []
n_unique_peptide = 0
for peptide_elem in protein_elem.findall(
parse.fixtag('', 'peptide', nsmap)):
peptide = parse.parse_attrib(peptide_elem)
protein['peptides'].append(peptide)
peptide['modifications'] = []
peptide['modified_sequence'] = peptide['peptide_sequence']
for modified_elem in peptide_elem.findall(
parse.fixtag('', 'modification_info', nsmap)):
attr = parse.parse_attrib(modified_elem)
peptide['modified_sequence'] = attr['modified_peptide']
for modification_elem in modified_elem.findall(
parse.fixtag('', 'mod_aminoacid_mass', nsmap)):
attr = parse.parse_attrib(modification_elem)
peptide['modifications'].append(attr)
group['proteins'].append(protein)
return group
def parse_protein_probabilities(elem, nsmap):
probs = []
for data_point in elem.findall(parse.fixtag('', 'protein_summary_data_filter', nsmap)):
attrib = parse.parse_attrib(data_point)
probs.append({
'error': attrib['false_positive_error_rate'],
'prob': attrib['min_probability'],
})
probs.sort(key=lambda d:d['error'])
return probs
def parse_protein_probabilities(elem, nsmap):
probs = []
for data_point in elem.findall(
parse.fixtag('', 'protein_summary_data_filter', nsmap)):
attrib = parse.parse_attrib(data_point)
probs.append({
'error': attrib['false_positive_error_rate'],
'prob': attrib['min_probability'],
})
probs.sort(key=lambda d: d['error'])
return probs
def parse_protein_probabilities(self, elem):
self.distribution = []
for data_point in self.findall(elem, 'protein_summary_data_filter'):
attrib = parse.parse_attrib(data_point)
self.distribution.append({
'error':
attrib['false_positive_error_rate'],
'prob':
attrib['min_probability'],
})
self.distribution.sort(key=lambda d: d['error'])
def parse_scan(scan_elem, nsmap):
scan = parse.parse_attrib(scan_elem)
scan['matches'] = []
tag = lambda tag_id: parse.fixtag('', tag_id, nsmap)
for search_elem in scan_elem.findall(
parse.fixtag('', "search_result", nsmap)):
search_hit_elem = search_elem[0]
match = parse.parse_attrib(search_hit_elem)
match['modified_sequence'] = match['peptide']
match['other_seqids'] = []
for alt_protein in search_hit_elem.findall(
parse.fixtag('', 'alternative_protein', nsmap)):
match['other_seqids'].append(alt_protein.attrib['protein'])
match['modifications'] = []
for modified_elem in search_hit_elem.findall(
parse.fixtag('', 'modification_info', nsmap)):
attr = parse.parse_attrib(modified_elem)
match['modified_sequence'] = attr['modified_peptide']
for modification_elem in modified_elem.findall(
parse.fixtag('', 'mod_aminoacid_mass', nsmap)):
attr = parse.parse_attrib(modification_elem)
attr['i'] = attr['position'] - 1
del attr['position']
match['modifications'].append(attr)
for score_elem in search_hit_elem.findall(tag('search_score')):
match.update(parse.parse_name_value(score_elem))
for analysis_elem in search_hit_elem.find(
parse.fixtag('', 'analysis_result', nsmap)):
if analysis_elem.tag == parse.fixtag('', 'peptideprophet_result',
nsmap):
match.update(parse.parse_attrib(analysis_elem))
for param_elem in analysis_elem[0]:
match.update(parse.parse_name_value(param_elem))
scan['matches'].append(match)
return scan
def parse_scan(self, scan_elem):
scan = parse.parse_attrib(scan_elem)
scan['matches'] = []
for search_elem in self.findall(scan_elem, "search_result"):
search_hit_elem = search_elem[0]
pepxml_match = parse.parse_attrib(search_hit_elem)
pepxml_match['modified_sequence'] = pepxml_match['peptide']
pepxml_match['other_seqids'] = []
for alt_protein in self.findall(search_hit_elem,
'alternative_protein'):
pepxml_match['other_seqids'].append(
alt_protein.attrib['protein'])
pepxml_match['modifications'] = []
for modified_elem in self.findall(search_hit_elem,
'modification_info'):
attr = parse.parse_attrib(modified_elem)
pepxml_match['modified_sequence'] = attr['modified_peptide']
for modification_elem in self.findall(modified_elem,
'mod_aminoacid_mass'):
attr = parse.parse_attrib(modification_elem)
attr['i'] = attr['position'] - 1
del attr['position']
pepxml_match['modifications'].append(attr)
for score_elem in self.findall(search_hit_elem, 'search_score'):
pepxml_match.update(parse.parse_name_value(score_elem))
for analysis_elem in self.find(search_hit_elem, 'analysis_result'):
if analysis_elem.tag == self.search_tag(
'peptideprophet_result'):
pepxml_match.update(parse.parse_attrib(analysis_elem))
for param_elem in analysis_elem[0]:
pepxml_match.update(parse.parse_name_value(param_elem))
scan['matches'].append(pepxml_match)
return scan
def parse_peptide_probabilities(elem, nsmap):
# try with error_point
error_points = elem.findall(parse.fixtag('', 'error_point', nsmap))
if len(error_points) == 0:
charge = 0
for charge_elem in elem.findall(parse.fixtag('', 'roc_error_data', nsmap)):
if charge_elem.attrib['charge'] == 'all':
error_points = charge_elem.findall(parse.fixtag('', 'error_point', nsmap))
break
probs = []
for elem in error_points:
attrib = parse.parse_attrib(elem)
probs.append({
'error': attrib['error'],
'prob': attrib['min_prob'],
})
probs.sort(key=lambda d:d['error'])
return probs
def parse_peptide_probabilities(elem, nsmap):
# try with error_point
error_points = elem.findall(parse.fixtag('', 'error_point', nsmap))
if len(error_points) == 0:
charge = 0
for charge_elem in elem.findall(
parse.fixtag('', 'roc_error_data', nsmap)):
if charge_elem.attrib['charge'] == 'all':
error_points = charge_elem.findall(
parse.fixtag('', 'error_point', nsmap))
break
probs = []
for elem in error_points:
attrib = parse.parse_attrib(elem)
probs.append({
'error': attrib['error'],
'prob': attrib['min_prob'],
})
probs.sort(key=lambda d: d['error'])
return probs
def parse_peptide_probabilities(self, elem):
# try with error_point
error_points = self.findall(elem, 'error_point')
if len(error_points) == 0:
charge = 0
for charge_elem in self.findall(elem, 'roc_error_data'):
if charge_elem.attrib['charge'] == 'all':
error_points = self.findall(charge_elem, 'error_point')
break
self.distribution = []
for elem in error_points:
attrib = parse.parse_attrib(elem)
self.distribution.append({
'error': attrib['error'],
'prob': attrib['min_prob'],
})
self.distribution.sort(key=lambda d: d['error'])
peptide_probability = error_to_probability(self.distribution, 0.01)
logger.info('Peptide probability cutoff for 0.01 fpe: %f' %
peptide_probability)