def make_peptide(pepxml_match, pepxml_scan, source):
peptide = {
'sequence': pepxml_match['peptide'],
'modified_sequence': pepxml_match['modified_sequence'],
'intensity': pepxml_match['probability'],
'mask': pepxml_match['fpe'],
'attr': {
'pepxml_id': pepxml_scan['index'],
'scan_id': pepxml_scan['start_scan'],
'charge': pepxml_scan['assumed_charge'],
'expect': pepxml_match['expect'],
'modifications': pepxml_match['modifications'],
'probability': pepxml_match['probability'],
'missed_cleavages': pepxml_match['num_missed_cleavages'],
'mass': pepxml_scan['precursor_neutral_mass'],
'mass_diff': pepxml_match['massdiff'],
'source': parse.basename(source),
}
}
def grab_opt(peptide_key, scan_key, source_dict):
if scan_key in source_dict:
peptide['attr'][peptide_key] = source_dict[scan_key]
grab_opt('retention_time', 'retention_time_sec', pepxml_scan)
grab_opt('score', 'ionscore', pepxml_match)
grab_opt('homology', 'homologyscore', pepxml_match)
grab_opt('identity', 'identityscore', pepxml_match)
peptide['attr']['matched_ions'] = str(pepxml_match['num_matched_ions'])
peptide['attr']['matched_ions'] += '/'
peptide['attr']['matched_ions'] += str(pepxml_match['tot_num_ions'])
return peptide
def make_peptide(pepxml_match, pepxml_scan, source):
peptide = {
'sequence': pepxml_match['peptide'],
'modified_sequence': pepxml_match['modified_sequence'],
'intensity': pepxml_match['probability'],
'mask': pepxml_match['fpe'],
'attr': {
'pepxml_id': pepxml_scan['index'],
'scan_id': pepxml_scan['start_scan'],
'charge': pepxml_scan['assumed_charge'],
'expect': pepxml_match['expect'],
'modifications': pepxml_match['modifications'],
'probability': pepxml_match['probability'],
'missed_cleavages': pepxml_match['num_missed_cleavages'],
'mass': pepxml_scan['precursor_neutral_mass'],
'mass_diff': pepxml_match['massdiff'],
'source': parse.basename(source),
}
}
def grab_opt(peptide_key, scan_key, source_dict):
if scan_key in source_dict:
peptide['attr'][peptide_key] = source_dict[scan_key]
grab_opt('retention_time', 'retention_time_sec', pepxml_scan)
grab_opt('score', 'ionscore', pepxml_match)
grab_opt('homology', 'homologyscore', pepxml_match)
grab_opt('identity', 'identityscore', pepxml_match)
peptide['attr']['matched_ions'] = str(pepxml_match['num_matched_ions'])
peptide['attr']['matched_ions'] += '/'
peptide['attr']['matched_ions'] += str(pepxml_match['tot_num_ions'])
return peptide
def make_match(psm, modification_table):
extracted_peptide_sequence, modifications = parse_peptide(
psm['peptide sequence'], modification_table)
peptide_sequence = psm['base peptide sequence']
if extracted_peptide_sequence != peptide_sequence:
logger.debug("Peptide sequences don't match: " + psm['peptide sequence'] + " " + extracted_peptide_sequence + " " + peptide_sequence)
q_value = float(psm['q-value (%)'])
if 'scan number' in psm:
scan_id = psm['scan number']
elif 'spectrum number' in psm:
scan_id = psm['spectrum number']
else:
scan_id = ''
if 'retention time (min)' in psm:
time = parse.round_decimal(psm['retention time (min)'], 4)
elif 'retention time (minutes)' in psm:
time = parse.round_decimal(psm['retention time (minutes)'], 4)
else:
time = ''
match = {
'sequence': peptide_sequence,
'attr': {
'scan_id': scan_id,
'retention_time': time,
'morpheus_score': parse.round_decimal(psm['morpheus score'], 4),
'charge': int(psm['precursor charge']),
'mass': parse.round_decimal(psm['precursor mass (da)'], 4),
'mass_diff': parse.round_decimal(psm['precursor mass error (da)'], 4),
'm/z': parse.round_decimal(psm['precursor m/z'], 4),
'source': parse.basename(psm['filename']),
'missed_cleavages': int(psm['missed cleavages']),
'q_value': q_value,
},
'modifications': [],
'intensity': 1.0,
'i': -1,
}
if modifications:
for modification in modifications:
modification['mass'] = parse.round_decimal(modification['mass'], 4)
match['modifications'] = modifications
modified_sequence = psm['peptide sequence'].split('.')[1]
match['attr']['modified_sequence'] = modified_sequence
return match
def make_peptide(pepxml_peptide, pepxml_scan, source):
peptide = {
'sequence': pepxml_peptide['peptide'],
'attr': {
'pepxml_id': pepxml_scan['index'],
'scan_id': pepxml_scan['start_scan'],
'expect': pepxml_peptide['expect'],
'retention_time': pepxml_scan['retention_time_sec'],
'modifications': pepxml_peptide['modifications'],
'source': parse.basename(source),
}
}
peptide['attr']['matched_ions'] = str(pepxml_peptide['num_matched_ions'])
peptide['attr']['matched_ions'] += '/'
peptide['attr']['matched_ions'] += str(pepxml_peptide['tot_num_ions'])
peptide['attr']['probability'] = pepxml_peptide['probability']
peptide['attr']['missed_cleavages'] = pepxml_peptide['num_missed_cleavages']
peptide['attr']['mass'] = pepxml_scan['precursor_neutral_mass']
peptide['attr']['mass_diff'] = pepxml_peptide['massdiff']
peptide['intensity'] = pepxml_peptide['probability']
return peptide
def make_peptide(pepxml_peptide, pepxml_scan, source):
peptide = {
'sequence': pepxml_peptide['peptide'],
'attr': {
'pepxml_id': pepxml_scan['index'],
'scan_id': pepxml_scan['start_scan'],
'expect': pepxml_peptide['expect'],
'retention_time': pepxml_scan['retention_time_sec'],
'modifications': pepxml_peptide['modifications'],
'source': parse.basename(source),
}
}
peptide['attr']['matched_ions'] = str(pepxml_peptide['num_matched_ions'])
peptide['attr']['matched_ions'] += '/'
peptide['attr']['matched_ions'] += str(pepxml_peptide['tot_num_ions'])
peptide['attr']['probability'] = pepxml_peptide['probability']
peptide['attr']['missed_cleavages'] = pepxml_peptide[
'num_missed_cleavages']
peptide['attr']['mass'] = pepxml_scan['precursor_neutral_mass']
peptide['attr']['mass_diff'] = pepxml_peptide['massdiff']
peptide['intensity'] = pepxml_peptide['probability']
return peptide
def get_proteins(protein_groups_fname, psm_fname, modifications_fname=None):
dump_dir = os.path.dirname(protein_groups_fname)
if modifications_fname is not None:
modification_table = read_modification_dict(modifications_fname)
else:
modification_table = {}
peptides = parse.read_tsv(psm_fname)
protein_groups = parse.read_tsv(protein_groups_fname)
if logger.root.level <= logging.DEBUG:
dump = os.path.join(dump_dir, 'peptides.dump')
logger.debug('Dumping peptides data structure to ' + dump)
parse.save_data_dict(peptides, dump)
dump = os.path.join(dump_dir, 'protein_groups.dump')
logger.debug('Dumping protein_groups data structure to ' + dump)
parse.save_data_dict(protein_groups, dump)
proteins = {}
for i_group, protein_group in enumerate(protein_groups):
descriptions = protein_group['protein description'].split(' / ')
seqids = [desc.split()[0] for desc in descriptions]
for seqid in seqids:
if seqid in proteins:
logger.warning(
"Different protein groups claim same first seqid", seqid)
protein = {
'description': descriptions[0],
'sequence': protein_group['protein sequence'],
'attr': {
'coverage':
protein_group['protein sequence coverage (%)'],
'morpheus-score':
parse.round_decimal(protein_group['summed morpheus score'], 4),
'i_group':
i_group,
'other_seqids':
seqids[1:],
'seqid':
seqids[0],
},
'sources': [{
'peptides': []
}]
}
proteins[seqids[0]] = protein
protein_by_seqid = {}
for seqid in proteins:
protein = proteins[seqid]
protein_by_seqid[seqid] = protein
for alt_seqid in protein['attr']['other_seqids']:
protein_by_seqid[alt_seqid] = protein
unmatched_peptides = []
n_peptide_matched = 0
for src_peptide in peptides:
descriptions = src_peptide['protein description'].split(' / ')
peptide_seqids = [d.split()[0] for d in descriptions]
protein = None
for peptide_seqid in peptide_seqids:
if peptide_seqid in protein_by_seqid:
protein = protein_by_seqid[peptide_seqid]
break
if protein is None:
unmatched_peptides.append(src_peptide)
continue
n_peptide_matched += 1
sequence = protein['sequence']
peptide_sequence, modifications = parse_peptide(
src_peptide['peptide sequence'], modification_table)
peptide_sequence = src_peptide['base peptide sequence']
i = sequence.index(peptide_sequence)
peptide = {
'sequence': peptide_sequence,
'attr': {
'scan_id':
src_peptide['scan number'],
'retention_time':
parse.round_decimal(src_peptide['retention time (min)'], 4),
'morpheus_score':
parse.round_decimal(src_peptide['morpheus score'], 4),
'mass':
parse.round_decimal(src_peptide['precursor mass (da)'], 4),
'mass_diff':
parse.round_decimal(src_peptide['precursor mass error (da)'],
4),
'm/z':
parse.round_decimal(src_peptide['precursor m/z'], 4),
'source':
parse.basename(src_peptide['filename']),
},
'intensity': src_peptide['morpheus score'] / len(peptide_sequence),
'i': i,
}
if modifications:
for modification in modifications:
modification['mass'] = parse.round_decimal(
modification['mass'], 4)
peptide['attr']['modifications'] = modifications
protein['sources'][0]['peptides'].append(peptide)
dump = os.path.join(dump_dir, 'proteins.dump')
logger.debug('Dumping proteins data structure to ' + dump)
if logger.root.level <= logging.DEBUG:
parse.save_data_dict(proteins, dump)
logger.info("Assigned {}/{} of PSMs.tsv to protein_groups.tsv".format(
n_peptide_matched, len(unmatched_peptides)))
return proteins
def get_proteins(protein_groups_fname, psm_fname, modifications_fname=None):
is_debug = logger.root.level <= logging.DEBUG
dump_dir = os.path.dirname(protein_groups_fname)
if modifications_fname is not None:
modification_table = read_modification_dict(modifications_fname)
else:
modification_table = {}
proteins = {}
dict_dump_writer = DictListWriter(
is_debug, os.path.join(dump_dir, 'protein_groups.dump'))
for i_group, protein_group in enumerate(
read_tsv_iter(protein_groups_fname)):
descriptions = protein_group['protein description'].split(' / ')
coverage_str = str(protein_group['protein sequence coverage (%)'])
if ';' in coverage_str:
coverage = float(get_first(coverage_str, ';'))
else:
coverage = float(get_first(coverage_str, '/'))
seqs = protein_group['protein sequence'].split('/')
seqids = [desc.split()[0] for desc in descriptions]
for seqid in seqids:
if seqid in proteins:
logger.warning(
"Different protein groups claim same first seqid", seqid)
protein = {
'description': descriptions[0],
'sequence': seqs[0],
'other_sequences': seqs[1:],
'attr': {
'coverage':
parse.round_decimal(coverage, 4),
'morpheus-score':
parse.round_decimal(protein_group['summed morpheus score'], 4),
'i_group':
i_group,
'other_seqids':
seqids[1:],
'seqid':
seqids[0],
},
'sources': [{
'peptides': []
}]
}
proteins[seqids[0]] = protein
dict_dump_writer.dump_dict(protein_group)
dict_dump_writer.close()
protein_by_seqid = {}
for seqid in proteins:
protein = proteins[seqid]
protein_by_seqid[seqid] = protein
for alt_seqid in protein['attr']['other_seqids']:
protein_by_seqid[alt_seqid] = protein
dict_dump_writer = DictListWriter(is_debug,
os.path.join(dump_dir, 'peptides.dump'))
n_peptide = 0
n_peptide_matched = 0
for src_peptide in read_tsv_iter(psm_fname):
dict_dump_writer.dump_dict(src_peptide)
descriptions = src_peptide['protein description'].split(' / ')
peptide_seqids = [d.split()[0] for d in descriptions]
protein = None
for peptide_seqid in peptide_seqids:
if peptide_seqid in protein_by_seqid:
protein = protein_by_seqid[peptide_seqid]
break
n_peptide += 1
if protein is None:
continue
n_peptide_matched += 1
sequence = protein['sequence']
extracted_peptide_sequence, modifications = parse_peptide(
src_peptide['peptide sequence'], modification_table)
peptide_sequence = src_peptide['base peptide sequence']
if extracted_peptide_sequence != peptide_sequence:
logger.warning("Peptide sequences don't match: " +
src_peptide['peptide sequence'] + " " +
extracted_peptide_sequence + " " + peptide_sequence)
i = sequence.find(peptide_sequence)
if i < 0:
logger.warning(peptide_sequence + ' not found in ' +
protein['attr']['seqid'])
continue
q_value = float(src_peptide['q-value (%)'])
if 'scan number' in src_peptide:
scan_id = src_peptide['scan number']
elif 'spectrum number' in src_peptide:
scan_id = src_peptide['spectrum number']
else:
scan_id = ''
if 'retention time (min)' in src_peptide:
time = parse.round_decimal(src_peptide['retention time (min)'], 4)
elif 'retention time (minutes)' in src_peptide:
time = parse.round_decimal(src_peptide['retention time (minutes)'],
4)
else:
time = ''
peptide = {
'sequence': peptide_sequence,
'attr': {
'scan_id':
scan_id,
'retention_time':
time,
'morpheus_score':
parse.round_decimal(src_peptide['morpheus score'], 4),
'mass':
parse.round_decimal(src_peptide['precursor mass (da)'], 4),
'mass_diff':
parse.round_decimal(src_peptide['precursor mass error (da)'],
4),
'm/z':
parse.round_decimal(src_peptide['precursor m/z'], 4),
'source':
parse.basename(src_peptide['filename']),
'q_value':
q_value,
},
'intensity': 1.0 - q_value / 100.0,
'i': i,
}
if modifications:
for modification in modifications:
modification['mass'] = parse.round_decimal(
modification['mass'], 4)
peptide['attr']['modifications'] = modifications
protein['sources'][0]['peptides'].append(peptide)
dict_dump_writer.close()
dump = os.path.join(dump_dir, 'proteins.dump')
if logger.root.level <= logging.DEBUG:
logger.debug('Dumping proteins data structure to ' + dump)
parse.save_data_dict(proteins, dump)
logger.info("Assigned {}/{} of PSMs.tsv to protein_groups.tsv".format(
n_peptide_matched, n_peptide))
return proteins
def get_proteins(xtandem_fname,
n_peak=50,
good_expect=1E-8,
cutoff_expect=1E-2):
proteins = {}
i_source = 0
print_scan = True
for scan in read_xtandem(xtandem_fname):
scan_id = scan['id']
x_vals = map(float, scan['masses'].split())
y_vals = map(float, scan['intensities'].split())
ions = [(x, y) for x, y in zip(x_vals, y_vals)]
ions.sort(key=lambda i: -i[1])
for xtandem_match in scan['matches']:
expect = xtandem_match['expect']
if cutoff_expect < expect:
continue
intensity = proteins_module.calc_minus_log_intensity(
expect, good_expect, cutoff_expect)
seqid = xtandem_match['seqid']
if seqid not in proteins:
protein = proteins_module.new_protein(seqid)
protein.update({
'sequence': xtandem_match['sequence'],
'description': xtandem_match['description'],
})
proteins[seqid] = protein
protein = proteins[seqid]
source = protein['sources'][i_source]
match = {
'sequence': xtandem_match['seq'],
'intensity': intensity,
'modifications': [],
'spectrum': ions[:n_peak],
'attr': {
'scan_id': scan['id'],
'charge': scan['charge'],
'expect': expect,
'missed_cleavages': xtandem_match['missed_cleavages'],
'mass': scan['mass'],
'source': parse.basename(xtandem_fname),
}
}
if xtandem_match['modifications']:
for mod in xtandem_match['modifications']:
i_mod_in_full_seq = int(mod['at']) - 1
full_seq = xtandem_match['sequence']
i_pep_seq = int(xtandem_match['start']) - 1
aa = mod['type']
if aa in peptidemass.aa_monoisotopic_mass:
mass = peptidemass.aa_monoisotopic_mass[aa]
else:
mass = 0.0
match['modifications'].append({
'i':
i_mod_in_full_seq - i_pep_seq,
'mass':
mod['modified'] + mass,
})
source['matches'].append(match)
proteins_module.calculate_peptide_positions(proteins)
return proteins
def get_proteins(protein_groups_fname, psm_fname, modifications_fname=None):
is_debug = logger.root.level <= logging.DEBUG
dump_dir = os.path.dirname(protein_groups_fname)
if modifications_fname is not None:
modification_table = read_modification_dict(modifications_fname)
else:
modification_table = {}
proteins = {}
dict_dump_writer = DictListWriter(is_debug, os.path.join(dump_dir, 'protein_groups.dump'))
for i_group, protein_group in enumerate(read_tsv_iter(protein_groups_fname)):
descriptions = protein_group['protein description'].split(' / ')
coverage_str = str(protein_group['protein sequence coverage (%)'])
if ';' in coverage_str:
coverage = float(get_first(coverage_str, ';'))
else:
coverage = float(get_first(coverage_str, '/'))
seqs = protein_group['protein sequence'].split('/')
seqids = [desc.split()[0] for desc in descriptions]
for seqid in seqids:
if seqid in proteins:
logger.warning("Different protein groups claim same first seqid", seqid)
protein = {
'description': descriptions[0],
'sequence': seqs[0],
'other_sequences': seqs[1:],
'attr': {
'coverage': parse.round_decimal(coverage, 4),
'morpheus-score': parse.round_decimal(protein_group['summed morpheus score'], 4),
'i_group': i_group,
'other_seqids': seqids[1:],
'seqid': seqids[0],
},
'sources': [{ 'peptides':[] }]
}
proteins[seqids[0]] = protein
dict_dump_writer.dump_dict(protein_group)
dict_dump_writer.close()
protein_by_seqid = {}
for seqid in proteins:
protein = proteins[seqid]
protein_by_seqid[seqid] = protein
for alt_seqid in protein['attr']['other_seqids']:
protein_by_seqid[alt_seqid] = protein
dict_dump_writer = DictListWriter(is_debug, os.path.join(dump_dir, 'peptides.dump'))
n_peptide = 0
n_peptide_matched = 0
for src_peptide in read_tsv_iter(psm_fname):
dict_dump_writer.dump_dict(src_peptide)
descriptions = src_peptide['protein description'].split(' / ')
peptide_seqids = [d.split()[0] for d in descriptions]
protein = None
for peptide_seqid in peptide_seqids:
if peptide_seqid in protein_by_seqid:
protein = protein_by_seqid[peptide_seqid]
break
n_peptide += 1
if protein is None:
continue
n_peptide_matched += 1
sequence = protein['sequence']
extracted_peptide_sequence, modifications = parse_peptide(
src_peptide['peptide sequence'],
modification_table)
peptide_sequence = src_peptide['base peptide sequence']
if extracted_peptide_sequence != peptide_sequence:
logger.warning("Peptide sequences don't match: " + src_peptide['peptide sequence'] + " " + extracted_peptide_sequence + " " + peptide_sequence)
i = sequence.find(peptide_sequence)
if i < 0:
logger.warning(peptide_sequence + ' not found in ' + protein['attr']['seqid'])
continue
q_value = float(src_peptide['q-value (%)'])
if 'scan number' in src_peptide:
scan_id = src_peptide['scan number']
elif 'spectrum number' in src_peptide:
scan_id = src_peptide['spectrum number']
else:
scan_id = ''
if 'retention time (min)' in src_peptide:
time = parse.round_decimal(src_peptide['retention time (min)'], 4)
elif 'retention time (minutes)' in src_peptide:
time = parse.round_decimal(src_peptide['retention time (minutes)'], 4)
else:
time = ''
peptide = {
'sequence': peptide_sequence,
'attr': {
'scan_id': scan_id,
'retention_time': time,
'morpheus_score': parse.round_decimal(src_peptide['morpheus score'], 4),
'mass': parse.round_decimal(src_peptide['precursor mass (da)'], 4),
'mass_diff': parse.round_decimal(src_peptide['precursor mass error (da)'], 4),
'm/z': parse.round_decimal(src_peptide['precursor m/z'], 4),
'source': parse.basename(src_peptide['filename']),
'q_value': q_value,
},
'intensity': 1.0 - q_value/100.0,
'i': i,
}
if modifications:
for modification in modifications:
modification['mass'] = parse.round_decimal(modification['mass'], 4)
peptide['attr']['modifications'] = modifications
protein['sources'][0]['peptides'].append(peptide)
dict_dump_writer.close()
dump = os.path.join(dump_dir, 'proteins.dump')
if logger.root.level <= logging.DEBUG:
logger.debug('Dumping proteins data structure to ' + dump)
parse.save_data_dict(proteins, dump)
logger.info("Assigned {}/{} of PSMs.tsv to protein_groups.tsv".format(n_peptide_matched, n_peptide))
return proteins