def parse_morpheus_psm(fname):
if 'hela' in fname:
celltype = 'Hela'
elif 'ecoli' in fname:
celltype = 'Ecoli'
values = []
key = 'Precursor Mass Error (ppm)'
for entry in datafile.read_csv(fname):
if entry['Target?'].lower() != "true":
continue
if float(entry['Q-Value (%)']) > 1:
continue
value = float(entry[key])
if abs(value) > 20:
continue
values.append(value)
if len(values) == 0:
return {}
else:
average_param = celltype + ' ' + key
upper_param = celltype + ' ' + key + ' Upper'
avg, std = datafile.get_avg_std(values)
result = {
average_param: avg,
upper_param: avg + std
}
return result
def read_peptides_from_csv(fname):
"""
Read peptide sequences, sorts by shortest first, and
removes repeats.
"""
peptides = []
for entry in datafile.read_csv(fname):
if 'Sequence' not in entry:
continue
peptide = {
'sequence': entry['Sequence'],
'overlaps': [],
'subsets': [],
'supersets': [],
'groups': [],
'attr': {}
}
for key in entry:
if not key == 'Sequence':
peptide['attr'][key] = entry[key]
peptides.append(peptide)
peptides.sort()
peptides.sort(key=lambda p: len(p['sequence']))
for i, peptide in enumerate(peptides):
peptide['i_peptide'] = i
return peptides
def add_modifications(modifications_tsv, extra_modifications_tsv):
mods = list(datafile.read_csv(modifications_tsv))
default_mods = [mod['Description'] for mod in mods]
logger.debug("Adding %s to %s" % (extra_modifications_tsv, modifications_tsv))
new_mod_lines = []
with open(extra_modifications_tsv, 'Ur') as f:
for line in f.readlines()[1:]:
mod = line.split('\t')[0]
if mod not in default_mods:
new_mod_lines.append(line)
with open(modifications_tsv, 'a') as f:
for line in new_mod_lines:
f.write(line)
def read_freqs_csv(freqs_csv):
freqs = {}
for entry in datafile.read_csv(freqs_csv):
aa = entry["Amino Acid"]
if aa == "Total":
continue
for key, val in entry.items():
if key.startswith("P"):
i = int(key[1:])
if i not in freqs:
freqs[i] = {}
freqs[i][aa] = float(val)/100.
return freqs
def parse_morpheus_summary(fname):
result = {}
if 'hela' in fname:
celltype = 'Hela'
elif 'ecoli' in fname:
celltype = 'Ecoli'
for entry in datafile.read_csv(fname):
for key, val in entry.items():
if not key:
result[key] = None
else:
result[celltype + ' ' + key] = datafile.parse_string(val)
break
return result
def read_peptides_from_csv(fname):
peptides = {}
for entry in datafile.read_csv(fname):
if int(entry['Uni']) == 0:
continue
n = entry['Uni']
seqids = entry['protein'].split(',')
sequence = entry['Sequence'].split('.')[1]
if sequence.startswith('n'):
sequence = sequence[1:]
if sequence in peptides:
existing_seqids = set(peptides[sequence]['seqids'])
if existing_seqids != seqids:
peptides[sequence]['seqids'] = existing_seqids.union(seqids)
else:
peptides[sequence] = {
'sequence': sequence,
'seqids': seqids,
}
return peptides
def read_peptides_from_csv(fname):
peptides = {}
for entry in datafile.read_csv(fname):
if int(entry['Uni']) == 0:
continue
n = entry['Uni']
seqids = entry['protein'].split(',')
sequence = entry['Sequence'].split('.')[1]
if sequence.startswith('n'):
sequence = sequence[1:]
if sequence in peptides:
existing_seqids = set(peptides[sequence]['seqids'])
if existing_seqids != seqids:
peptides[sequence]['seqids'] = existing_seqids.union(seqids)
else:
peptides[sequence] = {
'sequence': sequence,
'seqids': seqids,
}
return peptides
def get_peptide_by_seq(fnames):
times = []
peptide_by_seq = {}
for fname in fnames:
date = datafile.get_date_from_fname(fname)
logger.debug("Reading peptides in %s" % date)
seqs = []
for entry in datafile.read_csv(fname):
seq = entry['Peptide Sequence']
if seq not in peptide_by_seq:
peptide_by_seq[seq] = {
'sequence': seq,
'base_sequence': entry['Base Peptide Sequence'],
'intensity_fractions': [],
'ion_fractions': [],
'n_log': 0
}
peptide = peptide_by_seq[seq]
peptide['intensity_fractions'].append(
float(entry['Fraction of Intensity Matching']))
peptide['ion_fractions'].append(
float(entry['Ratio of Matching Products']))
seqs.append(seq)
for seq in set(seqs):
peptide_by_seq[seq]['n_log'] += 1
for peptide in peptide_by_seq.values():
ion_avg, ion_stdv = datafile.get_avg_std(peptide['ion_fractions'])
intensity_avg, intensity_stdv = datafile.get_avg_std(
peptide['intensity_fractions'])
peptide.update({
'ion_avg': ion_avg,
'ion_stdv': ion_stdv,
'intensity_avg': intensity_avg,
'intensity_stdv': intensity_stdv,
'n_psm': len(peptide['intensity_fractions']),
})
return peptide_by_seq
def read_peptides_from_csv(fname):
"""
Read peptide sequences, sorts by shortest first, and
removes repeats.
"""
peptides = []
for entry in datafile.read_csv(fname):
sequence = entry['Sequence']
if not sequence:
continue
peptides.append({
'sequence': sequence,
'modifications': entry['Modifications'],
'protein': entry['Protein ID'],
'overlaps': [],
'groups': [],
})
peptides.sort()
peptides.sort(key=lambda p: len(p['sequence']))
for i, peptide in enumerate(peptides):
peptide['i_peptide'] = i
return peptides
def read_peptides_from_csv(fname):
"""
Read peptide sequences, sorts by shortest first, and
removes repeats.
"""
peptides = []
for entry in datafile.read_csv(fname):
sequence = entry['Sequence']
if not sequence:
continue
peptides.append({
'sequence': sequence,
'modifications': entry['Modifications'],
'protein': entry['Protein ID'],
'overlaps': [],
'groups': [],
})
peptides.sort()
peptides.sort(key=lambda p: len(p['sequence']))
for i, peptide in enumerate(peptides):
peptide['i_peptide'] = i
return peptides
('DR', 'green', 2, list(range(1, 5)))]
source_sets = []
for cell_type, color, i_source, repeats in exp_sets:
for i_repeat in repeats:
if skip == '_skip_dr4':
if cell_type == 'DR' and i_repeat == 4:
continue
source_sets.append(
('%s%d' % (cell_type, i_repeat), color, i_source))
print source_sets
protein = {}
for exp, color, i_source in source_sets:
fname = 'Data_PeptideOverlay/%s_motif.csv' % exp
for entry in datafile.read_csv(fname):
seqid = uniprot.parse_fasta_header(entry['Accessions'])[0]
if seqid not in protein:
protein[seqid] = default_protein()
protein[seqid]['description'] = entry['Names']
source = protein[seqid]['sources'][i_source]
source['color'] = color
source['peptides'].append(entry['sequence'])
seqids, fasta = uniprot.read_fasta('../db/uniprot_sprot.fasta')
for seqid in protein:
sequence = fasta[seqid]['sequence']
protein[seqid]['sequence'] = sequence
protein[seqid]['length'] = len(sequence)
for source in protein[seqid]['sources']:
def print_modifications(modifications_tsv):
logger.debug('modifications.tsv: ' + os.path.relpath(modifications_tsv))
mods = [g['Description'] for g in datafile.read_csv(modifications_tsv)]
logger.debug('modifications: %s' % mods)