def forge_prediction_record(sequence, reference):
seq_obj = Sequence(sequence) if isinstance(
sequence, basestring) else copy.deepcopy(sequence)
calculated_mass = seq_obj.mass
glycan_map = {}
modifications = []
loc_rules = modification.get_position_modifier_rules_dict(seq_obj)
for i, (aa, mods) in enumerate(seq_obj):
for mod in mods:
if "Glycan" in mod.name:
glycan_map[i] = mod.name
else:
# Construct the set of acceptable reasons why this modification is here.
# Infer the least permissive modification rule.
try:
why = mod.why_valid(aa, loc_rules[i])
modifications.append(modification.Modification(why, (i,)))
except AttributeError:
print(mod)
raise
# Remove glycans from the sequence string to conform to the SequenceSpace
# expectations
for site, glycan in glycan_map.items():
# Don't discard anonymous HexNAcs. Downstream functions can handle them
if glycan != "HexNAc":
seq_obj.drop_modification(site, glycan)
seq_obj.add_modification(site, "HexNAc")
# Build the semicolon separated string for glycan compositions
glycan_composition = []
glycan_composition = [map(int, glycan.replace("Glycan", '').replace("[", "").replace("]", "").split(";"))
for glycan in glycan_map.values()]
glycan_composition = map(sum, zip(*glycan_composition))
glycan_composition_restring = "[" + \
";".join(map(str, glycan_composition)) + "]"
forgery = reference.copy()
forgery["_old_Glycopeptide_identifier"] = str(sequence)
forgery.Calc_mass = calculated_mass
forgery.Obs_Mass = forgery.Calc_mass - reference.ppm_error
forgery.Glycopeptide_identifier = str(seq_obj)
forgery.Glycan = glycan_composition_restring
forgery.glyco_sites = len(glycan_map)
forgery.Seq_with_mod = str(seq_obj)
return forgery
def random_glycopeptide_to_fragments(sequence_record):
try:
seq_obj = Sequence(sequence_record.Glycopeptide_identifier)
except:
print(sequence_record)
raise
glycan_map = {}
modifications = []
loc_rules = modification.get_position_modifier_rules_dict(seq_obj)
for i, (aa, mods) in enumerate(seq_obj):
for mod in mods:
if "Glycan" in mod.name or "HexNAc" in mod.name:
glycan_map[i] = mod.name
else:
# Construct the set of acceptable reasons why this modification is here.
# Infer the least permissive modification rule.
try:
why = mod.why_valid(aa, loc_rules[i])
modifications.append(modification.Modification(why, (i,)))
except AttributeError:
print(mod)
raise
# Remove glycans from the sequence string to conform to the SequenceSpace expectations
for site, glycan in glycan_map.items():
# Don't discard anonymous HexNAcs. Downstream functions can handle them
if glycan != "HexNAc":
seq_obj.drop_modification(site, glycan)
glycosylation_sites = glycan_map.keys()
if sequence_record.Glycan is None:
# Build the semicolon separated string for glycan compositions
glycan_composition = []
glycan_composition = [map(int, glycan.replace("Glycan", '').replace("[", "").replace("]", "").split(";"))
for glycan in glycan_map.values()]
glycan_composition = map(sum, zip(*glycan_composition))
glycan_composition_restring = "[" + ";".join(map(str, glycan_composition)) + "]"
else:
glycan_composition_restring = sequence_record.Glycan
# Begin generating fragment ions
b_type = seq_obj.get_fragments('B')
b_ions = []
b_ions_HexNAc = []
for b in b_type:
for fm in b:
key = fm.get_fragment_name()
if key == "B1" or re.search(r'B1\+', key):
# B1 Ions aren't actually seen in reality, but are an artefact of the generation process
# so do not include them in the output
continue
mass = fm.get_mass()
if "HexNAc" in key:
b_ions_HexNAc.append({"key": key, "mass": mass})
else:
b_ions.append({"key": key, "mass": mass})
y_type = seq_obj.get_fragments('Y')
y_ions = []
y_ions_HexNAc = []
for y in y_type:
for fm in y:
key = fm.get_fragment_name()
mass = fm.get_mass()
if "HexNAc" in key:
y_ions_HexNAc.append({"key": key, "mass": mass})
else:
y_ions.append({"key": key, "mass": mass})
peptide_seq = strip_modifications(seq_obj.get_sequence())
pep_stubs = StubGlycopeptide(peptide_seq, glycan_composition_restring,
len(glycosylation_sites), glycan_composition_restring)
stub_ions = pep_stubs.get_stubs()
assert len(stub_ions) > 1
oxonium_ions = pep_stubs.get_oxonium_ions()
ions = {
"MS1_Score": sequence_record.MS1_Score,
"Obs_Mass": sequence_record.Obs_Mass,
"Calc_mass": sequence_record.Calc_mass,
"ppm_error": sequence_record.ppm_error,
"Peptide": peptide_seq,
"Peptide_mod": sequence_record.Peptide_mod,
"Glycan": glycan_composition_restring,
"vol": sequence_record.vol,
"glyco_sites": len(glycan_map),
"startAA": None,
"endAA": None,
"Seq_with_mod": seq_obj.get_sequence(include_glycan=False),
"bare_b_ions": b_ions,
"b_ions_with_HexNAc": b_ions_HexNAc,
"bare_y_ions": y_ions,
"y_ions_with_HexNAc": y_ions_HexNAc,
"pep_stub_ions": stub_ions,
"Oxonium_ions": oxonium_ions,
"Glycopeptide_identifier": seq_obj.get_sequence(include_glycan=False) + glycan_composition_restring,
"_batch_id": int(sequence_record._batch_id)
}
return ions
