def get_theor_spectrum(peptide,
acc_frag,
types=('b', 'y'),
maxcharge=None,
**kwargs):
"""
Calculates theoretical spectra in two ways: usual one. and formatter in integer (mz / frag_acc).
`peptide` -peptide sequence
`acc_frag` - accuracy of matching.
`types` - ion types.
`maxcharge` - maximum charge.
----------
Returns spectra in two ways (usual, integer)
"""
peaks = {}
theoretical_set = defaultdict(set)
pl = len(peptide) - 1
if not maxcharge:
maxcharge = 1 + int(ec.charge(peptide, pH=2))
for charge in range(1, maxcharge + 1):
for ion_type in types:
nterminal = ion_type[0] in 'abc'
if nterminal:
maxpart = peptide[:-1]
maxmass = cmass.fast_mass(maxpart,
ion_type=ion_type,
charge=charge,
**kwargs)
marr = np.zeros((pl, ), dtype=float)
marr[0] = maxmass
for i in range(1, pl):
marr[i] = marr[i - 1] - mass.fast_mass2(
[maxpart[-i]]) / charge ### recalculate
else:
maxpart = peptide[1:]
maxmass = cmass.fast_mass(maxpart,
ion_type=ion_type,
charge=charge,
**kwargs)
marr = np.zeros((pl, ), dtype=float)
marr[pl - 1] = maxmass
for i in range(pl - 2, -1, -1):
marr[i] = marr[i + 1] - mass.fast_mass2(
[maxpart[-(i + 2)]]) / charge ### recalculate
tmp = marr / acc_frag
tmp = tmp.astype(int)
theoretical_set[ion_type].update(tmp)
marr.sort()
peaks[ion_type, charge] = marr
return peaks, theoretical_set
def generate_Yion(Ytype, aa_mass, ion_maxcharge, ion_type, labels, temp):
str_seq = parser.tostring(temp, False)
for m in range(ion_maxcharge):
if "Y" in ion_type:
yield Ion(seq=temp, charge=m + 1, ion_type="Y",
mz=mass.fast_mass2(str_seq, charge=m + 1, aa_mass=aa_mass, labels=labels),
fragment_number=int(Ytype[1:]) + 1)
def generate_byion(aa_mass, b_stop_at, ion_maxcharge, ion_type, labels, seq_len, temp, y_stop_at, b_selected, y_selected):
for s in range(seq_len):
if not (s + 1) == 1:
for m in range(ion_maxcharge):
for ion in ion_type:
if ion == "y":
position = seq_len - s - 1
if y_stop_at > -1:
if position <= y_stop_at:
continue
if y_selected:
if position in y_selected:
seq = temp[position:]
str_seq = parser.tostring(seq, False)
# print(str_seq)
yield Ion(seq=temp, ion_type=ion, charge=m + 1,
mz=mass.fast_mass2(str_seq, ion_type=ion, charge=m + 1, aa_mass=aa_mass,
labels=labels), fragment_number=s + 1)
else:
seq = temp[position:]
str_seq = parser.tostring(seq, False)
# print(str_seq)
yield Ion(seq=temp, ion_type=ion, charge=m + 1,
mz=mass.fast_mass2(str_seq, ion_type=ion, charge=m + 1, aa_mass=aa_mass,
labels=labels), fragment_number=s + 1)
elif ion == "b":
position = s + 1
if position == seq_len:
continue
if b_stop_at > -1:
if position > b_stop_at:
continue
if b_selected:
if s in b_selected:
seq = temp[:position]
str_seq = parser.tostring(seq, False)
# print(str_seq)
yield Ion(seq=temp, ion_type=ion, charge=m + 1,
mz=mass.fast_mass2(str_seq, ion_type=ion, charge=m + 1, aa_mass=aa_mass,
labels=labels), fragment_number=s + 1)
else:
seq = temp[:position]
str_seq = parser.tostring(seq, False)
# print(str_seq)
yield Ion(seq=temp, ion_type=ion, charge=m + 1,
mz=mass.fast_mass2(str_seq, ion_type=ion, charge=m + 1, aa_mass=aa_mass,
labels=labels), fragment_number=s + 1)
def theor_spectrum(self, types=('b', 'y'), maxcharge=None, **kwargs):
peaks = {}
if not maxcharge:
maxcharge = max(self.pcharge, 1)
for ion_type in types:
ms = []
for i in range(1, len(self.modified_sequence)):
if self.modified_sequence[i - 1] in parser.std_amino_acids and self.modified_sequence[i] != '-':
for charge in range(1, maxcharge + 1):
if ion_type[0] in 'abc':
ms.append(mass.fast_mass2(
str(self.modified_sequence)[:i], ion_type=ion_type, charge=charge,
**kwargs))
else:
ms.append(mass.fast_mass2(
str(self.modified_sequence)[i:], ion_type=ion_type, charge=charge,
**kwargs))
marr = np.array(ms)
marr.sort()
peaks[ion_type] = marr
return peaks
def fragmenting(i,
rec,
ms,
mv,
query_unique,
protein_id,
unique_q3=None,
y=None):
labels, aa_mass, seq = prepare_libraries(i['_protein']['_sequence'],
static=ms,
variable=mv)
seq_len = len(seq)
result = []
msMap = {m['label']: m['m_label'] for m in ms}
for f in generate_fragments(seq[:], static=ms, variable=mv):
# print(f)
new_seq = parser.tostring(f, False)
print('Current: ' + new_seq)
precursor_mz = mass.fast_mass2(new_seq,
charge=i['_precursor_charge'],
aa_mass=aa_mass,
labels=labels)
variable = ''
for r in i['_rt']:
for ion in fragments_by(aa_mass,
i['_charge'],
i['_protein']['_ion_type'],
labels,
seq_len,
f,
mv,
Ytype=y,
b_stop_at=i['_b_stop_at'],
y_stop_at=i['_y_stop_at'],
by_static=i['_by_run'],
b_selected=i['_b_selected'],
y_selected=i['_y_selected']):
create_row(f, i, ion, msMap, mv, precursor_mz, query_unique, r,
rec, result, seq, seq_len, unique_q3, variable,
protein_id)
return result, query_unique
def test_fast_mass2(self):
for pep in self.random_peptides:
self.assertAlmostEqual(
mass.fast_mass2(pep, aa_mass=self.test_aa_mass),
sum(pep.count(aa) * m for aa, m in self.test_aa_mass.items()) +
self.mass_H * 2.0 + self.mass_O)