def iterlines(self):
"""Iterates standard lines."""
for subs, inst in self.subsproduct():
full_name = inst.getname()
chains = tuple(
s.attrs.chain for s in subs if hasattr(s.attrs, 'chain')
)
chainsum = lipproc.sum_chains(chains)
name = (
(lipproc.summary_str(self.hg, chainsum),)
if self.hg and self.sum_only else
(lipproc.full_str(self.hg, chains),)
if self.hg and not self.sum_only else
()
)
lab = lipproc.LipidLabel(
db_id = None,
db = 'lipyd.lipid',
names = name,
formula = inst.formula,
)
rec = lipproc.LipidRecord(
lab = lab,
hg = self.hg,
chainsum = chainsum if chainsum.c else None,
chains = () if self.sum_only else chains,
)
yield inst.mass, rec
def _getname(parent, subs):
"""
Parameters
----------
parent :
subs :
Returns
-------
"""
return (lipproc.summary_str(
self.hg,
lipproc.sum_chains(
tuple(s.attrs.chain for s in subs
if hasattr(s.attrs, 'chain'))))
if self.sum_only else lipproc.full_str(
self.hg,
tuple(s.attrs.chain for s in subs
if hasattr(s.attrs, 'chain'))))
def collect_scan_lines(mzs, rts, mgfs):
"""
Parameters
----------
mzs :
rts :
mgfs :
Returns
-------
"""
result = []
for mz, rt in zip(mzs, rts):
fe = ms2.MS2Feature(mz, ionmode = 'pos', mgfs = mgfs, rt = rt)
fe.build_scans()
for sc, drt in zip(fe.scans, fe.deltart):
ids = sc.identify()
smrec = list(
sc.get_ms1_records(hg = 'SM', databases = {'lipyd.lipid'})
)
chainid = set()
for sm in smrec:
ccomb = sc.chain_combinations(sm[0])
for cc in ccomb:
chainid.add(
'%s[full=%s,frags=[%s,%s],i=[%.03f,%.03f]]' % (
sm[0].summary_str(),
lipproc.full_str(sm[0].hg, cc[0]),
cc[1].fragtype[0],
cc[1].fragtype[1],
cc[1].i[0] * 100,
cc[1].i[1] * 100,
)
)
for ms1id, ms2ids in ids.items():
if ms2ids and ms2ids[0].hg.main == 'SM':
this_line = [
'%.07f' % mz,
'%.02f - %.02f' % rt,
'%u' % sc.scan_id,
'%.02f' % drt,
sc.sample,
ms1id,
'%u' % ms2ids[0].score_pct,
]
isph = min(
sc.fragments_by_chain_type(frag_type = 'Sph-2xH2O+H'),
default = None
)
irel_sph = (
'%.03f' % (sc.inorm[isph] * 100)
if isph is not None
else ''
)
this_line.append(irel_sph)
for frag in sm_frags:
i = sc.fragment_by_name(frag)
reli = (
'%.03f' % (sc.inorm[i] * 100)
if i is not None
else ''
)
this_line.append(reli)
this_line.append(';'.join(chainid))
result.append(this_line)
return result
def make_index(self):
def cc2str(cc):
return ('%s%s%u:%u' %
(cc[0], '-' if cc[0] in {'O', 'P'} else '', cc[1], cc[2]))
self.close_plainfile()
self.load()
self.index = collections.defaultdict(lambda: set([]))
self.hg_index = collections.defaultdict(lambda: set([]))
self.species_index = collections.defaultdict(lambda: set([]))
self.subspec_index = collections.defaultdict(lambda: set([]))
self.isomer_index = collections.defaultdict(lambda: set([]))
if not self.silent:
self.prg = progress.Progress(self._curl.size,
'Indexing SwissLipids', 101)
self._plainfilename = '%s.extracted' % self._gzfile.name
with open(self._plainfilename, 'wb') as fpp:
offset = self._gzfile.tell()
for l in self._gzfile:
if not self.silent:
self.prg.step(len(l))
ll = l.decode('utf-8').split('\t')
if ll[1] in self.levels:
names = self.names(ll)
self.index[ll[0]].add(offset) # SwissLipids ID
self.index[ll[8]].add(offset) # SMILES
self.index[ll[10]].add(offset) # InChI key
for n in names.split('|'):
self.index[n].add(offset)
hg, chainsum, chains = self.nameproc.process(names)
if hg:
self.hg_index[hg].add(offset)
if hg and chainsum:
self.species_index[lipproc.summary_str(
hg, chainsum)].add(offset)
if hg and chains:
self.subspec_index[lipproc.full_str(
hg, chains)].add(offset)
if hg and chains:
self.isomer_index[lipproc.full_str(
hg, chains, iso=True)].add(offset)
offset = self._gzfile.tell()
fpp.write(l)
if not self.silent:
self.prg.terminate()
self.index = dict(self.index)
self._plainfile = open(self._plainfilename, 'r')