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 idx_from_name(
self,
name=None,
hg=None,
chainsum=None,
chains=(),
):
"""
For a lipid name returns the indices of the matching records.
Returns
-------
Set of indices. If name not in the database returns None.
"""
if name is None and hg is not None:
if chainsum is None:
chainsum = lipproc.sum_chains(chains)
name = lipproc.summary_str(hg=hg, chainsum=chainsum)
if name in self.names:
return self.names[name]
def fa_greek_cc(self, name):
"""
From the greek name of a fatty acyl, fatty alcohol or fatty acyl CoA
returns its carbon and unsaturation count.
Parameters
----------
name : str
A greek name.
Returns
-------
``ChainSummary`` and ``Chain`` objects with the carbon and
unsaturation counts.
"""
chainsum, chains = None, None
try:
name1 = name.split('-')[1] if '-' in name else name
for typ in {'FA', 'FAL', 'FACoA'}:
if name1 in getattr(self, '%s_greek' % typ.lower()):
cc1 = getattr(self, '%s_greek' % typ.lower())[name1]
iso = (tuple(name.split(')')[0][1:].split(','))
if self.iso and '(' in name else ())
chains = [
lipproc.Chain(c=cc1[0], u=cc1[1], typ=typ, iso=iso)
]
chainsum = lipproc.sum_chains(chains)
except:
pass
return chainsum, chains
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 carbon_counts(
self,
name,
ccexp=2,
chainsexp=None,
sphingo=False,
iso=False,
types=None,
accept_less_chains=False,
):
"""
Processes carbon and unsaturation counts from name.
Parameters
----------
str :
name:
Lipid name.
int :
ccexp:
Expected number of fatty acyl or other residues constaining
aliphatic chain. E.g. for DAG this should be 2 and for TAG 3
as the latter has 3 fatty acyls.
tuple :
chainsexp:
The type of the expected chains, e.g. `('Sph', 'FA')` for one
sphingosine and a fatty acyl. This matters only if the chain
types can not be inferred from the processed names.
bool :
sphingo:
Is this a sphingolipid, is the first chain a sphingosine base?
bool :
iso:
Process conformation isomer details for example 18:2(9E,11Z).
tuple :
types:
Explicit types for each chains.
name :
ccexp :
(Default value = 2)
chainsexp :
(Default value = None)
sphingo :
(Default value = False)
iso :
(Default value = False)
types :
(Default value = None)
"""
# number of groups in one regex match unit
_g = 5 if iso else 4
_i = 3
rechain = lipproc.rechainiso if iso else lipproc.rechain
# regex finds the total carbon count
cc1 = lipproc.rechainsum.search(name)
cc1 = cc1.groups() if cc1 else cc1
# regex finds 1-4 fatty acids
cc2 = rechain.findall(name)
if not cc2:
# if no result do an attempt with the isomeric regex
rechain = lipproc.rechainiso
cc2 = rechain.findall(name)
if cc2:
_g = 5
chains = []
if (ccexp and cc2 #and
#cc2[(ccexp - 1) * _g + 1] and
#cc2[(ccexp - 1) * _g + 2]
):
# in case of e.g. cardiolipin we have multiple groups of 2 chains
for _cc2 in cc2:
for i in xrange(len(_cc2) // _g):
if _cc2[i * _g + 1] and _cc2[i * _g + 2]:
c = int(_cc2[i * _g + 1])
u = int(_cc2[i * _g + 2])
attr = self.attr_proc(_cc2[i * _g:i * _g + _g], u)
# in lipidmaps one unsaturation
# for plasmalogens is implicit
if (self.database == 'lipidmaps'
and _cc2[i * _g] == 'P'):
u += 1
sphingo = sphingo or bool(attr.sph)
chains.append(
lipproc.Chain(
c=c,
u=u,
attr=attr,
typ=self.get_type(i, sphingo, attr.ether,
types, chainsexp),
# even if we used the isomeric regex
# we include the conformational isomer
# information only if requested
iso=(tuple(_cc2[i * _g + _i].split(','))
if iso and _cc2[i * _g + _i] else
()) if iso else None))
zerochains = sum(not c.c for c in chains)
# ccexp = ccexp - zerochains
chains = tuple(c for c in chains if c.c)
chains = (() if (
len(chains) != ccexp and
(not accept_less_chains or len(chains) < accept_less_chains))
else tuple(chains))
# the total carbon count
if chains:
chainsum = lipproc.sum_chains(chains)
elif cc1:
c = int(cc1[1])
u = int(cc1[2])
if not chainsexp:
attrs = (self.attr_proc(cc1, u), )
types = ()
else:
attrs, types = self.attrs_types(cc1, chainsexp)
chainsum = lipproc.ChainSummary(
c=c,
u=u,
attr=attrs,
typ=types,
)
else:
chainsum = None
return chainsum, chains