def headgroup_from_lipid_name(self, names, database=None):
"""For one database record attempts to identify the lipid class
by looking up keywords.
Calls greek name identification, greek fatty acid names are
identified as 'FA'.
Returns tuple of `lipproc.Headgroup` object and expected chain types.
Parameters
----------
names :
database :
(Default value = None)
Returns
-------
"""
if database is not None:
database = database.lower()
else:
database = self.database
names = '|'.join(names)
db = 'lmp' if database == 'lipidmaps' else 'swl'
for lipclass, spec in iteritems(self.lipnames):
for kwset in spec[db]:
matched = [kw in names for kw in kwset['pos']]
if sum(matched) == len(kwset['pos']) and \
sum(matched) > 0:
matched = [kw in names for kw in kwset['neg']]
if sum(matched) == 0:
return (
lipproc.Headgroup(
main=lipclass[1], # main class, e.g. Cer
sub=lipclass[0] # subclass, e.g. Hex
),
spec['chains'])
fa_name = self.process_fa_name(names)
if fa_name:
return (lipproc.Headgroup(main=fa_name), (fa_name, ))
return None, None
def test_aggregator_adduct_lookup(self):
""" """
result = self.mda.adduct_lookup(728.605042778354,
ionmode='pos')['[M+H]+'][1]
assert 'SLM:000391523' in [i.lab.db_id for i in result]
swl_cer1p = [i for i in result if i.lab.db_id == 'SLM:000391523'][0]
assert 'Ceramide phosphate (d42:2)' in swl_cer1p.lab.names
assert '' not in swl_cer1p.lab.names
assert '1P' in swl_cer1p.hg.sub
lyp_cer1p = lipproc.LipidRecord(
lab=lipproc.LipidLabel(
db_id=None,
db='lipyd.lipid',
names=('Cer-1P(DH42:2)', ),
formula='C42H82N1O6P1',
),
hg=lipproc.Headgroup(main='Cer', sub=('1P', )),
chainsum=lipproc.ChainSummary(
c=42,
u=2,
typ=('Sph', 'FA'),
attr=(lipproc.ChainAttr(sph='DH', ether=False, oh=()),
lipproc.ChainAttr(sph='', ether=False, oh=())),
),
chains=(),
)
assert lyp_cer1p in list(result)
def __init__(self,
sn1='H',
sn2='H',
sn3='H',
charge=None,
pcharge=0,
ncharge=0,
hg=None,
name='Glycerol',
typ='G',
**kwargs):
self.pcharge = pcharge
self.ncharge = ncharge
self.netcharge = (charge if charge is not None else self.pcharge -
self.ncharge)
self.typ = typ
hg = hg or lipproc.Headgroup(main=name)
metabolite.AbstractMetabolite.__init__(self,
core='C3H5O3',
subs=[
self.get_substituent(sn1),
self.get_substituent(sn2),
self.get_substituent(sn3)
],
charge=self.netcharge,
name=name,
hg=hg,
**kwargs)
def __init__(self,
c=None,
u=None,
fa_counts=None,
getname=None,
sub=(),
**kwargs):
def _getname(parent, subs):
"""
Parameters
----------
parent :
subs :
Returns
-------
"""
return '%s(%u:%u)' % (parent.name, subs[0].c, subs[0].u)
metabolite.AbstractMetabolite.__init__(
self,
core='OH',
subs=[substituent.FattyAcyl(c=c, u=u, counts=fa_counts)],
name='FA',
hg=lipproc.Headgroup(main='FA', sub=()),
getname=getname or _getname,
**kwargs)
def __init__(self, **kwargs):
AbstractGPL.__init__(self,
headgroup='C2H4NH2',
name='PE',
typ='PE',
hg=lipproc.Headgroup('PE'),
pcharge=1,
ncharge=1,
**kwargs)
def __init__(self,
sn1_1=None,
sn2_1=None,
sn1_2=None,
sn2_2=None,
sn1_1_fa_args=None,
sn1_2_fa_args=None,
sn2_1_fa_args=None,
sn2_2_fa_args=None,
fa_args=None,
charge=None,
pcharge=0,
ncharge=0,
name='CL',
typ='CL',
sub=(),
**kwargs):
"""
Example: http://swisslipids.org/#/entity/SLM:000528939/
>>> list(lipid.Cardiolipin(fa_args = {'c': 19, 'u': 2}))[0].mass
Exact mass: 1505.0348
"""
self.fa_args = fa_args or {}
sn1_1_fa_args = self.fa_args or {}
sn1_2_fa_args = self.fa_args or {}
sn2_1_fa_args = self.fa_args or {}
sn2_2_fa_args = self.fa_args or {}
self.pcharge = pcharge
self.ncharge = ncharge
self.netcharge = (charge if charge is not None else self.pcharge -
self.ncharge)
self.typ = typ
hg = lipproc.Headgroup(main=name, sub=sub)
metabolite.AbstractMetabolite.__init__(
self,
core='C9H18O13P2',
subs=[
self._get_substituent(sn1_1, args=sn1_1_fa_args),
self._get_substituent(sn1_2, args=sn1_2_fa_args),
self._get_substituent(sn2_1, args=sn2_1_fa_args),
self._get_substituent(sn2_2, args=sn2_2_fa_args),
],
charge=self.netcharge,
name=name,
hg=hg,
**kwargs,
)
def get_hg_obmol(self, hg, sub=()):
"""
Parameters
----------
hg :
sub :
(Default value = ())
Returns
-------
"""
if isinstance(hg, common.basestring):
hg = lipproc.Headgroup(hg, sub=sub)
return self.get_obmol(hg, index='hg')
def __init__(self,
c=None,
u=None,
fa_counts=None,
getname=None,
sub=(),
**kwargs):
def _getname(parent, subs):
return '%s(%u:%u)' % (parent.name, subs[0].c, subs[0].u)
metabolite.AbstractMetabolite.__init__(
self,
core='C19OH31',
subs=[substituent.FattyAcyl(c=c, u=u, counts=fa_counts)],
name='SE',
hg=lipproc.Headgroup(main='SE', sub=()),
getname=getname or _getname,
**kwargs,
)
def __init__(self,
core='',
sph_args=None,
fa_args=None,
o='H',
lyso=False,
fa_hydroxy=False,
lcb_type='d',
name='Cer',
hg=None,
getname=None,
**kwargs):
sph_args = sph_args or {}
fa_args = fa_args or {}
if lyso:
fa = 'H'
elif fa_hydroxy:
fa = substituent.HydroxyFattyAcyl(**fa_args)
else:
fa = substituent.FattyAcyl(**fa_args)
self.fa_hydroxy = fa_hydroxy
hg = hg or lipproc.Headgroup(main='Cer')
AbstractSphingolipid.__init__(self,
o=o,
n=fa,
sph_args=sph_args,
name=name,
getname=getname,
lcb_type=lcb_type,
hg=hg,
**kwargs)
def __init__(self,
c=None,
u=None,
fa_counts=None,
getname=None,
sub=(),
**kwargs):
def _getname(parent, subs):
"""
Parameters
----------
parent :
subs :
Returns
-------
"""
return 'Ret'
metabolite.AbstractMetabolite.__init__(
self,
core='C19H26',
subs=(metabolite.AbstractSubstituent(
cores=('CH2OH', ),
c=(0, ),
u=(0, ),
), ),
name='Ret',
hg=lipproc.Headgroup(main='Ret', sub=('Ol', )),
getname=getname or _getname,
**kwargs)
'[M+HCOO]-',
},
},
'pos': {
1: {
'[M+H]+',
'[M+NH4]+',
'[M+Na]+',
},
}
},
# additional constraints for adduct lookups at various species
# e.g. by default `[M-H2O+H]+` is not used but at Vitamin A we use it:
'adduct_constraints': {
'pos': {
lipproc.Headgroup(main = 'VA'): {
'[M+H]+',
'[M+NH4]+',
'[M+Na]+',
'[M-H2O+H]+',
},
lipproc.Headgroup(main = 'Cer', sub = ('Hex',)): {
'[M+H]+',
'[M+NH4]+',
'[M+Na]+',
'[M-H2O+H]+',
},
lipproc.Headgroup(main = 'Cer', sub = ('Hex2',)): {
'[M+H]+',
'[M+NH4]+',
'[M+Na]+',
def process(self, names, database=None, iso=None):
"""The main method of this class. Processes a lipid name string
and returns a standard name, prefix, carbon counts and
unsaturations.
Args
----
Parameters
----------
list :
names:
One or more names to process. Single result will be returned
and names will be attempted to be processed one after the other
until processing is successful.
names :
database :
(Default value = None)
Returns
-------
"""
iso = iso if iso is not None else self.iso
if hasattr(names, 'lower'):
# ok, if one passes a string let us still process it
names = (names, )
database = database or self.database
hg, chainsum, chains, chainsiso, chainsexp = (None, None, None, None,
None)
hg, chainsexp = self.headgroup_from_lipid_name(names,
database=database)
hg_modified = None
# try greek fatty acyl carbon counts:
if not hg and iso and database == 'swisslipids':
try:
for name0 in names:
fa_greek = name0.split('-')
if len(fa_greek) > 1:
hg = self.process_fa_name(fa_greek[1])
if hg:
hg = lipproc.Headgroup(main=fa_name)
chainsexp = (fa_name, )
break
except:
pass
for n in names:
lyso = hg and 'Lyso' in hg.sub
# how many aliphatic chains this molecule has
ccexp = (
2
if not hg else 1 if hg.main in {'FA', 'MAG'} or lyso else 3 if
hg.main == 'TAG' else 4 # cardiolipin has 4 aliphatic chains,
# but lyso or monolyso have 2 or 3,
# respectively
if hg.main == 'CL' else 2)
# for CL accept also 2 or 3 chains
accept_less_chains = 2 if hg and hg.main in {'CL'} else False
_chainsum, _chains = self.carbon_counts(
n,
ccexp=ccexp,
chainsexp=chainsexp,
iso=iso,
accept_less_chains=accept_less_chains,
)
chains = chains or _chains
chainsum = chainsum or _chainsum
if self.iso and _chains and any(c.iso for c in _chains):
chains = _chains
# TODO:
# this part I turned off, CL name processing should be
# imporoved later
if False and hg and hg.main == 'CL':
if _chains and len(_chains) == 2:
hg_modified = lipproc.Headgroup(
main='CL',
sub=('Lyso', ),
)
elif len(_chains) == 3:
hg_modified = lipproc.Headgroup(
main='CL',
sub=('Monolyso', ),
)
if (chainsum and chains
and (not iso or any(c.iso for c in chains))):
break
if hg and hg.main in {'FA', 'FAL', 'FACoA'} and not chainsum:
for name0 in names:
chainsum, chains = self.fa_greek_cc(name0)
if chainsum:
break
return hg_modified or hg, chainsum, chains
def __init__(self,
core='',
o='H',
n='H',
sph=None,
sph_args=None,
charge=None,
pcharge=None,
ncharge=None,
lcb_type='d',
name='Sphingolipid',
subtype=(),
hg=None,
getname=None,
**kwargs):
sph_args = sph_args or {}
self.pcharge = pcharge or 0
self.ncharge = ncharge or 0
self.charge = (charge if type(charge) is int else self.pcharge -
self.ncharge)
self.lcb_type = lcb_type
self.hg = hg or lipproc.Headgroup(main=name, sub=subtype)
if not sph:
# this is necessary because we have only 8:0 standard for kSph
# TODO: find a better solution for this
if (lcb_type == 'k'
and ('keto8' not in sph_args or sph_args['keto8'])):
if 'c' in sph_args:
sph_args['c'] = ((8, sph_args['c'])
if type(sph_args['c']) is int else
(8, sph_args['c'][1]))
else:
sph_args['c'] = (8, 20)
sph = substituent.Sphingosine(lcb_type=lcb_type, **sph_args)
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'))))
metabolite.AbstractMetabolite.__init__(
self,
core='',
subs=[sph, self.get_substituent(n),
self.get_substituent(o)],
name=name,
hg=hg,
charge=self.charge,
getname=getname or _getname,
**kwargs)
