def __init__(
self,
protein,
id_type='uniprot',
ncbi_tax_id=9606,
typ='unknown',
motif=None,
residue=None,
isoform=1,
evidences=None,
seq=None,
):
self.non_digit = re.compile(r'[^\d.-]+')
self.protein = (protein
if hasattr(protein, 'identifier') else entity.Entity(
identifier=protein,
id_type=id_type,
taxon=ncbi_tax_id,
))
self.id_type = id_type
self.typ = typ.lower()
self.seq = seq
self.motif = motif
self.residue = residue
self.isoform = (isoform if type(isoform) is int else int(
self.non_digit.sub('', isoform)))
self.isoforms = set()
self.add_isoform(isoform)
self.evidences = evidence.Evidences()
self.add_evidences(evidences)
def __init__(self, domain, ptm, evidences = None, pdbs = None):
self.ptm = ptm
self.domain = domain
self.pdbs = set()
self.pnetw_score = None
self.add_pdbs(pdbs)
self.evidences = evidences or evidence.Evidences()
def __init__(
self,
protein,
start,
end,
id_type = 'uniprot',
ncbi_tax_id = 9606,
regex = None,
instance = None,
isoform = 1,
motif_name = None,
prob = None,
elm = None,
description = None,
seq = None,
evidences = None,
):
non_digit = re.compile(r'[^\d.-]+')
self.protein = (
protein
if hasattr(protein, 'identifier') else
entity.Entity(
protein,
id_type = id_type,
taxon = ncbi_tax_id,
)
)
self.id_type = id_type
self.seq = seq
self.isoform = (
isoform
if isinstance(isoform, int) else
int(non_digit.sub('', isoform))
)
self.start = (
start
if not isinstance(start, common.basestring) else
int(non_digit.sub('', start))
)
self.end = (
end
if not isinstance(end, common.basestring) else
int(non_digit.sub('', end))
)
self.regex = None if regex is None else re.compile(regex)
self.instance = instance
self.motif_name = motif_name
self.prob = prob
self.elm = elm
self.description = description
self.evidences = evidence.Evidences()
self.add_evidences(evidences)
def _process(self, p):
# human leukocyte antigenes result a result an
# extremely high number of combinations
if (not p['kinase'] or (isinstance(p['substrate'], common.basestring)
and p['substrate'].startswith('HLA'))):
return
if not isinstance(p['kinase'], list):
p['kinase'] = [p['kinase']]
kinase_ups = mapping.map_names(
p['kinase'],
self.id_type_enzyme,
'uniprot',
ncbi_tax_id=self.ncbi_tax_id,
)
substrate_ups_all = set()
for sub_id_type in self.id_type_substrate:
if isinstance(sub_id_type, (list, tuple)):
sub_id_type, sub_id_attr = sub_id_type
else:
sub_id_attr = 'substrate'
substrate_ups_all.update(
set(
mapping.map_name(
p[sub_id_attr],
sub_id_type,
'uniprot',
self.ncbi_tax_id,
)))
# looking up sequences in all isoforms:
substrate_ups = []
for s in substrate_ups_all:
if 'substrate_isoform' in p and p['substrate_isoform']:
substrate_ups.append((s, p['substrate_isoform']))
else:
se = self.get_seq(s)
if se is None:
continue
for isof in se.isoforms():
if 'instance' in p and p['instance'] is not None:
if se.match(
p['instance'],
p['start'],
p['end'],
isoform=isof,
):
substrate_ups.append((s, isof))
else:
if se.match(
p['resaa'],
p['resnum'],
isoform=isof,
):
substrate_ups.append((s, isof))
if self.trace:
if p['substrate'] not in self.sub_ambig:
self.sub_ambig[p['substrate']] = substrate_ups
for k in p['kinase']:
if k not in self.kin_ambig:
self.kin_ambig[k] = kinase_ups
# generating report on non matching substrates
if len(substrate_ups) == 0:
for s in substrate_ups_all:
se = self.get_seq(s[0])
if se is None:
continue
self.nomatch.append((
s[0],
s[1],
(
p['substrate_refseq']
if 'substrate_refseq' in p else '',
s,
p['instance'],
se.get(p['start'], p['end']),
),
))
# building objects representing the enzyme-substrate interaction(s)
if 'typ' not in p:
p['typ'] = 'phosphorylation'
_resources = tuple(
(self.input_param.
get_via(name) if hasattr(self.input_param, 'get_via') else name)
for name in (p['databases'] if 'databases' in p else ()))
_resources += ((self.name, ) if isinstance(
self.input_param, common.basestring) else (self.input_param, ))
# collecting the evidences
evidences = evidence.Evidences(
evidence.Evidence(resource=_res,
references=p['references'] if 'references' in
p else None) for _res in _resources)
for s in substrate_ups:
# building the objects representing the substrate
se = self.get_seq(s[0])
if se is None:
continue
res = intera.Residue(
p['resnum'],
p['resaa'],
s[0],
isoform=s[1],
ncbi_tax_id=self.ncbi_tax_id,
)
if 'instance' not in p or p['instance'] is None:
reg = se.get_region(
p['resnum'],
p['start'] if 'start' in p else None,
p['end'] if 'end' in p else None,
isoform=s[1],
)
if reg is not None:
p['start'], p['end'], p['instance'] = reg
mot = intera.Motif(
s[0],
p['start'],
p['end'],
instance=p['instance'],
isoform=s[1],
ncbi_tax_id=self.ncbi_tax_id,
)
ptm = intera.Ptm(
s[0],
motif=mot,
residue=res,
typ=p['typ'],
evidences=evidences,
isoform=s[1],
ncbi_tax_id=self.ncbi_tax_id,
)
for k in kinase_ups:
if (not self.allow_mixed_organisms
and (self.get_taxon(k) != self.ncbi_tax_id
or self.get_taxon(s[0]) != self.ncbi_tax_id)):
continue
# the enzyme (kinase)
dom = intera.Domain(
protein=k,
ncbi_tax_id=self.ncbi_tax_id,
)
dommot = intera.DomainMotif(
domain=dom,
ptm=ptm,
evidences=evidences,
)
if hasattr(self.input_param, 'extra_attrs'):
for attr, key in iteritems(self.input_param.extra_attrs):
if key in p:
setattr(dommot, attr, p[key])
yield dommot