print >>sys.stderr, "Skip: Unsupported circular glycan"
# dodge this case...
continue
except UnsupportedMonoError, e:
print >>sys.stderr, "Skip:",e.message
continue
except GlycanParseError, e:
print >>sys.stderr, "Skip:",e.message
continue
if not gtopo or not gcomp:
continue
try:
g3 = glycoct_parser.toGlycan(zf1.read(acc+'.txt'))
except (KeyError, GlycoCTParseError):
g3 = None
if acc in ('G24172ZD',):
# this is a (1+1) two monosaccharide case in which the
# GlyTouCan topology has (-1+1). Other (1+1) two
# monosaccharide caes have GlyTouCan topology with (1+1) -
# and this second decision seems to be more common.
# Can't figure out how to decide whicoh one to do.
print >>sys.stderr, "Skip: (1+1) two monosaccharide issue"
continue
if False and acc in ('G64632PP',):
# the toplogy G83908ZR breaks my equality algorithm when compared to the
# topology generated from G64632PP - multiple valid id
class GlycoCTDatabase:
prefix = ""
extn = "gct"
source = ""
def __init__(self, filename):
self.filename = filename
self.name, extn = filename.rsplit('.', 1)
self.name = os.path.split(self.name)[1]
assert extn == self.extn
self.fmt = GlycoCTFormat()
def getraw(self, accession):
zf = zipfile.ZipFile(self.filename, "r")
fn = accession[len(self.prefix):] + '.txt'
try:
zf.getinfo(fn)
except KeyError:
return None
gct = self._getraw(zf, fn)
zf.close()
return gct
def _getraw(self, zf, name):
return zf.read(name)
def _get(self, zf, name):
try:
glystr = self._getraw(zf, name)
g = self.fmt.toGlycan(glystr)
except GlycoCTParseError:
# f.write('> GlycanFormatter > GLYCANPARSEERROR > '+self.message)###
# print '> GlycanFormatter > GLYCANPARSEERROR > ',self.message###
# time.sleep(2)###
return None
except:
print >> sys.stderr, "Problem with GlycoCT file " + name
# time.sleep(2)###
traceback.print_exc()
sys.exit(1)
kwargs = {}
try:
attstr = zf.read(name.rsplit('.', 1)[0] + '.att')
for r in csv.reader(StringIO(attstr)):
kwargs[r[0]] = copy.copy(r[1:])
except KeyError:
pass
try:
kwargs['image'] = zf.read(name.rsplit('.', 1)[0] + '.png')
except KeyError:
pass
gr = GlyRecord(source=self.source if self.source else self.name,
accession=self.prefix + name.rsplit('.', 1)[0],
glycan=g,
name=self.name,
**kwargs)
# print gr
return gr
def get(self, accession):
zf = zipfile.ZipFile(self.filename, "r")
fn = accession[len(self.prefix):] + '.txt'
try:
zf.getinfo(fn)
except KeyError:
return None
gr = self._get(zf, fn)
zf.close()
return gr
def __iter__(self):
return self.next()
def next(self):
zf = zipfile.ZipFile(self.filename, "r")
for name in zf.namelist():
if not name.endswith('.txt'):
continue
gr = self._get(zf, name)
if gr:
yield gr
zf.close()
class GlycanFactory(ReferenceTable):
def __init__(self):
self.fmt = GlycoCTFormat()
self.mf = MonoFactory()
super(GlycanFactory, self).__init__()
def new(self, key):
return self[key].clone()
def parseSection(self, name, kv):
aliases = [name]
g = self.fmt.toGlycan('\n'.join(kv['GlycoCT'].split()))
aliases.extend(map(str.strip, kv.get('Aliases', '').split(';')))
return [(a, g) for a in aliases]
def add_mono(self,
parent,
name,
parent_pos,
child_pos=1,
anomer=Anomer.beta,
parent_type=Linkage.oxygenPreserved,
child_type=Linkage.oxygenLost):
m = self.mf.new(name)
m.set_anomer(anomer)
parent.add_child(m,
parent_pos=parent_pos,
child_pos=child_pos,
parent_type=parent_type,
child_type=child_type)
return m
def oxford2Glycan(self, name):
if name in self:
return self.new(name)
p = 0
if name[p] == 'F':
g = self.new('FM3')
p += 1
else:
g = self.new('M3')
# print repr(g)
# print self.fmt.toStr(g)
# print self
r = g.root()
glcnac2 = filter(lambda m: m.compatible(self.mf['GlcNAc']),
r.children())[0]
man1 = glcnac2.children()[0]
man16 = [l.child() for l in man1.links() if l.parent_pos() == 6][0]
man13 = [l.child() for l in man1.links() if l.parent_pos() == 3][0]
assert name[p] == 'A'
nant = int(name[p + 1])
ant = [None]
if nant in (1, 2, 3, 4):
ant.append(self.add_mono(man13, 'GlcNAc', parent_pos=2))
if nant in (2, 3, 4):
ant.append(self.add_mono(man16, 'GlcNAc', parent_pos=2))
if nant in (3, 4):
ant.append(self.add_mono(man13, 'GlcNAc', parent_pos=4))
if nant in (4, ):
ant.append(self.add_mono(man16, 'GlcNAc', parent_pos=6))
p += 2
if p >= len(name):
return g
if name[p] == 'B':
b = self.add_mono(man1, 'GlcNAc', 4)
name[p] += 1
if p >= len(name):
return g
if name[p] == 'F':
nfuc = int(name[p + 1])
assert (nfuc <= nant)
for fi in range(1, nfuc + 1):
self.add_mono(ant[fi],
'Fuc',
parent_pos=6,
anomer=Anomer.alpha)
p += 2
if p >= len(name):
return g
assert (name[p] == 'G')
ngal = int(name[p + 1])
gal = [None]
assert (ngal <= nant)
for gi in range(1, ngal + 1):
gal.append(self.add_mono(ant[gi], 'Gal', parent_pos=4))
p += 2
if p >= len(name):
return g
assert (name[p] == 'S')
nsia = int(name[p + 1])
sia = [None]
assert (nsia <= ngal)
for si in range(1, nsia + 1):
sia.append(
self.add_mono(gal[si],
'Neu5Ac',
parent_pos=6,
child_pos=2,
anomer=Anomer.alpha))
return g