from analysis.fisher import lod, fisher_exact_low, fisher_exact_high
from transitionspecificity import specscore
w = GPTWiki()
trspec = {}
labelspec = {}
glycanlabelspec = {}
smallwindow = 1
largewindow = 6
threshold = 50
nonspeccount = 0
alltrs = set()
for tg in w.itertgs(acqtype='DIA'):
pepid = tg.get('peptide')
pepage = w.get(pepid)
z1 = tg.get('z1')
spectra = tg.get('spectra')
if spectra.find('DIA') == -1:
continue
try:
glycan = re.search('\[H(.)*?\]', pepage.get('name')).group(0)[1:-1]
except:
continue
filename = pepid + '.' + str(z1) + '.50.json'
onlinefile = 'http://edwardslab.bmcb.georgetown.edu/~nedwards/dropbox/pBYmLSkGeq/' + spectra + '/' + filename
json_file = urllib.urlopen(onlinefile)
try:
continue
if '?' in glyspec:
continue
seen.add((seq, glyspec, modspec))
# print >>sys.stderr, seq,glyspec,modspec
glycan = []
if glyspec != "-":
glycan = map(lambda t: (t[1], seq[int(t[0]) - 1] + str(t[0])),
map(lambda s: s.split(':'), glyspec.split(',')))
badglys = set()
for glyacc in set(map(itemgetter(0), glycan)):
if not w.get(glyacc):
badglys.add(glyacc)
if len(badglys) > 0:
print >> sys.stderr, "Warning: Can't resolve glycan accession(s):", ", ".join(
sorted(badglys))
continue
alignments = []
for pracc in map(str.strip, praccs.split(',')):
if not w.get(pracc):
print >> sys.stderr, "Warning: Can't resolve protein accession:", pracc
continue
prseq = "".join(w.get(pracc).get('sequence').split())
nrtonly = True
w = GPTWiki()
seenpeps = set()
sites = set()
prot2site = defaultdict(set)
samples = set()
glycans = set()
glysites = set()
site2gly = defaultdict(set)
for sp in w.iterspec(type='DDA'):
for tg in w.itertgs(spectra=sp.get('name')):
tgid = tg.get('id')
if tg.get('peptide') in seenpeps:
continue
pep = w.get(tg.get('peptide'))
if nrtonly and pep.get('nrt') == None:
continue
seenpeps.add(pep.get('id'))
gly = pep.get('glycan')[0][0]
glycans.add(gly)
for al in pep.get('alignments', []):
site = al.get('prsites')
prot = al.get('protein')
print pep.get('id'), prot, site, gly
sites.add((prot, site))
prot2site[prot].add(site)
site2gly[(prot, site)].add(gly)
glysites.add((prot, site, gly))
print "Proteins:", len(prot2site)
z1=z1,
spectra=spectra,
mz1=mz1,
nrt=nrt,
rt=rt,
prt=rt,
transitions=trans,
ntransition=len(trans),
**extras)
if 'lccalibration' in data and spectra not in lccal:
nrtslope, nrtintercept = map(float,
data['lccalibration'].split(":"))
lccal[spectra]['nrtslope'] = nrtslope
lccal[spectra]['nrtintercept'] = nrtintercept
tgs[spectra].add(tg.get('id'))
for spectra in allspec:
spec = w.get(spectra)
if spectra in lccal:
spec.set("nrtslope", lccal[spectra]['nrtslope'])
spec.set("nrtintercept", lccal[spectra]['nrtintercept'])
else:
spec.delete("nrtslope")
spec.delete("nrtintercept")
w.put(spec)
for tg in w.itertgs(spectra=spectra, all=True):
if tg.get('id') not in tgs[spectra]:
if w.cleartransgroup(tg):
print "Clear", tg.get("id")
#!/bin/env python2
import sys
from collections import defaultdict
from getwiki import GPTWiki
w=GPTWiki()
tla = {'N': 'Asn'}
sampleids = set(sys.argv[1:])
prsites = defaultdict(dict)
for i,tg in enumerate(w.itertransgroups()):
pep = w.get(tg.get('peptide'))
# if not pep.get('nrt'):
# continue
glyid = pep.get('glycan')[0][0]
gly = w.get(glyid)
name = ""
for n in gly.get('name',[]):
if 'Fuc' in n:
continue
if 'HexNAc' in n:
name = n
break
cls = gly.get('class')
if len(cls) != 1:
continue
cls = cls[0]
if modified:
print t.get('id')
if (pid, z1, spectra) not in tgroup:
tgroup[(pid, z1, spectra)] = dict(transitions=[],
nrt=nrt,
rt=rt,
prt=prt,
mz1=mz1,
scans=scans,
gphash=gphash)
tgroup[(pid, z1, spectra)]['transitions'].append((t.get('id'), relint))
for pid, z1, spectra in tgroup:
tgroup[(pid, z1, spectra)]['ntransition'] = len(
tgroup[(pid, z1, spectra)]['transitions'])
tg, mod = w.addtransgroup(peptide=pid,
z1=z1,
spectra=spectra,
**tgroup[(pid, z1, spectra)])
if mod:
print tg.get('id')
if tg.get('id') in spectra2tg[spectra]:
spectra2tg[spectra].remove(tg.get('id'))
for spectra in allspectra:
for tgid in spectra2tg[spectra]:
tg = w.get(tgid)
if w.cleartransgroup(tg):
print "Clear", tg.get("id")
import sys, urllib, string, csv, os.path
from collections import defaultdict
import Bio.SeqIO
from util import peptide_mw, mod_mw
def asscan(s):
t = s.rstrip(')').split('(')
return [int(t[0])] + t[1].split(',')
w = GPTWiki()
spectra2tg = defaultdict(set)
for tgpage in w.iterpages(include_categories=['TransitionGroup']):
tg = w.get(tgpage.name)
spectra = tg.get('spectra')
spectra2tg[spectra].add(tg.get('id'))
allspectra = set()
for transfile in sys.argv[1:]:
spectra, sample, method, index, extn = transfile.rsplit('.', 4)
spectra = os.path.split(spectra)[1]
allspectra.add(spectra)
w.addacquisition(name=spectra, method=method, sample=sample)
tgroup = defaultdict(dict)
for l in csv.DictReader(open(transfile), dialect='excel-tab'):
seq = l['PeptideSequence']
glyspec = l['Glycans']
modspec = l['Mods']
def label2series(lab):
return lab[0].lower()
w = GPTWiki()
peps = defaultdict(lambda: defaultdict(dict))
for sp in w.iterspec(sample=opts.sample,
acqtype=opts.acqtype,
method=opts.method,
inst=opts.inst):
print >> sys.stderr, sp.get('name')
for i, tg in enumerate(w.itertgs(spectra=sp.get('name'))):
pep = w.get(tg.get('peptide'))
pepid = pep.get('id')
if pep.get('nrt') == None:
continue
z1 = tg.get('z1')
ntrans = len(tg.get('transitions', []))
if ntrans == 0:
continue
proteins = set()
for al in pep.get("alignments", []):
proteins.add(al.get("protein"))
for trid, trint in tg.get('transitions', []):
tr = w.get(trid)
if trid not in peps[(pepid, z1)]:
peps[(pepid, z1)][trid]['pepname'] = pep.get('name')
peps[(pepid, z1)][trid]['prot'] = ";".join(sorted(proteins))
#!/bin/env python27
from getwiki import GPTWiki, Peptide
import sys
from collections import defaultdict
w = GPTWiki()
currtrans = set()
for tgpage in w.iterpages(include_categories=['TransitionGroup']):
tg = w.get(tgpage.name)
tgid = tg.get('id')
for t, i in tg.get('transitions'):
tr = w.get(t)
if tr:
currtrans.add(t)
for tpage in w.iterpages(include_categories=['Transition']):
t = w.get(tpage.name)
tid = t.get('id')
if tid not in currtrans:
print >> sys.stderr, "Delete transition " + tid
w.delete(tid)
print 'please enter the spectra file name(s)'
exit(1)
spectrafiles = sys.argv[1:]
pepnrtpairs = defaultdict(list)
nrtobsgt0 = {}
for tgpage in w.itertransgroups():
spectra = tgpage.get('spectra')
if spectra not in spectrafiles:
continue
tgid = tgpage.get('id')
pepid = tgpage.get('peptide')
peppage = w.get(pepid)
pepnrt = peppage.get('nrt')
nrtobs = peppage.get('nrtobs')
peakrt = tgpage.get('prt')
if peakrt != None and pepnrt != None:
if nrtobs == '0':
continue
else:
if spectra not in nrtobsgt0:
nrtobsgt0[spectra] = 1
else:
nrtobsgt0[spectra] += 1
if spectra not in pepnrtpairs:
pepnrtpairs[spectra] = [(float(peakrt), float(pepnrt))]
towiki = sys.argv[2].upper()
assert fromwiki in ("PROD", "DEV", "TEST")
assert towiki in ("PROD", "DEV", "TEST")
assert fromwiki != towiki
w1 = GPTWiki(smwenv=fromwiki, quiet=True)
print >> sys.stderr, "from: %s" % (w1.title(), )
w2 = GPTWiki(smwenv=towiki, quiet=True)
print >> sys.stderr, " to: %s" % (w2.title(), )
dummy = raw_input("Enter to proceed, <Ctrl-C> to abort:")
currentids = set()
for page in w1.iterpages(include_categories=('Transition', 'Peptide',
'TransitionGroup')):
id = page.name
currentids.add(id)
it = w1.get(id)
if w2.put(it):
print >> sys.stderr, "Pushing %s to %s" % (id, w2.title())
else:
print >> sys.stderr, "No change to %s in %s" % (id, w2.title())
for page in w2.iterpages(include_categories=('Transition', 'Peptide',
'TransitionGroup')):
id = page.name
if id not in currentids:
w2.delete(id)
print >> sys.stderr, "Delete %s from %s" % (id, w2.title())
opts, args = parser.parse_args()
if opts.cachefile and os.path.exists(opts.cachefile):
data = json.loads(open(opts.cachefile).read())
rows = data['rows']
tgs = data['tgs']
origpepnrt = data['pepnrt']
else:
monos = "NHFS"
tgs = dict()
origpepnrt = dict()
rows = []
w = GPTWiki(quiet=True)
for tg in w.itertransgroups():
pepid = tg.get('peptide')
p = w.get(pepid)
pepacc = p.get('id')
pepseq = p.get('sequence')
pepname = p.get('name')
pepnrt = p.get('nrt')
if not tg.has('nrt'):
continue
nrt = float(tg.get('nrt'))
glyacc = p.get('glycan')[0][0]
g = w.get(glyacc)
gsym = g.get('sym')
mcnt = {}
for mono in monos:
mcnt[mono] = 0
m = re.search(mono + r'(\d+)', gsym)
if m:
#!/bin/env python27
from getwiki import GPTWiki, Peptide
import sys, urllib, string
import Bio.SeqIO
from util import peptide_mw, mod_mw
seen = set()
w = GPTWiki()
peps = []
for p in w.iterpages(include_categories=['Peptide']):
pep = w.get(p.name)
pepkey = Peptide.key(pep.get('sequence'), pep.get('glycan', []),
pep.get('mod', []))
if pepkey in seen:
print >> sys.stderr, p.name
w.delete(p.name)
seen.add(pepkey)
#!/bin/env python27
from getwiki import GPTWiki, Peptide
import sys, urllib, string
import Bio.SeqIO
print "\t".join(
map(str, [
"Spectra", "Accession", "Peptide", "Site", "Glycan", "TGAccession",
"Charge", "PrecursorMZ"
]))
seen = set()
w = GPTWiki()
for tgpage in w.iterpages(include_categories=['TransitionGroup']):
tg = w.get(tgpage.name)
pep = w.get(tg.get('peptide'))
pepid = pep.get('id')
z1 = tg.get('z1')
mz1 = tg.get('mz1')
spectra = tg.get('spectra')
if (spectra, pepid, z1) not in seen:
pepseq = list(pep.get('sequence'))
for deltastr, pos in pep.get('mod', []):
aa = pos[0]
pos = int(pos[1:]) - 1
if round(deltastr, 3) == 57.021:
pepseq[pos] += ":m"
elif round(deltastr, 3) in (15.995, 15.996):
pepseq[pos] += ":o"
pepseq = "".join(pepseq)
