def importFile(self):
def convert(s):
if len(s) == 1: return s, s
elif len(s) == 2: return s[0], s[1]
""" Merges duplicate samples """
self.ped = libPed.Ped()
head, firstline = False, True
for line in self.fgeno:
l = line.strip()
if firstline:
l = l.strip(',').split(',')
self.mark = libMark.Mark(l)
firstline = False
else:
l = l.split(',')
animal, genolist, genos = l[0], l[1:], []
for el in genolist:
if len(el) == 1: el = el + el
if len(el) == 0:
sys.stderr.write('Error in importFile: %s\n' %
(animal))
sys.exit(1)
genos.append(el[0])
genos.append(el[1])
self.addGenotype(animal, genos)
self.ped.addAnimal(animal, '0', '0', 'F0', '3')
self.ped.updateSex()
def importFile(self):
""" Merges duplicate samples """
self.ped = libPed.Ped()
firstline = True
for line in self.fgeno:
if line.strip().startswith('#'):
if firstline:
line = line.strip('#').strip().split()
self.mark = libMark.Mark(line)
firstline = False
continue
l = line.strip().split()
if len(l) == len(
self.mark) + 1 or len(l) == len(self.mark) * 2 + 1:
animal, sire, dam, geno = l[0], '0', '0', l[1:]
elif len(l) == len(
self.mark) + 3 or len(l) == len(self.mark) * 2 + 3:
animal, sire, dam, geno = l[0], l[1], l[2], l[3:]
elif len(l) == 0:
pass
else:
sys.stdout.write(
'Found %d genotypes for %s, expected %d markers\n' %
(len(l) - 3, l[0], len(self.mark)))
continue
#sys.exit(1)
newgeno = []
# This part is to fix files where the 2 alleles have been combined into one
if len(geno) == len(self.mark):
for a in geno:
if len(a) == 2:
newgeno += [trans(a[0])] + [trans(a[1])]
elif len(a) == 1:
newgeno += [trans(a)] + [trans(a)]
elif len(a) == 4: # Includes either 'DEL' or 'INS'
if a[:3] == 'DEL': newgeno += ['D'] + [trans(a[3])]
elif a[1:] == 'DEL': newgeno += [trans(a[0])] + ['D']
elif a[:3] == 'INS': newgeno += ['I'] + [trans(a[3])]
elif a[1:] == 'INS': newgeno += [trans(a[0])] + ['I']
else:
sys.stderr.write(
'Error in importing, unknown allele: %s\n' % a)
sys.exit(1)
elif len(a) == 6:
if a == 'DELDEL': newgeno += ['D'] + ['D']
elif a == 'DELINS': newgeno += ['D'] + ['I']
elif a == 'INSDEL': newgeno += ['I'] + ['D']
elif a == 'INSINS': newgeno += ['I'] + ['I']
else:
sys.stderr.write(
'Error in importing, unknown allele: %s\n' % a)
sys.exit(1)
geno = newgeno
self.ped.addAnimal(animal, dam, sire, 'F0', '3')
self.addGenotype(animal, geno)
self.ped.updateSex()
def importFile(self):
""" Merges duplicate samples """
print "This option is not available at this point in time"
sys.exit(1)
self.ped = libPed.Ped()
self.mark = libMark.Mark()
self.ped.addAnimal(animal, dam, sire, family, sex)
self.addGenotype(animal, geno.split())
self.ped.updateSex()
def importFile(self):
""" Merges duplicate samples """
self.ped = libPed.Ped()
self.mark = libMark.Mark()
marks = {}
genos = {}
reading = False
firstline = True
self.gcdict = {}
for line in self.fgeno:
l = line.strip().split(',')
if line.startswith('#'): continue
if len(l) == 6:
(animal, marker, a, t1, gc,
t3) = l[0], l[1], l[2], l[3], l[4], l[5]
elif len(l) < 4:
continue
else:
sys.stderr.write('Unexpected number of elements: %s \n' %
line.strip())
sys.exit(1)
if animal == 'Sample': continue
if a == 'NA': a1, a2 = '0', '0'
elif len(a) == 1: a1, a2 = a, a
elif a == 'DEL': a1, a2 = 'D', 'D'
elif a == 'INS': a1, a2 = 'I', 'I'
elif '.' in a:
try:
a1, a2 = a.split('.')
except ValueError:
sys.stderr.write('Unknown allele %s\n' % a)
sys.exit(1)
if a2 == 'DEL': a2 = 'D'
if a1 == 'DEL': a1 = 'D'
if a2 == 'INS': a2 = 'I'
if a1 == 'INS': a1 = 'I'
elif len(a) == 2: a1, a2 = a[0], a[1]
else:
sys.stderr.write('Unknown allele %s\n' % a)
sys.exit(1)
if animal not in genos: genos[animal], self.gcdict[animal] = {}, {}
if marker not in marks: marks[marker] = 1
genos[animal][marker] = [a1, a2]
self.gcdict[animal][marker] = gc
for mark in marks:
self.mark.addMarker(mark, '99')
for animal in genos:
gen = []
for mark in marks:
gen += genos[animal][mark]
self.ped.addAnimal(animal, '0', '0', 'F0')
self.addGenotype(animal, gen)
def pedcheck(options):
#if options.genotypefile[-3:0] == '.gz': fin = gzip.open(options.genotypefile,'r')
#else: fin = open(options.genotypefile,'r')
#***** Check requirements and read data *****
if options.pedigree: pedigree = libPed.Ped(options.pedigree)
else:
print "Gathering pedigree from data"
pedigree = libGeno.extractPedigree(options.genofile)
#sys.stderr.write('Pedigree file needed.\n')
#sys.exit(1)
if options.markers: markers = libMark.Mark(options.markers)
else:
print "Gathering markers from data"
markers = libGeno.extractBglMark(options.genofile)
#sys.stderr.write('Marker file needed.\n')
#sys.exit(1)
checkAll, checkOrphans = False, False
if ',' in options.pedlims: lim = options.pedlims.split(',')
else: lim = (options.pedlims, options.pedlims)
oldLim, newLim = float(lim[0]), float(lim[1])
r = {}
hits = {}
count = 0
for anim in pedigree.getAnimals():
r[anim] = count
count += 1
#****** Reading data, converting if needed
gen2 = None
fout = None
if options.reportfile: fout = open(options.reportfile, 'w')
fin = open(options.genofile, 'r')
rows = len(pedigree)
columns = len(markers)
gen = np.zeros((rows, columns))
marklist = fin.next().strip('#').strip().split()
newanims = {}
for line in fin:
l = line.strip().split()
icolumn = 0
irow = r[l[0]]
newanims[l[0]] = 1
for i in xrange(0, len(marklist) * 2, 2):
a1, a2 = l[i + 3:i + 5]
try:
m1, m2 = markers.getMAlleles(marklist[icolumn])
except KeyError:
if marklist[icolumn] not in markers.getMarkers():
print "WARNING! Incomplete markerlist"
a1, a2 = '0', '0'
else:
print "ERROR! Failed to assign marker alleles", marklist[
icolumn]
sys.exit(1)
if m1 == m2 or '0' in m1 + m2: gen[irow, icolumn] = 1
else:
gen[irow, icolumn] = tbase012(a1, a2, tbasenum(m1),
tbasenum(m2))
icolumn += 1
fin.close()
if options.genofile2:
fin = open(options.genofile2, 'r')
for line in fin:
if line.startswith('#'): continue
l = line.strip().split()
irow = r[l[0]]
icolumn = 0
for i in xrange(0, len(marklist) * 2, 2):
a1, a2 = l[i + 3:i + 5]
try:
m1, m2 = markers.getMAlleles(marklist[icolumn])
except KeyError:
if marklist[icolumn] not in markers.getMarkers():
print "WARNING! Incomplete markerlist"
a1, a2 = '0', '0'
else:
print "ERROR! Failed to assign marker alleles"
sys.exit(1)
if m1 == m2 or '0' in m1 + m2: gen[irow, icolumn] = 1
else:
gen[irow, icolumn] = tbase012(a1, a2, tbasenum(m1),
tbasenum(m2))
icolumn += 1
irow += 1
fin.close()
out = ''
sep = '\t'
if fout:
fout.write(
'#ID\tparent\tdiscords\tinfo_sites\tdiscord%\tcategory_sex\n')
for anim in pedigree.getAnimals():
if anim not in newanims: continue
mismatch = False
sire, dam = pedigree.getSire(anim), pedigree.getDam(anim)
if sire != '0' and sire in r and not checkAll:
res = gen[r[sire], :] - gen[r[anim], :]
wrong = len(res[res == 2]) + len(res[res == -2])
identical = len(res[res == 0])
info = len(res) - len(res[np.isnan(res)])
if info == 0:
out += anim + sep + sire + sep + str(wrong) + sep + str(
info) + sep + '-1' + sep + '1' + '\n'
else:
out += anim + sep + sire + sep + str(wrong) + sep + str(
info) + sep + str(wrong * 100.0 / info) + sep + '1' + '\n'
if info == 0: pass
elif wrong * 100.0 / info > oldLim:
mismatch = True
if fout:
fout.write(
'%s\t%s\t%d\t%d\t%.3f\t%s\n' %
(anim, sire, wrong, info, wrong * 100.0 / info, 'W1'))
hits[anim, sire] = wrong, info
if dam != '0' and dam in r and not checkAll:
res = gen[r[dam], :] - gen[r[anim], :]
wrong = len(res[res == 2]) + len(res[res == -2])
identical = len(res[res == 0])
info = len(res) - len(res[np.isnan(res)])
if info == 0:
out += anim + sep + dam + sep + str(wrong) + sep + str(
info) + sep + '-1' + sep + '0' + '\n'
else:
out += anim + sep + dam + sep + str(wrong) + sep + str(
info) + sep + str(wrong * 100.0 / info) + sep + '0' + '\n'
if info == 0: pass
elif wrong * 100.0 / info > oldLim:
mismatch = True
if fout:
fout.write(
'%s\t%s\t%d\t%d\t%.3f\t%s\n' %
(anim, dam, wrong, info, wrong * 100.0 / info, 'W0'))
hits[anim, dam] = wrong, info
if sire == '0' and dam == '0' and len(
pedigree.getOffspring(anim)) == 0:
mismatch = True
if mismatch: # Search for better matches
for anim2 in pedigree.getAnimals():
if anim2 == anim: continue
sex = pedigree.getSex(anim2)
if (anim, anim2) in hits:
wrong, info = hits[anim, anim2]
rep = True
elif (anim2, anim) in hits:
wrong, info = hits[anim2, anim]
rep = True
else:
res = gen[r[anim2], :] - gen[r[anim], :]
wrong = len(res[res == 2]) + len(res[res == -2])
identical = len(res[res == 0])
info = len(res) - len(res[np.isnan(res)])
rep = False
hits[anim, anim2] = wrong, info
if info == 0 or rep: pass
elif wrong * 100.0 / info <= newLim:
if fout:
fout.write('%s\t%s\t%d\t%d\t%.3f\t%s\n' %
(anim, anim2, wrong, info,
wrong * 100.0 / info, 'N' + sex))
if fout: fout.close()
if len(out) > 0 and options.reportped:
fout = open(options.reportped, 'w')
fout.write(
'#ID\tparent\tdiscords\tinfo_sites\tdiscord%\tparent_sex\n' + out)
fout.close()