def __init__(self, f, fname, default_sample_id):
TextParser.__init__(self, f, fname, default_sample_id)
self.__format = None
# Metadata and comments
line = self._readline()
while len(line) > 0 and line.startswith("#"):
if line.startswith("##fileformat="):
self.__format = line[13:]
line = self._readline()
if len(line) > 0: # Header
column_indices = {}
columns = line.rstrip().split("\t")
self._col_size = len(columns)
for i, name in enumerate(columns):
name = name.lower()
if name in _COLUMNS:
column_indices[name] = i
try:
self._col_name_indices = column_indices
self._col_indices = [column_indices[name] for name in _COLUMNS]
except KeyError as ex:
raise ParserException("Header column not found: {0}".format(ex.args[0]), self._location())
else:
raise ParserException("Header not found", (self._fname))
def __init__(self, f, fname, default_sample_id):
TextParser.__init__(self, f, fname, default_sample_id)
self.__format = None
# Metadata and comments
line = self._readline()
while len(line) > 0 and line.startswith("#"):
if line.startswith("##INDIVIDUAL="):
self._default_sample_id = line[13:]
elif line.startswith("##fileformat="):
self.__format = line[13:]
line = self._readline()
if len(line) > 0: # First line
self._queue_line(line)
def next(self):
TextParser.next(self)
var = None
while var is None:
line = self._readline()
if len(line) == 0:
raise StopIteration()
fields = line.rstrip("\n").split("\t")
chr, start, strand, vtype, ref, alt1, alt2, sample = [
fields[i] if i < self._col_size else None for i in self._col_indices]
#print ">>>", chr, start, strand, vtype, ref, alt1, alt2, sample
# Chromosome
chr = parse_chromosome(chr)
if chr is None:
self._discard_line()
continue
# Start
try:
start = int(start)
except:
self._discard_line()
continue
# Strand
if len(strand) == 0 or strand == "1" or strand == "+1":
strand = "+"
elif strand == "-1":
strand = "-"
elif strand not in ["+", "-"]:
self._discard_line()
continue
# Ref & alt
if ref is None or alt1 is None or alt2 is None:
self._discard_line()
continue
try:
for i, x in enumerate([ref, alt1, alt2]):
if _ALLELE_RE.match(x) is None:
self._discard_line()
raise SkipLine()
except SkipLine:
continue
alt = alt1
if ref == "-":
# [1 2] --> [1 2] 3
# . - . N
# . T . N T
ref = "N"
alt = "N" + alt if alt != "-" else "N"
elif alt == "-":
# 1 [2] 3 --> [1 2] 3
# . T . N T
# . - . N
start -= 1
ref = "N" + ref
alt = "N"
ref_len = len(ref)
vtype = Variant.SUBST if ref_len == len(alt) else Variant.INDEL
if len(sample) == 0:
sample = self._default_sample_id
if alt1 != alt2:
fields[self._col_name_indices[_COL_ALLELE1]] = fields[self._col_name_indices[_COL_ALLELE2]]
self._queue_line("\t".join(fields))
if ref == alt:
continue
var = Variant(type=vtype, chr=chr, start=start, ref=ref, alt=alt, strand=strand,
samples=[Sample(name=sample)])
#from intogensm.variants.utils import var_to_tab
#print "***", var
#print "+++", var_to_tab(var)
return var
def next(self):
TextParser.next(self)
var = None
while var is None:
fields = self.__read_fields()
if len(fields) < 5:
self._discard_line()
continue
if len(fields) < 6:
fields += [self._default_sample_id]
chr, start, end, strand, allele, sample = fields[0:6]
#print ">>>", chr, start, end, strand, allele, sample
# Chromosome
chr = parse_chromosome(chr)
if chr is None:
self._discard_line()
continue
# Start and end
try:
start = int(start)
except:
self._discard_line()
continue
try:
end = int(end)
except:
self._discard_line()
continue
if start > end:
start, end = end, start
# Strand
if len(strand) == 0 or strand == "1" or strand == "+1":
strand = "+"
elif strand == "-1":
strand = "-"
elif strand not in ["+", "-"]:
self._discard_line()
continue
# Alleles
alleles = allele.split(">")
if len(alleles) != 2:
self._discard_line()
continue
ref, alt = alleles
# Check that are well formed
try:
for a in [ref, alt]:
if _ALLELE_RE.match(a) is None:
self._discard_line()
raise SkipLine()
except SkipLine:
continue
# Special cases
if ref == "-" or alt == "-": # ->A, GCT>-
if ref == "-":
# [1 2] --> [1 2] 3
# . - . N
# . T . N T
ref = "N"
alt = "N" + alt if alt != "-" else "N"
elif alt == "-":
# 1 [2] 3 --> [1 2] 3
# . T . N T
# . - . N
start -= 1
ref = "N" + ref
alt = "N"
elif ref == "*" and len(alt) > 1 and alt[0] in ["-", "+"]: # *>-ACG, *>+CG
if alt[0] == "-":
start -= 1
ref = "N" + alt[1:]
alt = "N"
elif alt[0] == "+":
ref = "N"
alt = "N" + alt[1:]
elif "/" in ref or "/" in alt: # A/A>-/GGT, C/C>A/T, C/C>G/G
ref = ref.split("/")
alt = alt.split("/")
if len(ref) != 2 or len(ref) != len(alt):
self._discard_line()
continue
if ref[0] == ref[1] and alt[0] == alt[1]:
ref.pop()
alt.pop()
for i in range(len(ref)):
allele = "{0}>{1}".format(ref[i], alt[i])
self._queue_line("\t".join([chr, str(start), str(end), strand, allele, sample]))
continue
ref_len = len(ref)
alt_len = len(alt)
vtype = Variant.SUBST if ref_len == alt_len else Variant.INDEL
# Sample
if len(sample) == 0:
sample = self._default_sample_id
var = Variant(type=vtype, chr=chr, start=start, ref=ref, alt=alt, strand=strand,
samples=[Sample(name=sample)])
#from intogensm.variants.utils import var_to_tab
#print "***", var
#print "+++", var_to_tab(var)
return var
def __init__(self, f, fname, default_sample_id): TextParser.__init__(self, f, fname, default_sample_id)
def next(self):
TextParser.next(self)
var = None
while var is None:
line = self._readline()
while len(line) > 0 and line.lstrip().startswith("#"):
if line.startswith("##INDIVIDUAL="):
self._default_sample_id = line[13:]
line = self._readline()
if len(line) == 0:
raise StopIteration()
fields = line.rstrip("\n").split("\t")
if len(fields) < 5:
self._discard_line()
continue
chr, start, external_id, ref, alt = fields[0:5]
# Chromosome
chr = parse_chromosome(chr)
if chr is None:
self._discard_line()
continue
# Start
try:
start = int(start)
except:
self._discard_line()
continue
# Check ref and alt
if _REF_RE.match(ref) is None:
self._discard_line()
continue
if _ALT_RE.match(alt) is None:
self._discard_line()
continue
ref_len = len(ref)
if "," in alt:
vtype = None
s = alt.split(",")
for allele in s:
if ref_len == len(allele):
t = Variant.SUBST
else:
t = Variant.INDEL
if vtype is None:
vtype = t
elif vtype != t:
vtype = Variant.COMPLEX
break
else:
if ref_len == len(alt):
vtype = Variant.SUBST
else:
vtype = Variant.INDEL
var = Variant(type=vtype, chr=chr, start=start, ref=ref, alt=alt, strand="+",
samples=[Sample(name=self._default_sample_id)])
return var