def _run(self, _config, temp):
def keyfunc(bed):
return (bed.contig, bed.name, bed.start)
fastafile = pysam.Fastafile(self._reference)
seqs = collections.defaultdict(list)
with open(self._intervals) as bedfile:
intervals = text.parse_lines_by_contig(bedfile, pysam.asBed()).items()
for (contig, beds) in sorted(intervals):
beds.sort(key = keyfunc)
for (gene, gene_beds) in itertools.groupby(beds, lambda x: x.name):
gene_beds = tuple(gene_beds)
for bed in gene_beds:
seqs[(contig, gene)].append(fastafile.fetch(contig, bed.start, bed.end))
seq = "".join(seqs[(contig, gene)])
if any((bed.strand == "-") for bed in gene_beds):
assert all((bed.strand == "-") for bed in gene_beds)
seq = sequences.reverse_complement(seq)
seqs[(contig, gene)] = seq
temp_file = os.path.join(temp, "sequences.fasta")
with open(temp_file, "w") as out_file:
for ((_, gene), sequence) in sorted(seqs.items()):
fasta.print_fasta(gene, sequence, out_file)
move_file(temp_file, self._outfile)
def main(argv):
parser = argparse.ArgumentParser()
parser.add_argument("--genotype",
help="Tabix indexed pileup file.",
required=True)
parser.add_argument("--intervals", help="BED file.", required=True)
parser.add_argument("--padding",
type=int,
default=10,
help="Number of bases to expand intervals, when "
"filtering based on adjacent indels [%default]")
parser.add_argument("--min-distance-to-indels",
type=int,
default=5,
help="Variants closer than this distance from indels "
"are filtered [%default].")
args = parser.parse_args(argv)
genotype = pysam.Tabixfile(args.genotype)
with open(args.intervals) as bed_file:
intervals = text.parse_lines_by_contig(bed_file, pysam.asBed())
for (_, beds) in sorted(intervals.items()):
for (name, sequence) in build_genes(args, genotype, beds):
FASTA(name, None, sequence).write(sys.stdout)
return 0
def _run(self, _config, temp):
def keyfunc(bed):
return (bed.contig, bed.name, bed.start)
fastafile = pysam.Fastafile(self._reference)
seqs = collections.defaultdict(list)
with open(self._intervals) as bedfile:
intervals = text.parse_lines_by_contig(bedfile,
pysam.asBed()).items()
for (contig, beds) in sorted(intervals):
beds.sort(key=keyfunc)
for (gene,
gene_beds) in itertools.groupby(beds, lambda x: x.name):
gene_beds = tuple(gene_beds)
for bed in gene_beds:
seqs[(contig, gene)].append(
fastafile.fetch(contig, bed.start, bed.end))
seq = "".join(seqs[(contig, gene)])
if any((bed.strand == "-") for bed in gene_beds):
assert all((bed.strand == "-") for bed in gene_beds)
seq = sequences.reverse_complement(seq)
seqs[(contig, gene)] = seq
temp_file = os.path.join(temp, "sequences.fasta")
with open(temp_file, "w") as out_file:
for ((_, gene), sequence) in sorted(seqs.items()):
fasta.print_fasta(gene, sequence, out_file)
move_file(temp_file, self._outfile)
def test_parse_lines__two_contigs():
lines = ["abc line1 \n", "def line2 \n"]
def _parse(line, length):
assert_equal(len(line), length)
return _RecordMock(*line.split())
expected = {"abc" : [_RecordMock("abc", "line1")],
"def" : [_RecordMock("def", "line2")]}
assert_equal(parse_lines_by_contig(lines, _parse), expected)
def test_parse_lines__two_contigs():
lines = ["abc line1 \n", "def line2 \n"]
def _parse(line, length):
assert_equal(len(line), length)
return _RecordMock(*line.split())
expected = {
"abc": [_RecordMock("abc", "line1")],
"def": [_RecordMock("def", "line2")]
}
assert_equal(parse_lines_by_contig(lines, _parse), expected)
def read_intervals(filename):
with open(filename) as bed_file:
intervals = text.parse_lines_by_contig(bed_file, BEDRecord)
for (key, beds) in intervals.iteritems():
bed_tuples = []
for bed in beds:
if len(bed) < 6:
sys.stderr.write(("ERROR: Invalid BED record '%r', must "
"have at least 6 fields ...\n") % (bed,))
return None
bed_tuples.append(bed)
intervals[key] = bed_tuples
return intervals
def read_intervals(filename):
with open(filename) as bed_file:
intervals = text.parse_lines_by_contig(bed_file, BEDRecord)
for (key, beds) in intervals.iteritems():
bed_tuples = []
for bed in beds:
if len(bed) < 6:
sys.stderr.write(
("ERROR: Invalid BED record '%r', must "
"have at least 6 fields ...\n") % (bed, ))
return None
bed_tuples.append(bed)
intervals[key] = bed_tuples
return intervals
def _run(self, _config, temp):
def _by_name(bed):
return bed.name
fastafile = pysam.Fastafile(self._reference)
seqs = collections.defaultdict(list)
with open(self._bedfile) as bedfile:
bedrecords = text.parse_lines_by_contig(bedfile, BEDRecord)
for (contig, beds) in sorted(bedrecords.iteritems()):
beds.sort(key=lambda bed: (bed.contig, bed.name, bed.start))
for (gene, gene_beds) in itertools.groupby(beds, _by_name):
gene_beds = tuple(gene_beds)
sequence = self._collect_sequence(fastafile, gene_beds)
seqs[(contig, gene)] = sequence
temp_file = os.path.join(temp, "sequences.fasta")
with open(temp_file, "w") as out_file:
for ((_, gene), sequence) in sorted(seqs.items()):
FASTA(gene, None, sequence).write(out_file)
fileutils.move_file(temp_file, self._outfile)
def _run(self, _config, temp):
def _by_name(bed):
return bed.name
fastafile = pysam.Fastafile(self._reference)
seqs = collections.defaultdict(list)
with open(self._bedfile) as bedfile:
bedrecords = text.parse_lines_by_contig(bedfile, pysam.asBed())
for (contig, beds) in sorted(bedrecords.iteritems()):
beds.sort(key=lambda bed: (bed.contig, bed.name, bed.start))
for (gene, gene_beds) in itertools.groupby(beds, _by_name):
gene_beds = tuple(gene_beds)
sequence = self._collect_sequence(fastafile, gene_beds)
seqs[(contig, gene)] = sequence
temp_file = os.path.join(temp, "sequences.fasta")
with open(temp_file, "w") as out_file:
for ((_, gene), sequence) in sorted(seqs.items()):
FASTA(gene, None, sequence).write(out_file)
fileutils.move_file(temp_file, self._outfile)
def main(argv):
parser = argparse.ArgumentParser()
parser.add_argument("--genotype", help="Tabix indexed pileup file.",
required=True)
parser.add_argument("--intervals", help="BED file.", required=True)
parser.add_argument("--padding", type=int, default=10,
help="Number of bases to expand intervals, when "
"filtering based on adjacent indels [%default]")
parser.add_argument("--min-distance-to-indels", type=int, default=5,
help="Variants closer than this distance from indels "
"are filtered [%default].")
args = parser.parse_args(argv)
genotype = pysam.Tabixfile(args.genotype)
with open(args.intervals) as bed_file:
intervals = text.parse_lines_by_contig(bed_file, pysam.asBed())
for (_, beds) in sorted(intervals.items()):
for (name, sequence) in build_genes(args, genotype, beds):
FASTA(name, None, sequence).write(sys.stdout)
return 0
def read_intervals(filename):
with open(filename) as bed_file:
intervals = text.parse_lines_by_contig(bed_file, pysam.asBed())
for (key, beds) in intervals.iteritems():
bed_tuples = []
for bed in beds:
if len(bed) < 6:
sys.stderr.write(("ERROR: Invalid BED record '%s', must "
"have at least 6 fields ...\n") %
("\\t".join(bed),))
return None
# Transform to a named tuple, as Pysam has a tendency to
# segfault if you do anything wrong
bed = list(bed)[:6] # BED6 only
bed[1] = int(bed[1]) # start
bed[2] = int(bed[2]) # end
bed[4] = int(bed[4]) # score
bed_tuples.append(BEDTuple(*bed))
intervals[key] = bed_tuples
return intervals
