def create_merged_seq(self):
"""Writes merged target-ordered query sequence"""
align = pd.read_table(os.path.join(self.out_dir, "longest_segments.txt"))
align.sort(columns=["target_name", "target_final_start"], inplace=True)
out = open(os.path.join(self.out_dir, "merged_seq.fa"), "w")
curr_target_name = ""
i = 0
start = 0
for row in align.iterrows():
query = row[1]['query_name']
target = row[1]['target_name']
fa = FastaHack(self.query_fas[query])
if curr_target_name != target:
if i > 0:
out.write("\n")
i = 1
out.write(">{0}\n".format(target))
curr_target_name = target
fasub = fa.get_sequence(query)
if row[1]['query_strand'] == "-":
fasub = futil.reverse_complement(fasub)
fasub1, start = futil.format_fasta(fasub, start)
[out.write(x) for x in fasub1]
ns = "N" * 1000
ns1, start = futil.format_fasta(ns, start)
[out.write(x) for x in ns1]
out.write("\n")
class FastaTestMore(FastaHackTest):
def setUp(self):
self.fa = FastaHack(FA)
def test_query(self):
self.assertEqual(self.fa.get_sub_sequence("1", 0, 4), "TAACC")
self.assertEqual(self.fa.get_sequence("1:1-5"), "TAACC")
self.assertEqual(self.fa["1:1-5"], "TAACC")
self.assertEqual(self.fa["1:4-5"], "CC")
def tab_to_vcf(input_file, output_file, reference_file, columns, info_fields, convert_iupac=False):
"""
Convert tab-delimited file to VCF.
Support for the fixed VCF fields: #CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO
PyVCF's _Record class requires the following arguments:
CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO, FORMAT, sample_indexes
convert_iupac (bool) : When present, convert IUPAC codes to the non-reference allele.
This is only possible for when the reference and IUPAC-determined alternates share
at least one allele. Tri-allelic conversion is not supported and will emit a warning.
IUPAC codes: http://www.bioinformatics.org/sms/iupac.html
"""
reference_dict = FastaHack(reference_file)
with open(input_file, "r") as input_fh:
reader = csv.DictReader(input_fh, delimiter="\t")
with open(TEMPLATE_VCF_FILE, "r") as template_fh:
vcf_reader = vcf.Reader(template_fh)
with open(output_file, "w") as output_fh:
vcf_writer = vcf.Writer(output_fh, vcf_reader, lineterminator='\n')
for row in reader:
args = [row.get(columns.get(f,None), ".") for f in VCF_COLUMN_ORDER]
# Convert position to an integer.
args[POSITION_INDEX] = int(args[POSITION_INDEX])
# Convert indels from GATK to VCF format.
if args[ALT_INDEX].startswith(("+", "-")) and not "/" in args[ALT_INDEX]:
args = gatk_indel_to_vcf(args, reference_dict)
# Optionally convert IUPAC code
if convert_iupac:
args = _convert_iupac(args)
# Convert alternate allele scalar to a list.
args[ALT_INDEX] = [args[ALT_INDEX]]
# Convert info fields
if info_fields:
INFO = {}
for vcf_field,tab_field in info_fields.items():
if tab_field in row:
INFO[vcf_field] = row[tab_field]
else:
INFO = {}
# Add empty entries for INFO, FORMAT, and sample_indexes.
args.extend([INFO, ".", []])
record = _Record(*args)
vcf_writer.write_record(record)
def tab_to_vcf(input_file, output_file, reference_file):
"""
Convert tab-delimited file to VCF.
Support for the fixed VCF fields: #CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO
PyVCF's _Record class requires the following arguments:
CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO, FORMAT, sample_indexes
"""
reference_dict = FastaHack(reference_file)
with open(input_file, "r") as input_fh:
reader = csv.DictReader(input_fh, delimiter="\t")
with open(TEMPLATE_VCF_FILE, "r") as template_fh:
vcf_reader = vcf.Reader(template_fh)
with open(output_file, "w") as output_fh:
vcf_writer = vcf.Writer(output_fh,
vcf_reader,
lineterminator='\n')
for row in reader:
args = [
row.get(tab_field, ".")
for vcf_field, tab_field in VCF_TO_FIELDS
]
# Convert position to an integer.
args[POSITION_INDEX] = int(args[POSITION_INDEX])
# Convert indels from GATK to VCF format.
if args[ALT_INDEX].startswith(
("+", "-")) and not "/" in args[ALT_INDEX]:
args = gatk_indel_to_vcf(args, reference_dict)
# Convert alternate allele scalar to a list.
args[ALT_INDEX] = [args[ALT_INDEX]]
# Add empty entries for INFO, FORMAT, and sample_indexes.
args.extend([{}, ".", []])
record = _Record(*args)
vcf_writer.write_record(record)
def setUp(self):
self.fa = FastaHack(FA)
if __name__=='__main__':
opts = OptionParser()
opts.add_option('','--contig_file',dest='fn_contig_file')
opts.add_option('','--wnd_width',dest='wnd_width',default=None,type=int)
opts.add_option('','--wnd_slide',dest='wnd_slide',default=None,type=int)
opts.add_option('','--wnd_pickle',dest='wnd_pickle',default=None)
opts.add_option('','--wnd_contig_file',dest='fn_wnd_contig_file')
opts.add_option('','--fasta',dest='fn_fasta')
opts.add_option('','--fn_out',dest='fn_out',default=None)
opts.add_option('','--sunk_based',dest='sunk_based',action='store_true',default=False)
(o, args) = opts.parse_args()
print "loading fasta %s..."%(o.fn_fasta)
fa = FastaHack(o.fn_fasta)
print>>stderr, "%s"%(o.fn_out)
out_wnd_DTS = DenseTrackSet(o.fn_wnd_contig_file,
"%s"%(o.fn_out),
overwrite=True,
openMode='w')
out_wnd_DTS.addGroup("GC")
out_wnd_DTS.addGroup("starts")
out_wnd_DTS.addGroup("ends")
out_wnd_DTS['GC'].addArray(tables.Float32Atom(),[])
out_wnd_DTS['starts'].addArray(tables.UInt32Atom(),[])
out_wnd_DTS['ends'].addArray(tables.UInt32Atom(),[])
###WE ONLY NEED THE STARTS because start[k],start[k+1] == start[k], end[k]
