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")
def main(argv):
infile = open(argv[1])
outfile = open(os.path.splitext(argv[1])[0] + ".fa", 'w', -1)
pos = 0
curr_target_name = ""
start = 0
a_pattern = re.compile("^a")
s_pattern = re.compile("^s")
for line in infile:
if s_pattern.search(line):
sline = line.split()
if pos == 0:
# target
pos = 1
target = sline[1]
if curr_target_name != target:
if curr_target_name != "":
outfile.write("\n")
start = 0
outfile.write(">{0}\n".format(target))
curr_target_name = target
elif pos == 1:
# query
pos = 0
seq = sline[6].translate(None, "-")
if sline[4] == "-":
seq = futil.reverse_complement(seq)
start = futil.write_segment(seq, start, outfile)
outfile.write("\n")