def extract_breakpoint_sequence(tx_name_5p, tx_end_5p,
tx_name_3p, tx_start_3p,
ref_fa, max_read_length,
homology_mismatches):
tx_start_5p = max(0, tx_end_5p - max_read_length + 1)
tx_end_3p = tx_start_3p + max_read_length - 1
# fetch sequence
seq5p = ref_fa.fetch(tx_name_5p, tx_start_5p, tx_end_5p).upper()
seq3p = ref_fa.fetch(tx_name_3p, tx_start_3p, tx_end_3p).upper()
# pad sequence if too short
if len(seq5p) < (max_read_length - 1):
logging.warning("Could not extract sequence of length >%d from "
"5' partner at %s:%d-%d, only retrieved "
"sequence of length %d" %
(max_read_length-1, tx_name_5p, tx_start_5p,
tx_end_5p, len(seq5p)))
# pad sequence
padding = (max_read_length - 1) - len(seq5p)
seq5p = ("N" * padding) + seq5p
if len(seq3p) < max_read_length - 1:
logging.warning("Could not extract sequence of length >%d from "
"3' partner at %s:%d-%d, only retrieved "
"sequence of length %d" %
(max_read_length-1, tx_name_3p, tx_start_3p,
tx_end_3p, len(seq3p)))
# pad sequence
padding = (max_read_length - 1) - len(seq3p)
seq3p = seq3p + ("N" * padding)
# if 5' partner continues along its normal transcript
# without fusing, get the sequence that would result
homolog_end_5p = tx_end_5p + max_read_length - 1
homolog_seq_5p = ref_fa.fetch(tx_name_5p, tx_end_5p, homolog_end_5p).upper()
# if 3' partner were to continue in the 5' direction,
# grab the sequence that would be produced
homolog_start_3p = max(0, tx_start_3p - max_read_length + 1)
homolog_seq_3p = ref_fa.fetch(tx_name_3p, homolog_start_3p, tx_start_3p).upper()
# count number of bases in common between downstream 5' sequence
# and the sequence of the 3' partner in the chimera
homology_right = calc_homology(homolog_seq_5p, seq3p,
homology_mismatches)
# count number of bases in common between upstream 3' sequence
# and the sequence of the 5' partner in the chimera
homology_left = calc_homology(homolog_seq_3p[::-1], seq5p[::-1],
homology_mismatches)
return seq5p, seq3p, homology_left, homology_right
def extract_breakpoint_sequence(tx_id_5p, tx_end_5p, tx_id_3p, tx_start_3p,
ref_fa, max_read_length, homology_mismatches):
tx_start_5p = max(0, tx_end_5p - max_read_length + 1)
tx_end_3p = tx_start_3p + max_read_length - 1
# fetch sequence
seq5p = ref_fa.fetch(tx_id_5p, tx_start_5p, tx_end_5p).upper()
seq3p = ref_fa.fetch(tx_id_3p, tx_start_3p, tx_end_3p).upper()
# pad sequence if too short
if len(seq5p) < (max_read_length - 1):
logging.warning("Could not extract sequence of length >%d from "
"5' partner at %s:%d-%d, only retrieved "
"sequence of length %d" %
(max_read_length - 1, tx_id_5p, tx_start_5p, tx_end_5p,
len(seq5p)))
# pad sequence
padding = (max_read_length - 1) - len(seq5p)
seq5p = ("N" * padding) + seq5p
if len(seq3p) < max_read_length - 1:
logging.warning("Could not extract sequence of length >%d from "
"3' partner at %s:%d-%d, only retrieved "
"sequence of length %d" %
(max_read_length - 1, tx_id_3p, tx_start_3p, tx_end_3p,
len(seq3p)))
# pad sequence
padding = (max_read_length - 1) - len(seq3p)
seq3p = seq3p + ("N" * padding)
# if 5' partner continues along its normal transcript
# without fusing, get the sequence that would result
homolog_end_5p = tx_end_5p + max_read_length - 1
homolog_seq_5p = ref_fa.fetch(tx_id_5p, tx_end_5p, homolog_end_5p).upper()
# if 3' partner were to continue in the 5' direction,
# grab the sequence that would be produced
homolog_start_3p = max(0, tx_start_3p - max_read_length + 1)
homolog_seq_3p = ref_fa.fetch(tx_id_3p, homolog_start_3p,
tx_start_3p).upper()
# count number of bases in common between downstream 5' sequence
# and the sequence of the 3' partner in the chimera
homology_right = calc_homology(homolog_seq_5p, seq3p, homology_mismatches)
# count number of bases in common between upstream 3' sequence
# and the sequence of the 5' partner in the chimera
homology_left = calc_homology(homolog_seq_3p[::-1], seq5p[::-1],
homology_mismatches)
return seq5p, seq3p, homology_left, homology_right
def testHomology(self):
a = "AAAAGGGGTTTTCCCC"
b = "AAAAGGGGTTTTCCCC"
self.assertEquals(calc_homology(a, b, 0), 16)
b = "AAAAGGGGTTTTCCCG"
self.assertEquals(calc_homology(a, b, 0), 15)
b = "AAATTTGGTTTTCCCC"
self.assertEquals(calc_homology(a, b, 0), 3)
self.assertEquals(calc_homology(a, b, 1), 4)
self.assertEquals(calc_homology(a, b, 2), 5)
self.assertEquals(calc_homology(a, b, 3), 16)
def testHomology(self):
a = "AAAAGGGGTTTTCCCC"
b = "AAAAGGGGTTTTCCCC"
self.assertEquals(calc_homology(a, b, 0), 16)
b = "AAAAGGGGTTTTCCCG"
self.assertEquals(calc_homology(a, b, 0), 15)
b = "AAATTTGGTTTTCCCC"
self.assertEquals(calc_homology(a, b, 0), 3)
self.assertEquals(calc_homology(a, b, 1), 4)
self.assertEquals(calc_homology(a, b, 2), 5)
self.assertEquals(calc_homology(a, b, 3), 16)