def testSplit(self):
self.b = [[5, 12, 'I'], [20, 22, 'II'], [23, 28, 'III']]
r = Interval.split(self.b, 21, 24, flag='left')
self.assertListEqual(r, [[5, 12, 'I'], [20, 21, 'II']],
'Failed in Split')
r = Interval.split(self.b, 24, 26, flag='middle')
self.assertListEqual(r, [[24, 26, 'III']], 'Failed in Split')
r = Interval.split(self.b, 8, 21, flag='right')
self.assertListEqual(r, [[21, 22, 'II'], [23, 28, 'III']],
'Failed in Split')
def testInstanceSubstract(self):
self.a = Interval(self.a) # self.a is an instance
self.b = Interval(self.b) # self.b is an instance
c = self.a - self.b
self.assertListEqual(c.interval, [[1, 5, 'a', 'c'], [17, 20, 'b', 'd'],
[22, 23, 'b', 'd'], [30, 35, 'e']],
'Failed in instance c = a - b')
c = self.b - self.a
self.assertListEqual(c.interval, [[25, 28, 'III']],
'Failed in instance c = b - a')
self.a -= self.b
self.assertListEqual(self.a.interval,
[[1, 5, 'a', 'c'], [17, 20, 'b', 'd'],
[22, 23, 'b', 'd'], [30, 35, 'e']],
'Failed in instance a -= b')
def check_gDNA_reads(read_info, read_seq, target_fa):
read_len = len(read_seq)
# add padding
read_lst = deepcopy(read_info)
for item in read_lst:
item[1] += 10
if len(Interval(read_lst)) < len(read_lst): # realign
new_read_info = []
for read in target_fa.map(read_seq, MD=True):
left_cigar, right_cigar = '', ''
if read.q_st > 0:
left_cigar = '{}S'.format(read.q_st)
if read_len > read.q_en:
right_cigar = '{}S'.format(read_len - read.q_en)
if read.strand == 1: # plus strand
strand, reverse = '+', False
cigar = left_cigar + read.cigar_str + right_cigar
else: # minus strand
strand, reverse = '-', True
cigar = right_cigar + read.cigar_str + left_cigar
# reverse alignment if possible
alignment = convert_CIGAR(cigar, read.MD, reverse=reverse)
aln = ''.join('{}{}'.format(x[0], x[1]) for x in alignment)
index1, index2 = index_alignment(aln)
new_read_info.append(
[index1, index2, read.r_st, read.r_en, strand, aln, alignment])
return new_read_info
else:
return read_info
def testMapto(self):
r = Interval.mapto(self.d, self.c)
self.assertListEqual(
r,
[[3, 4, 'a', 'I'], [3, 7, 'b', 'I'], [4, 6, 'e', 'I'],
[4, 7, 'd', 'I'], [5, 7, 'f', 'I'], [10, 11, 'd', 'II'],
[10, 12, 'x', 'II'], [16, 20, 'x', 'III'], [16, 17, 'h', 'III'],
[18, 20, 'i', 'III'], [23, 24, 'x', 'IV']], 'Failed in Mapto')
def testInstanceAdd(self):
self.a = Interval(self.a) # self.a is an instance
self.b = Interval(self.b) # self.a is an instance
c = self.a + self.b
self.assertListEqual(c.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in instance c = a + b')
c = self.b + self.a # TODO: why this addition did not change the order
self.assertListEqual(c.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in instance c = b + a')
self.a += self.b
self.assertListEqual(self.a.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in instance a += b')
def testInstanceMultiple(self):
self.a = Interval(self.a) # self.a is an instance
self.b = Interval(self.b) # self.b is an instance
c = self.a * self.b
self.assertListEqual(c.interval,
[[5, 12, 'a', 'c', 'I'], [20, 22, 'b', 'd', 'II'],
[23, 25, 'b', 'd', 'III']],
'Failed in instance c = a * b')
c = self.b * self.a
self.assertListEqual(c.interval,
[[5, 12, 'I', 'a', 'c'], [20, 22, 'II', 'b', 'd'],
[23, 25, 'III', 'b', 'd']],
'Failed in instance c = b * a')
self.a *= self.b
self.assertListEqual(self.a.interval,
[[5, 12, 'a', 'c', 'I'], [20, 22, 'b', 'd', 'II'],
[23, 25, 'b', 'd', 'III']],
'Failed in instance a *= b')
def testAdd(self):
self.a = Interval(self.a) # self.a is an instance
c = self.a + self.b
self.assertListEqual(c.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in c = a + b')
c = self.b + self.a
self.assertListEqual(c.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in c = b + a')
self.a += self.b
self.assertListEqual(self.a.interval,
[[1, 12, 'a', 'c', 'I'],
[17, 28, 'b', 'd', 'II', 'III'], [30, 35, 'e']],
'Failed in a += b')
def testSlice(self):
self.a = Interval(self.a)
self.assertEqual(self.a[1], [17, 25, 'b', 'd'], 'Failed in a[1]')
self.assertListEqual(self.a[:2],
[[1, 12, 'a', 'c'], [17, 25, 'b', 'd']],
'Failed in a[:2]')
self.assertTrue([27, 34] in self.a, 'Failed in [27, 34] in a')
self.assertTrue([31, 34] in self.a, 'Failed in [31, 34] in a')
self.assertTrue([31, 37] in self.a, 'Failed in [31, 37] in a')
self.assertTrue([27, 37] in self.a, 'Failed in [27, 37] in a')
self.assertTrue([[27, 32], [33, 34]] in self.a,
'Failed in [[27, 32], [33, 34]] in a')
self.assertTrue([[31, 32], [33, 34]] in self.a,
'Failed in [[31, 32], [33, 34]] in a')
self.assertTrue([[31, 32], [33, 37]] in self.a,
'Failed in [[31, 32], [33, 37]] in a')
self.assertTrue([[27, 32], [33, 37]] in self.a,
'Failed in [[27, 32], [33, 37]] in a')
def parse_indel(read_info, target_name, target, bond_name=None, gDNA=False):
min_indel = None
indel_info = []
target_len = target[target_name]['total']
if not gDNA: # donor
left_bond = target[bond_name]['left_bond']
right_bond = target[bond_name]['right_bond']
dis = 5
else: # gDNA
left_bond = 0
right_bond = 0
cut_left = target[target_name]['cut_left']
cut_right = target[target_name]['cut_right']
dis = 20
for index1, index2, start, end, strand, aln, alignment in read_info:
if strand == '+':
interval = target[target_name]['interval']
else:
interval = target[target_name]['rev_interval']
# reverse start and end
start, end = target_len - end, target_len - start
aln_info = '{}|{}|{}|{}|{}|{}'.format(strand, start, end, index1,
index2, aln)
if start >= dis or target_len - end >= dis:
target_tag = 'partially'
seg_type = '-'.join(
x[2] for x in Interval.mapto([start, end], interval))
else:
target_tag = 'full'
if strand == '+':
seg_type = 'L-LH-C-RH-R' if gDNA else 'LH-I-RH'
else:
seg_type = 'R-RH-C-LH-L' if gDNA else 'RH-I-LH'
cut_indel = 0
total_indel = 0
reference_pos = start
read_tag = 'full'
for n, (num, t) in enumerate(alignment):
if t in ('S', 'H'):
if n == 0: # first postion
if num - left_bond >= dis:
read_tag = 'partially'
else: # last position
if num - right_bond >= dis:
read_tag = 'partially'
if gDNA:
if t == 'I' and cut_left <= reference_pos <= cut_right:
cut_indel += num
if t in ('D', 'U'):
if (cut_left <= reference_pos + num
and reference_pos <= cut_right):
cut_indel += num
if t in ('M', 'U', 'D'):
reference_pos += num
else:
if t in ('D', 'U', 'I'):
cut_indel += num
if t in ('D', 'U', 'I'):
total_indel += num
if target_tag == 'partially' or read_tag == 'partially':
tag = 'partially'
else:
tag = 'full'
if min_indel is None or total_indel < min_indel:
min_indel = total_indel
min_cut_indel = cut_indel
final_tag = tag
indel_info.append('{}|{}|{}|{}|{}'.format(target_name, seg_type,
total_indel, cut_indel,
aln_info))
return [indel_info, final_tag, min_indel, min_cut_indel, target_name]
def testLen(self):
a = Interval(self.a)
self.assertEqual(len(a), 3, 'Failed in length')
b = Interval(self.b)
self.assertEqual(len(b), 3, 'Fbiled in length')
def testInit(self):
a = Interval(self.a) # merge intervals
self.assertListEqual(
a.interval, [[1, 12, 'a', 'c'], [17, 25, 'b', 'd'], [30, 35, 'e']],
'Failed in initiation')
def testOverlapwith(self):
r = Interval.overlapwith(self.c, self.d)
self.assertListEqual(
r, [[3, 7, 'I', 'a', 'b', 'e', 'd', 'f'], [10, 12, 'II', 'd', 'x'],
[16, 20, 'III', 'x', 'h', 'i'], [23, 25, 'IV', 'x']],
'Failed in Overlapwith')
def testConvert(self):
e = Interval(self.e) # convert list to nested list
self.assertListEqual(e.interval, [[10, 15, 't']], 'Failed in convert')