def setUp(self):
iv = IntervalNode(50, 59, Interval(50, 59))
for i in range(0, 110, 10):
if i == 50: continue
f = Interval(i, i + 9)
iv = iv.insert(f.start, f.end, f)
self.intervals = iv
def setUp(self):
iv = IntervalTree()
iv.add_interval(Interval(50, 59))
for i in range(0, 110, 10):
if i == 50: continue
f = Interval(i, i + 9)
iv.add_interval(f)
self.intervals = iv
def test_left(self):
iv = self.intervals
self.assertEqual(str(iv.left(60, n=2)),
str([Interval(50, 59),
Interval(40, 49)]))
for i in range(10, 100, 10):
r = iv.left(i, max_dist=10, n=1)
self.assertEqual(r[0].end, i - 1)
def setUp(self):
iv = IntervalNode(1, 2, Interval(1, 2))
self.max = 1000000
for i in range(0, self.max, 10):
f = Interval(i, i)
iv = iv.insert(f.start, f.end, f)
for i in range(600):
iv = iv.insert(0, 1, Interval(0, 1))
self.intervals = iv
def test_downstream(self):
iv = self.intervals
downstreams = iv.downstream_of_interval(Interval(59, 60),
num_intervals=200)
for d in downstreams:
self.assertTrue(d.start > 60)
downstreams = iv.downstream_of_interval(Interval(59, 60, strand=-1),
num_intervals=200)
for d in downstreams:
self.assertTrue(d.start < 59)
def test_upstream(self):
iv = self.intervals
upstreams = iv.upstream_of_interval(Interval(59, 60),
num_intervals=200)
for u in upstreams:
self.assertTrue(u.end < 59)
upstreams = iv.upstream_of_interval(Interval(60, 70, strand=-1),
num_intervals=200)
for u in upstreams:
self.assertTrue(u.start > 70)
upstreams = iv.upstream_of_interval(Interval(58, 58, strand=-1),
num_intervals=200)
for u in upstreams:
self.assertTrue(u.start > 59)
def test_right(self):
iv = self.intervals
self.assertEqual(str(iv.left(60, n=2)),
str([Interval(50, 59),
Interval(40, 49)]))
def get_right_start(b10):
r = iv.right(b10 + 1, n=1)
assert len(r) == 1
return r[0].start
for i in range(10, 100, 10):
self.assertEqual(get_right_start(i), i + 10)
for i in range(0, 100, 10):
r = iv.right(i - 1, max_dist=10, n=1)
self.assertEqual(r[0].start, i)
def setUp(self):
iv = IntervalTree()
n = 0
for i in range(1, 1000, 80):
iv.insert(i, i + 10, dict(value=i * i))
# add is synonym for insert.
iv.add(i + 20, i + 30, dict(astr=str(i * i)))
# or insert/add an interval object with start, end attrs.
iv.insert_interval(
Interval(i + 40, i + 50, value=dict(astr=str(i * i))))
iv.add_interval(
Interval(i + 60, i + 70, value=dict(astr=str(i * i))))
n += 4
self.intervals = self.iv = iv
self.nintervals = n
def test_n(self):
iv = self.intervals
for i in range(0, 90, 10):
r = iv.after(i, max_dist=20, num_intervals=2)
self.assertEqual(r[0].start, i + 10)
self.assertEqual(r[1].start, i + 20)
r = iv.after_interval(Interval(i, i), max_dist=20, num_intervals=2)
self.assertEqual(r[0].start, i + 10)
self.assertEqual(r[1].start, i + 20)
def _check_cdna_vs_utr(transcript):
"""
Verify that cDNA + UTR in the transcript add up.
:return:
"""
transcript.logger.debug("Checking the cDNA for %s", transcript.id)
if transcript.cdna_length > transcript.combined_utr_length + transcript.combined_cds_length:
if transcript.combined_utr == transcript.combined_cds == []:
# non-coding transcript
transcript.logger.debug("%s is non coding, returning",
transcript.id)
return
assert transcript.combined_cds != []
transcript.logger.debug(
"Recalculating the UTR for %s. Reason: cDNA length %s, UTR %s, CDS %s (total %s)",
transcript.id, transcript.cdna_length,
transcript.combined_utr_length, transcript.combined_cds_length,
transcript.combined_utr_length + transcript.combined_cds_length)
transcript.combined_utr = [] # Reset
transcript.combined_cds = sorted(transcript.combined_cds,
key=operator.itemgetter(0, 1))
combined_cds = IntervalTree.from_tuples(transcript.combined_cds)
orfs = [
IntervalTree.from_tuples([_[1] for _ in orf if _[0] == "CDS"])
for orf in transcript.internal_orfs
]
assert isinstance(combined_cds, IntervalTree)
exons = IntervalTree.from_intervals(
[Interval(*exon) for exon in transcript.exons])
mapper = defaultdict(list)
for cds in transcript.combined_cds:
fexon = exons.find(cds[0] - 1, cds[1], strict=False)
if len(fexon) > 1:
raise InvalidCDS(
"{} has a CDS ({}) which straddles {} different exons ({})."
.format(transcript.id, cds, len(fexon), fexon))
elif len(fexon) == 0:
raise InvalidCDS(
"{} has a CDS ({}) which is not mapped to any exon.".
format(transcript.id, cds, len(fexon), fexon))
mapper[fexon[0]].append(cds)
for exon in transcript.exons:
if exon not in mapper:
transcript.combined_utr.append(exon)
continue
elif len(mapper[exon]) == 1:
cds = mapper[exon][0]
if cds[0] == exon[0] and exon[1] == cds[1]:
continue
else:
before = None
after = None
if cds[0] < exon[0] or cds[1] > exon[1]:
raise InvalidCDS("{} in {} debords its exon {}".format(
cds, transcript.id, exon))
if cds[0] > exon[0]:
before = (exon[0], max(cds[0] - 1, exon[0]))
transcript.combined_utr.append(before)
if cds[1] < exon[1]:
after = (min(cds[1] + 1, exon[1]), exon[1])
transcript.combined_utr.append(after)
assert before or after, (exon, cds)
else:
transcript.logger.debug("Starting to find the UTRs for %s",
exon)
found = sorted(mapper[exon])
utrs = []
for pos, interval in enumerate(found):
if pos == len(found) - 1:
if exon[1] > interval[1]:
utrs.append((min(exon[1],
interval[1] + 1), exon[1]))
continue
if pos == 0 and exon[0] < interval[0]:
utrs.append((exon[0], max(exon[0], interval[0] - 1)))
next_interval = found[pos + 1]
if not (interval[1] + 1 <= next_interval[0] - 1):
raise InvalidCDS(
"Error while inferring the UTR for a transcript with multiple ORFs: overlapping CDS found."
)
utrs.append((interval[1] + 1, next_interval[0] - 1))
assert utrs, found
utr_sum = sum([_[1] - _[0] + 1 for _ in utrs])
cds_sum = sum(_[1] - _[0] + 1 for _ in found)
assert utr_sum + cds_sum == exon[1] - exon[0] + 1, (utr_sum,
cds_sum,
exon[1] -
exon[0] +
1, utrs,
found)
transcript.combined_utr.extend(utrs)
# If no CDS and no UTR are present, all good
equality_one = (transcript.combined_cds_length ==
transcript.combined_utr_length == 0)
# Otherwise, if cDNA length == UTR + CDS, all good
equality_two = (
transcript.cdna_length == transcript.combined_utr_length +
transcript.combined_cds_length)
if not (equality_one or equality_two):
# Something fishy going on
raise InvalidCDS(""""Failed to create the UTR:
ID: {}
Exons: {}
Combined CDS: {}
Combined UTR: {}
CDS == UTR == 0: {}
CDNA == CDS + UTR: {}
CDNA == {}
CDS == {}
UTR == {}""".format(transcript.id, transcript.exons, transcript.combined_cds,
transcript.combined_utr, equality_one, equality_two,
transcript.cdna_length, transcript.combined_cds_length,
transcript.combined_utr_length))