class TestIssue9(unittest.TestCase):
def setUp(self):
self.tree4 = IntervalTree()
self.tree4.insert(Interval(22, 33, data='example1'))
self.tree4.insert(Interval(22, 33, data='example2'))
def test_right(self):
self.assertEqual(0, len(self.tree4.right(Interval(44, 55))))
self.assertEqual(2, len(self.tree4.right(Interval(11, 12))))
def test_left(self):
self.assertEqual(2, len(self.tree4.left(Interval(44, 55))))
self.assertEqual(0, len(self.tree4.left(Interval(11, 12))))
class TestIssue9(unittest.TestCase):
def setUp(self):
self.tree4 = IntervalTree()
self.tree4.insert(Interval(22, 33, data='example1'))
self.tree4.insert(Interval(22, 33, data='example2'))
def test_right(self):
self.assertEqual(0, len(self.tree4.right(Interval(44,55))))
self.assertEqual(2, len(self.tree4.right(Interval(11,12))))
def test_left(self):
self.assertEqual(2, len(self.tree4.left(Interval(44,55))))
self.assertEqual(0, len(self.tree4.left(Interval(11,12))))
def _create_intervaltree(locs):
it = IntervalTree()
for k, (start, end) in locs.iterrows():
intervals = it.find(start, end)
if intervals:
continue
it.add(start, end, k)
return it
def __init__(self,
content,
doc_id=rand_id(),
language=lng.ENGLISH,
preprocessors=None):
super().__init__(self, 0, len(content))
self._content = preprocess(content,
preprocessors) if preprocessors else content
self._annotations = IntervalTree()
self._doc_id = rand_id(10) if doc_id is None else doc_id
self._completed = {}
self._next_id = 0
self[LANGUAGE] = language
self._aid_dict = {}
def createIntervalTreesFragmentResolution():
'''
creates one interval tree for quick lookups
returns
fragmentsMap[fragmentId] = [tuple(chrom, fragmentMidPoint)]
intersect_tree - intersect Tree for interval matching
'''
if (options.verbose):
print >> sys.stdout, "- %s START : populate intervaltree with given resolution for chromosomes matching pattern %s" % (timeStamp(), options.chromPattern)
intersect_tree = IntervalTree()
fragmentsCount = 0
fragmentsMap = {}
fragmentsChrom = {} # lookp table for fragment ranges of a chromosome
for line in fileinput.input([options.chromSizes]):
chrom=line.split("\t")[0]
# check if chromosome needs to be filtered out or not
if (options.chromPattern != "" and not re.match("^"+options.chromPattern+"$", chrom)):
# skip this one
if (options.vverbose):
print "skipping pattern %s" % (line)
continue
fragmentsStart=fragmentsCount
chromlen=int(line.split("\t")[1])
for i in range(0, chromlen, options.resolution):
start=i
end=min(i+ options.resolution, chromlen)
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, int(0.5*(start+end))])
fragmentsCount += 1
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom, start, end)
fragmentsEnd=fragmentsCount
fragmentsChrom[chrom] = tuple([fragmentsStart, fragmentsEnd])
if (options.verbose):
print >> sys.stdout, "- %s FINISHED: intervaltree populated" % (timeStamp())
return [ fragmentsMap, intersect_tree, fragmentsCount, fragmentsChrom ]
def test_tree_pickle(self):
a = IntervalTree()
for ichr in range(5):
for i in range(10, 100, 6):
f = Feature(i -4, i + 4, strand=1, chr=ichr)
a.insert(f)
a.dump('a.pkl')
b = IntervalTree()
b.load('a.pkl')
for ichr in range(5):
for i in range(10, 100, 6):
f = Feature(i -4, i + 4, strand=1, chr=ichr)
af = sorted(a.find(f), key=operator.attrgetter('start'))
bf = sorted(b.find(f), key=operator.attrgetter('start'))
assert len(bf) > 0
self.assertEqual(len(af), len(bf))
self.assertEqual(af[0].start, bf[0].start)
self.assertEqual(af[-1].start, bf[-1].start)
class EmptyTreeTestCase(unittest.TestCase):
""" test search on an empty tree."""
def setUp(self):
self.tree = IntervalTree()
def test_search(self):
self.tree.search(46, 47)
def test_find(self):
self.tree.find(Interval(46, 47))
def test_left(self):
self.tree.left(Interval(46, 47))
def test_right(self):
self.tree.right(Interval(46, 47))
class EmptyTreeTestCase(unittest.TestCase):
""" test search on an empty tree."""
def setUp(self):
self.tree = IntervalTree()
def test_search(self):
self.tree.search(46, 47)
def test_find(self):
self.tree.find(Interval(46, 47))
def test_left(self):
self.tree.left(Interval(46, 47))
def test_right(self):
self.tree.right(Interval(46, 47))
def join(leftSet, rightSet, mincols=1, leftfill=True, rightfill=True):
# Read rightSet into memory:
rightlen = 0
leftlen = 0
rightTree = IntervalTree()
for item in rightSet:
if isinstance(item, GenomicInterval):
rightTree.insert( item, rightSet.linenum, item.fields )
if rightlen == 0: rightlen = item.nfields
for interval in leftSet:
if leftlen == 0 and isinstance(interval, GenomicInterval):
leftlen = interval.nfields
if not isinstance(interval, GenomicInterval):
yield interval
else:
result = []
rightTree.intersect( interval, lambda node: result.append( node ) )
overlap_not_met = 0
for item in result:
if item.start in range(interval.start,interval.end+1) and item.end not in range(interval.start,interval.end+1):
overlap = interval.end-item.start
elif item.end in range(interval.start,interval.end+1) and item.start not in range(interval.start,interval.end+1):
overlap = item.end-interval.start
elif item.start in range(interval.start,interval.end+1) and item.end in range(interval.start,interval.end+1):
overlap = item.end-item.start
else: #the intersecting item's start and end are outside the interval range
overlap = interval.end-interval.start
if overlap < mincols:
overlap_not_met += 1
continue
outfields = list(interval)
map(outfields.append, item.other)
setattr( item, "visited", True )
yield outfields
if (len(result) == 0 or overlap_not_met == len(result)) and rightfill:
outfields = list(interval)
for x in range(rightlen): outfields.append(".")
yield outfields
if leftfill:
def report_unvisited( node, results ):
if not hasattr(node, "visited"):
results.append( node )
results = []
rightTree.traverse( lambda x: report_unvisited( x, results ) )
for item in results:
outfields = list()
for x in range(leftlen): outfields.append(".")
map(outfields.append, item.other)
yield outfields
def main(argv):
if len(argv) < 3:
print("Usage: bedcov.py <loaded.bed> <streamed.bed>")
sys.exit(1)
bed, i = {}, 0
start = timer()
with open(argv[1]) as fp:
for line in fp:
t = line[:-1].split("\t")
if not t[0] in bed:
bed[t[0]] = IntervalTree()
bed[t[0]].add(int(t[1]) + 1, int(t[2]))
sys.stderr.write("Read in {} sec\n".format(timer() - start))
start = timer()
with open(argv[2]) as fp:
for line in fp:
t = line[:-1].split("\t")
if not t[0] in bed:
print("{}\t{}\t{}\t0".format(t[0], t[1], t[2]))
else:
r = bed[t[0]].search(int(t[1]) + 1, int(t[2]))
print("{}\t{}\t{}\t{}".format(t[0], t[1], t[2], len(r)))
sys.stderr.write("Query in {} sec\n".format(timer() - start))
def setUp(self):
self.tree = IntervalTree()
def test_tree_pickle(self):
a = IntervalTree()
for ichr in range(5):
for i in range(10, 100, 6):
f = Interval(i - 4, i + 4)
a.insert(f)
a.dump('a.pkl')
b = IntervalTree()
b.load('a.pkl')
for ichr in range(5):
for i in range(10, 100, 6):
f = Interval(i - 4, i + 4)
af = sorted(a.find(f), key=operator.attrgetter('start'))
bf = sorted(b.find(f), key=operator.attrgetter('start'))
assert len(bf) > 0
self.assertEqual(len(af), len(bf))
self.assertEqual(af[0].start, bf[0].start)
self.assertEqual(af[-1].start, bf[-1].start)
"Start": np.int32,
"End": np.int32
})
background = pd.read_table(b,
sep="\t",
usecols=[1, 2],
header=None,
names="Start End".split(),
engine="c",
dtype={
"Start": np.int32,
"End": np.int32
})
tree = IntervalTree()
for start_, end_ in zip(background.Start, background.End):
tree.add(start_, end_)
start = time()
results = []
for start_, end_ in zip(chip.Start, chip.End):
results.append(tree.search(start_, end_))
end = time()
# print(result)
total = end - start
total_dt = datetime.datetime.fromtimestamp(total)
def join(leftSet, rightSet, mincols=1, leftfill=True, rightfill=True, asfraction=False, matchStrand=STRAND_NEUTRAL, outColumns=[-1,-1]):
# Read rightSet into memory:
rightlen = 0
leftlen = 0
rightStrandCol = -1
minoverlap = mincols
rightTree = IntervalTree()
for item in rightSet:
if isinstance(item, GenomicInterval):
rightTree.insert( item, rightSet.linenum, item.fields )
if rightlen == 0: rightlen = item.nfields
if rightStrandCol == -1: rightStrandCol = item.strand_col
for interval in leftSet:
if leftlen == 0 and isinstance(interval, GenomicInterval):
leftlen = interval.nfields
if not isinstance(interval, GenomicInterval):
yield interval
else:
result = []
rightTree.intersect( interval, lambda node: result.append( node ) )
overlap_not_met = 0
leftbases = interval.end - interval.start
for item in result:
rightbases = item.end - item.start
if (asfraction==True):
if rightbases < leftbases:
mincols = rightbases
else:
mincols = leftbases
mincols = math.floor(mincols * minoverlap)
if item.start in range(interval.start,interval.end+1) and item.end not in range(interval.start,interval.end+1):
overlap = interval.end-item.start
elif item.end in range(interval.start,interval.end+1) and item.start not in range(interval.start,interval.end+1):
overlap = item.end-interval.start
elif item.start in range(interval.start,interval.end+1) and item.end in range(interval.start,interval.end+1):
overlap = item.end-item.start
else: #the intersecting item's start and end are outside the interval range
overlap = interval.end-interval.start
if overlap < mincols:
overlap_not_met += 1
continue
else:
#check strand
strandMatched = STRAND_INTEGER_VALUES[interval.strand] * STRAND_INTEGER_VALUES[item.other[rightStrandCol]]
if (strandMatched == -1 and matchStrand > 0):
#needed match but found a complement
overlap_not_met += 1
continue
if (strandMatched == 1 and matchStrand < 0):
#needed complement but found a match
overlap_not_met += 1
continue
if (strandMatched == 0 and (matchStrand < -1 or matchStrand > 1)):
#strict criteria but only permissive match found
overlap_not_met += 1
continue
#strand criteria met
setattr( item, "visited", True )
yield(getSelectedColumns( interval.fields, item.other, outColumns ))
if (len(result) == 0 or overlap_not_met == len(result)) and rightfill:
yield(getSelectedColumns( interval.fields, rightlen, outColumns ))
if leftfill:
def report_unvisited( node, results ):
if not hasattr(node, "visited"):
results.append( node )
results = []
rightTree.traverse( lambda x: report_unvisited( x, results ) )
for item in results:
yield(getSelectedColumns( leftlen, item.other, outColumns))
def join(
leftSet,
rightSet,
mincols=1,
leftfill=True,
rightfill=True,
asfraction=False,
matchStrand=STRAND_NEUTRAL,
outColumns=[-1, -1],
):
# Read rightSet into memory:
rightlen = 0
leftlen = 0
rightStrandCol = -1
minoverlap = mincols
rightTree = IntervalTree()
for item in rightSet:
if isinstance(item, GenomicInterval):
rightTree.insert(item, rightSet.linenum, item.fields)
if rightlen == 0:
rightlen = item.nfields
if rightStrandCol == -1:
rightStrandCol = item.strand_col
for interval in leftSet:
if leftlen == 0 and isinstance(interval, GenomicInterval):
leftlen = interval.nfields
if not isinstance(interval, GenomicInterval):
yield interval
else:
result = []
rightTree.intersect(interval, lambda node: result.append(node))
overlap_not_met = 0
leftbases = interval.end - interval.start
for item in result:
rightbases = item.end - item.start
if asfraction == True:
if rightbases < leftbases:
mincols = rightbases
else:
mincols = leftbases
mincols = math.floor(mincols * minoverlap)
if item.start in range(interval.start, interval.end + 1) and item.end not in range(
interval.start, interval.end + 1
):
overlap = interval.end - item.start
elif item.end in range(interval.start, interval.end + 1) and item.start not in range(
interval.start, interval.end + 1
):
overlap = item.end - interval.start
elif item.start in range(interval.start, interval.end + 1) and item.end in range(
interval.start, interval.end + 1
):
overlap = item.end - item.start
else: # the intersecting item's start and end are outside the interval range
overlap = interval.end - interval.start
if overlap < mincols:
overlap_not_met += 1
continue
else:
# check strand
strandMatched = (
STRAND_INTEGER_VALUES[interval.strand] * STRAND_INTEGER_VALUES[item.other[rightStrandCol]]
)
if strandMatched == -1 and matchStrand > 0:
# needed match but found a complement
overlap_not_met += 1
continue
if strandMatched == 1 and matchStrand < 0:
# needed complement but found a match
overlap_not_met += 1
continue
if strandMatched == 0 and (matchStrand < -1 or matchStrand > 1):
# strict criteria but only permissive match found
overlap_not_met += 1
continue
# strand criteria met
setattr(item, "visited", True)
yield (getSelectedColumns(interval.fields, item.other, outColumns))
if (len(result) == 0 or overlap_not_met == len(result)) and rightfill:
yield (getSelectedColumns(interval.fields, rightlen, outColumns))
if leftfill:
def report_unvisited(node, results):
if not hasattr(node, "visited"):
results.append(node)
results = []
rightTree.traverse(lambda x: report_unvisited(x, results))
for item in results:
yield (getSelectedColumns(leftlen, item.other, outColumns))
def add(self, chrom, start, end, val):
if chrom not in self.chroms:
self.chroms[chrom] = IntervalTree()
tree = self.chroms[chrom]
tree.add(start, end, val)
def createIntervalTreesFragmentFile(options):
'''
creates one interval tree for quick lookups
returns
fragmentsMap[fragmentId] = [tuple(chrom, start, end)]
intersect_tree - intersect Tree for interval matching
'''
if (options.verbose):
print >> sys.stdout, "- %s START : populate intervaltree from fragmented genome" % (
timeStamp())
intersect_tree = IntervalTree()
fragmentsCount = 0
fragmentsMap = {}
fragmentsChrom = {} # lookp table for fragment ranges of a chromosome
fragmentsStart = 0
start = 0
end = 0
counter = 0
chrom = ""
for line in fileinput.input([options.genomeFragmentFile]):
line = line.strip()
if (len(line) == 0 or line.startswith("Genome")
or line.startswith("Chromosome")):
continue
cols = line.split("\t")
try:
# check if chromosome changed from last
if (cols[0] != chrom):
# do we have to finish the last chromosome?
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, start, end])
fragmentsCount += 1
fragmentsChrom[chrom] = tuple(
[fragmentsStart, fragmentsCount])
fragmentsStart = fragmentsCount
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (
chrom, start, end)
# check if chromosome needs to be filtered out or not
if (options.chromPattern != ""
and not re.match(options.chromPattern, cols[0])):
chrom = ""
start = 0
end = 0
else:
chrom = cols[0]
start = int(cols[1])
end = int(cols[2])
counter = 0
# check if fragment aggregation is fulfilled
elif (counter >= options.fragmentAggregation):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (
chrom, start, end)
fragmentsMap[fragmentsCount] = tuple([chrom, start, end])
start = int(cols[1])
end = int(cols[2])
counter = 0
fragmentsCount += 1
else:
end = int(cols[2])
# increment counter
counter += 1
except:
if (options.verbose):
print >> sys.stderr, 'skipping line in options.genomeFragmentFile: %s' % (
line)
if (options.vverbose):
traceback.print_exc()
sys.exit(1)
# handle last fragment
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, start, end])
fragmentsCount += 1
fragmentsChrom[chrom] = tuple([fragmentsStart, fragmentsCount])
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom,
start, end)
if (options.verbose):
print >> sys.stdout, "- %s FINISHED: intervaltree populated" % (
timeStamp())
return [fragmentsMap, intersect_tree, fragmentsCount, fragmentsChrom]
class Document(HString):
def __init__(self,
content,
doc_id=rand_id(),
language=lng.ENGLISH,
preprocessors=None):
super().__init__(self, 0, len(content))
self._content = preprocess(content,
preprocessors) if preprocessors else content
self._annotations = IntervalTree()
self._doc_id = rand_id(10) if doc_id is None else doc_id
self._completed = {}
self._next_id = 0
self[LANGUAGE] = language
self._aid_dict = {}
@property
def content(self) -> str:
return self._content
@property
def doc_id(self):
return self._doc_id
def annotation(self,
annotation_type,
start=None,
end=None) -> typing.List[Annotation]:
try:
if end is None or start is None:
anno_iter = self._annotations.find(Interval(0, self.end))
else:
anno_iter = filter(
lambda x: x.data.overlaps(Span(start, end)),
self._annotations.find(Interval(start, end)))
except:
return []
if annotation_type:
annotation_type = annotation_type.lower()
return sorted([
x.data for x in anno_iter
if x.data.annotation_type.lower() == annotation_type
and x.data != self
])
return sorted([x.data for x in anno_iter if x.data != self])
def annotation_by_id(self, annotation_id: int):
return self._aid_dict[
annotation_id] if annotation_id in self._aid_dict else None
def previous_annotation(self,
annotation: Annotation,
annotation_type: str = None) -> 'Annotation':
if not annotation_type:
annotation_type = annotation.annotation_type
a = self.annotation(annotation_type, start=-1, end=annotation.start)
if len(a) == 0:
return Annotation(None, 0, 0, annotation_type, [])
return a[-1]
def next_annotation(self,
annotation: Annotation,
annotation_type: str = None) -> 'Annotation':
if not annotation_type:
annotation_type = annotation.annotation_type
a = self.annotation(annotation_type,
start=annotation.end,
end=self.end)
if len(a) == 0:
return Annotation(None, 0, 0, annotation_type, [])
return a[0]
def create_annotation(self,
type: str,
start: int,
end: int,
attributes=None) -> Annotation:
if attributes is None:
attributes = []
annotation = Annotation(self, start, end, type, attributes,
self._next_id)
self._next_id += 1
self._annotations.insert(
Interval(annotation.start, annotation.end, annotation))
self._aid_dict[annotation.annotation_id] = annotation
return annotation
def annotate(self, *args):
for arg in args:
if arg in self._completed:
continue
self.language().load()
annotator = self.language().get_annotator(arg)
if annotator:
annotator.annotate(self)
self._completed[arg] = '1.0'
else:
raise Exception("No annotator for {} annotations in {}".format(
arg, self.language()))
def language(self):
if LANGUAGE in self.attributes:
return self.attributes[LANGUAGE]
return lng.UNKNOWN
@staticmethod
def from_spacy(parsed):
document = Document(content=str(parsed))
for token in parsed:
if token.lemma_.strip() != "":
t = document.create_annotation(
"token", token.idx, token.idx + len(token),
[(type.INDEX, token.i), (type.LEMMA, token.lemma_),
("prob", token.prob),
(type.PART_OF_SPEECH, PartOfSpeech.of(token.tag_))])
if token.head is token:
head_idx = None
else:
head_idx = token.head.i
if head_idx:
t.add_relation(target=head_idx,
type="dep",
relation=token.dep_)
for entity in parsed.ents:
document.create_annotation(type.ENTITY, entity.start_char,
entity.end_char,
[(type.ENTITY_TYPE, entity.label_)])
for i, sentence in enumerate(parsed.sents):
document.create_annotation(type.SENTENCE, sentence.start_char,
sentence.end_char, [(type.INDEX, i)])
for np in parsed.noun_chunks:
document.create_annotation(
type.PHRASE_CHUNK, np.start_char, np.end_char,
[(type.PART_OF_SPEECH, PennTreebank.NP)])
@staticmethod
def from_json(json_str):
doc = Document(content='')
doc.__read_json(json.loads(json_str))
return doc
def __getstate__(self):
return self.to_dict()
def __setstate__(self, state):
self.__read_json(state)
def __read_json(self, obj):
self.__init__(content=obj['content'])
self._doc_id = obj.get('id', self._doc_id)
for (k, v) in obj.get("attributes", {}).items():
self[k] = get_decoder(k)(v)
for (k, v) in obj.get('completed', {}).items():
self._completed[k] = v
max_id = -1
for annotation in obj.get("annotations", []):
ann = Annotation(
document=self,
start=annotation["start"],
end=annotation["end"],
annotation_type=annotation["type"],
attributes=[
(k, get_decoder(k)(v))
for k, v in annotation.get("attributes", {}).items()
],
annotation_id=annotation["id"])
max_id = max(max_id, ann.annotation_id)
self._annotations.add(ann.start, ann.end, ann)
#self._annotations.add(ann)
for rel in annotation.get("relations", []):
ann.add_relation(target=rel["target"],
type=rel["type"],
relation=rel["value"])
self.language().load()
self._next_id = max_id + 1
def to_json(self) -> str:
return json.dumps(self.to_dict(), default=default)
def to_dict(self) -> typing.Dict[str, typing.Any]:
return dict([
("id", self._doc_id), ("content", self.content),
("attributes", self._attributes), ("completed", self._completed),
("annotations",
[a.as_dict() for a in self.annotation(annotation_type=None)])
])
def setUp(self):
self.tree4 = IntervalTree()
self.tree4.insert(Interval(22, 33, data='example1'))
self.tree4.insert(Interval(22, 33, data='example2'))
def setUp(self):
self.tree = IntervalTree()
def setUp(self):
self.tree4 = IntervalTree()
self.tree4.insert(Interval(22, 33, data='example1'))
self.tree4.insert(Interval(22, 33, data='example2'))
from matplotlib import pyplot as plt
from matplotlib_venn import venn3, venn3_circles, venn3_unweighted
from itertools import groupby
from operator import itemgetter
import numpy as np
# create list of tuple intervals for each ccr, pli, and missense z interval and a tree of those intervals
# search each tree with both lists
ccrs = open(sys.argv[1], "rb") #exacresiduals/gnomad10x.5syn-ccrs.bed.gz
mpcs = open(sys.argv[2], "rb") #essentials/mpc.regions.clean.sorted.bed.gz
plis = open(sys.argv[3],
"rb") #$HOME/software/pathoscore/score-sets/GRCh37/pLI/pLI.bed.gz
# generate data: we want to search the trees with the lists to get the numbers in each venn diagram
ccrtree = defaultdict(lambda: IntervalTree())
ccrlist = defaultdict(list)
sorter = itemgetter(0, 3, 6)
grouper = itemgetter(0, 3, 6)
ccrtemp = []
ccrgenes = set()
pligenes = set()
mpcgenes = set()
ccr99genes = set()
for ccr in ccrs:
ccr = ccr.strip().split("\t")
if float(ccr[-1]) < 95: continue
ccrtemp.append(ccr)
for key, grp in groupby(sorted(ccrtemp, key=sorter), grouper):
grp = list(grp)
chrom = grp[0][0]
def createIntervalTreesFragmentFile():
'''
creates one interval tree for quick lookups
returns
fragmentsMap[fragmentId] = [tuple(chrom, fragmentMidPoint)]
intersect_tree - intersect Tree for interval matching
'''
if (options.verbose):
print >> sys.stdout, "- %s START : populate intervaltree from fragmented genome" % (timeStamp())
intersect_tree = IntervalTree()
fragmentsCount = 0
fragmentsMap = {}
fragmentsChrom = {} # lookp table for fragment ranges of a chromosome
fragmentsStart = 0
start = 0
end = 0
counter = 0
chrom = ""
for line in fileinput.input([options.fragmentFile]):
line = line.strip()
if (len(line)==0 or line.startswith("Genome") or line.startswith("Chromosome")):
continue
cols = line.split("\t")
try:
# check if chromosome changed from last
if (cols[0] != chrom):
# do we have do finish the last chromosome?
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, int(0.5*(start+end))])
fragmentsCount += 1
fragmentsChrom[chrom] = tuple([fragmentsStart, fragmentsCount])
fragmentsStart = fragmentsCount
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom, start, end)
# check if chromosome needs to be filtered out or not
if (options.chromPattern != "" and not re.match(options.chromPattern, cols[0])):
chrom = ""
start = 0
end = 0
else:
chrom = cols[0]
start = int(cols[1])
end = int(cols[2])
counter = 0
# check if fragment aggregation is fulfilled
elif (counter >= options.fragmentAggregation):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom, start, end)
fragmentsMap[fragmentsCount] = tuple([chrom, int(0.5*(start+end))])
start = int(cols[1])
end = int(cols[2])
counter = 0
fragmentsCount += 1
else:
end = int(cols[2])
# increment counter
counter += 1
except:
if (options.verbose):
print >> sys.stderr, 'skipping line in options.fragmentFile: %s' % (line)
if (options.vverbose):
traceback.print_exc()
sys.exit(1)
# handle last fragment
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, int(0.5*(start+end))])
fragmentsCount += 1
fragmentsChrom[chrom] = tuple([fragmentsStart, fragmentsCount])
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom, start, end)
if (options.verbose):
print >> sys.stdout, "- %s FINISHED: intervaltree populated" % (timeStamp())
return [fragmentsMap, intersect_tree, fragmentsCount, fragmentsChrom]
def getpairs(leftSet,
rightSet,
leftCol,
mincols=1,
asfraction=False,
matchStrand=STRAND_NEUTRAL,
skipChrNames=True,
skipStrandNames=True):
# Read leftSet into memory:
leftlen = 0
rightlen = 0
leftStrandCol = -1
minoverlap = mincols
leftTree = IntervalTree()
rightCols = list()
for item in leftSet:
if type(item) is GenomicInterval:
leftTree.insert(item, leftSet.linenum, item.fields)
if leftlen == 0: leftlen = item.nfields
if leftStrandCol == -1: leftStrandCol = item.strand_col
for interval in rightSet:
if rightlen == 0 and type(interval) is GenomicInterval:
rightlen = interval.nfields
rightCols = range(rightlen)
#remove the useless columns
rightCols.remove(interval.start_col)
rightCols.remove(interval.end_col)
if skipChrNames:
rightCols.remove(interval.chrom_col)
if skipStrandNames:
rightCols.remove(interval.strand_col)
if not (type(interval) is GenomicInterval):
yield interval
else:
result = []
leftTree.intersect(interval, lambda node: result.append(node))
overlap_not_met = 0
rightbases = interval.end - interval.start
for item in result:
leftbases = item.end - item.start
if (asfraction == True):
if leftbases < rightbases:
mincols = leftbases
else:
mincols = rightbases
mincols = math.floor(mincols * minoverlap)
if (item.start >= interval.start and item.start <= interval.end
) and (item.end < interval.start
or item.end > interval.end):
overlap = interval.end - item.start
elif (item.end >= interval.start and item.end <= interval.end
) and (item.start < interval.start
or item.end > interval.end):
overlap = item.end - interval.start
elif item.start >= interval.start and item.start <= interval.end and item.end >= interval.start and item.end <= interval.end:
overlap = item.end - item.start
else: #the intersecting item's start and end are outside the interval range
overlap = interval.end - interval.start
if overlap < mincols:
overlap_not_met += 1
continue
else:
#check strand
strandMatched = STRAND_INTEGER_VALUES[
interval.strand] * STRAND_INTEGER_VALUES[
item.other[leftStrandCol]]
if (strandMatched == -1 and matchStrand > 0):
#needed match but found a complement
overlap_not_met += 1
continue
if (strandMatched == 1 and matchStrand < 0):
#needed complement but found a match
overlap_not_met += 1
continue
if (strandMatched == 0
and (matchStrand < -1 or matchStrand > 1)):
#strict criteria but only permissive match found
overlap_not_met += 1
continue
#strand criteria met
setattr(item, "visited", True)
leftTerm = item.other[leftCol]
for col in rightCols:
#take each field that's not a number
#split it on semicolons, commas, and spaces
#output the word and the leftTerm as being associated
#curcol = re.sub("\;|\,","\t",interval.fields[col])
curcol = interval.fields[col]
lexer = shlex.shlex(curcol)
lexer.whitespace = '\t\r\n\,\;'
lexer.wordchars += ":'"
lexer.whitespace_split = True
lexer.quotes = '"'
for item in lexer:
item = item.strip()
if (item == "."): continue
try:
float(item)
except ValueError:
yield [item, leftTerm]
def join(leftSet, rightSet, mincols=1, leftfill=True, rightfill=True):
# Read rightSet into memory:
rightlen = 0
leftlen = 0
rightTree = IntervalTree()
for item in rightSet:
if isinstance(item, GenomicInterval):
rightTree.insert(item, rightSet.linenum, item.fields)
if rightlen == 0: rightlen = item.nfields
for interval in leftSet:
if leftlen == 0 and isinstance(interval, GenomicInterval):
leftlen = interval.nfields
if not isinstance(interval, GenomicInterval):
yield interval
else:
result = []
rightTree.intersect(interval, lambda node: result.append(node))
overlap_not_met = 0
for item in result:
if item.start in range(interval.start, interval.end +
1) and item.end not in range(
interval.start, interval.end + 1):
overlap = interval.end - item.start
elif item.end in range(interval.start, interval.end +
1) and item.start not in range(
interval.start, interval.end + 1):
overlap = item.end - interval.start
elif item.start in range(interval.start, interval.end +
1) and item.end in range(
interval.start, interval.end + 1):
overlap = item.end - item.start
else: #the intersecting item's start and end are outside the interval range
overlap = interval.end - interval.start
if overlap < mincols:
overlap_not_met += 1
continue
outfields = list(interval)
map(outfields.append, item.other)
setattr(item, "visited", True)
yield outfields
if (len(result) == 0
or overlap_not_met == len(result)) and rightfill:
outfields = list(interval)
for x in range(rightlen):
outfields.append(".")
yield outfields
if leftfill:
def report_unvisited(node, results):
if not hasattr(node, "visited"):
results.append(node)
results = []
rightTree.traverse(lambda x: report_unvisited(x, results))
for item in results:
outfields = list()
for x in range(leftlen):
outfields.append(".")
map(outfields.append, item.other)
yield outfields
b = snakemake.input.background
background = pd.read_table(b,
sep="\t",
usecols=[1, 2],
header=None,
names="Start End".split(),
engine="c",
dtype={
"Start": np.int32,
"End": np.int32
})
start = time()
tree = IntervalTree()
for start_, end_ in zip(background.Start, background.End):
tree.add(start_, end_)
end = time()
# print(result)
total = end - start
total_dt = datetime.datetime.fromtimestamp(total)
minutes_seconds = total_dt.strftime('%-M.%-S.%f')
open(snakemake.output[0], "w+").write(minutes_seconds)
def createIntervalTrees():
'''
creates one interval tree for quick lookups
returns
fragmentsMap[fragmentId] = [tuple(chrom, fragmentMidPoint)]
intersect_tree - intersect Tree for interval matching
'''
if (options.verbose):
print >> sys.stdout, "- %s START : populate intervaltree from fragmented genome" % (
timeStamp())
intersect_tree = IntervalTree()
fragmentsCount = 0
fragmentsMap = {}
start = 0
end = 0
counter = 0
chrom = ""
for line in fileinput.input([options.fragmentFile]):
line = line.strip()
if (len(line) == 0 or line.startswith("Genome")
or line.startswith("Chromosome")):
continue
cols = line.split("\t")
try:
# check if chromosome changed from last
if (cols[0] != chrom):
# do we have do finish the last chromosome?
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple(
[chrom, int(0.5 * (start + end))])
fragmentsCount += 1
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (
chrom, start, end)
chrom = cols[0]
start = int(cols[1])
end = int(cols[2])
counter = 0
# check if fragement aggregation is fulfilled
elif (counter >= options.fragmentAggregation):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (
chrom, start, end)
fragmentsMap[fragmentsCount] = tuple(
[chrom, int(0.5 * (start + end))])
start = int(cols[1])
end = int(cols[2])
counter = 0
fragmentsCount += 1
else:
end = int(cols[2])
# increment counter
counter += 1
except:
if (options.verbose):
print >> sys.stderr, 'skipping line in options.fragmentFile: %s' % (
line)
if (options.vverbose):
traceback.print_exc()
# handle last fragment
if (end > 0):
interval = Interval(chrom, start, end)
intersect_tree.insert(interval, fragmentsCount)
fragmentsMap[fragmentsCount] = tuple([chrom, int(0.5 * (start + end))])
fragmentsCount += 1
if (options.vverbose):
print >> sys.stdout, "-- intervaltree.add %s:%d-%d" % (chrom,
start, end)
if (options.verbose):
print >> sys.stdout, "- %s FINISHED: intervaltree populated" % (
timeStamp())
return [fragmentsMap, intersect_tree]
def getpairs(leftSet, rightSet, leftCol, mincols=1, asfraction=False, matchStrand=STRAND_NEUTRAL, skipChrNames=True, skipStrandNames=True):
# Read leftSet into memory:
leftlen = 0
rightlen = 0
leftStrandCol = -1
minoverlap = mincols
leftTree = IntervalTree()
rightCols = list()
for item in leftSet:
if type( item ) is GenomicInterval:
leftTree.insert( item, leftSet.linenum, item.fields )
if leftlen == 0: leftlen = item.nfields
if leftStrandCol == -1: leftStrandCol = item.strand_col
for interval in rightSet:
if rightlen == 0 and type( interval ) is GenomicInterval:
rightlen = interval.nfields
rightCols = range(rightlen)
#remove the useless columns
rightCols.remove( interval.start_col )
rightCols.remove( interval.end_col )
if skipChrNames:
rightCols.remove( interval.chrom_col )
if skipStrandNames:
rightCols.remove( interval.strand_col )
if not (type( interval ) is GenomicInterval):
yield interval
else:
result = []
leftTree.intersect( interval, lambda node: result.append( node ) )
overlap_not_met = 0
rightbases = interval.end - interval.start
for item in result:
leftbases = item.end - item.start
if (asfraction==True):
if leftbases < rightbases:
mincols = leftbases
else:
mincols = rightbases
mincols = math.floor(mincols * minoverlap)
if (item.start >= interval.start and item.start <= interval.end) and (item.end < interval.start or item.end > interval.end):
overlap = interval.end-item.start
elif (item.end >= interval.start and item.end <= interval.end) and (item.start < interval.start or item.end > interval.end):
overlap = item.end-interval.start
elif item.start >= interval.start and item.start <= interval.end and item.end >= interval.start and item.end <= interval.end:
overlap = item.end-item.start
else: #the intersecting item's start and end are outside the interval range
overlap = interval.end-interval.start
if overlap < mincols:
overlap_not_met += 1
continue
else:
#check strand
strandMatched = STRAND_INTEGER_VALUES[interval.strand] * STRAND_INTEGER_VALUES[item.other[leftStrandCol]]
if (strandMatched == -1 and matchStrand > 0):
#needed match but found a complement
overlap_not_met += 1
continue
if (strandMatched == 1 and matchStrand < 0):
#needed complement but found a match
overlap_not_met += 1
continue
if (strandMatched == 0 and (matchStrand < -1 or matchStrand > 1)):
#strict criteria but only permissive match found
overlap_not_met += 1
continue
#strand criteria met
setattr( item, "visited", True )
leftTerm = item.other[leftCol]
for col in rightCols:
#take each field that's not a number
#split it on semicolons, commas, and spaces
#output the word and the leftTerm as being associated
#curcol = re.sub("\;|\,","\t",interval.fields[col])
curcol= interval.fields[col]
lexer = shlex.shlex(curcol)
lexer.whitespace='\t\r\n\,\;'
lexer.wordchars += ":'"
lexer.whitespace_split=True
lexer.quotes='"'
for item in lexer:
item = item.strip()
if (item == "."): continue
try:
float(item)
except ValueError:
yield [item, leftTerm]
