def _next(self, line):
if line.startswith('#'):
return
ge = GenomeElement(self._genome)
cols = line.split('\t')
if self._numCols is not None:
if len(cols) != self._numCols:
raise InvalidFormatError('Error: BED files must have the same number of columns in each data line.')
else:
self._numCols = len(cols)
if self._numCols < self.MIN_NUM_COLS or self._numCols > self.MAX_NUM_COLS:
raise InvalidFormatError('Error: BED file must contain between %s and %s columns.' % (self.MIN_NUM_COLS, self.MAX_NUM_COLS))
ge.chr = self._checkValidChr(cols[0])
ge.start = self._checkValidStart(ge.chr, int(cols[1]))
self._parseEnd( ge, self._checkValidEnd(ge.chr, int(cols[2]), start=ge.start))
self._parseName( ge, cols )
self._parseVal( ge, cols )
if self._numCols >= 6:
ge.strand = self._getStrandFromString(cols[5])
for i,extraCol in enumerate(self.BED_EXTRA_COLUMNS):
if self._numCols >= i+7:
setattr(ge, extraCol, cols[i+6])
return ge
def _next(self, line):
cols = line.split('\t')
ge = GenomeElement(self._genome)
ge.chr = self._checkValidChr(cols[0])
ge.start = int(cols[1])
ge.end = int(cols[2])
self._parseVal(ge, cols[3])
return ge
def _next(self, line):
cols = line.split('\t')
ge = GenomeElement(self._genome)
ge.chr = self._checkValidChr(cols[0])
ge.start = int(cols[1])
ge.end = int(cols[2])
self._parseVal(ge, cols[3])
return ge
def _next(self, line):
if line.startswith('>'):
self._appendBoundingRegionTuple()
self._elCount = 0
self._chr = self._checkValidChr(line[1:].split()[0])
else:
if self._chr is None:
raise InvalidFormatError('FASTA file does not start with the ">" character.')
self._elCount += len(line)
ge = GenomeElement(self._genome, self._chr)
ge.val = np.fromstring(line, dtype='S1')
return ge
def _next(self, line):
if line.startswith('>'):
self._appendBoundingRegionTuple()
self._elCount = 0
self._chr = self._checkValidChr(line[1:].split()[0])
else:
if self._chr is None:
raise InvalidFormatError(
'FASTA file does not start with the ">" character.')
self._elCount += len(line)
ge = GenomeElement(self._genome, self._chr)
ge.val = np.fromstring(line, dtype='S1')
return ge
def next(self):
self._curPos += 1
if self._curPos % 10e6 == 0:
print '.',
if self._curPos >= len(self._tv.genomeAnchor):
raise StopIteration
if self._exhausted:
return None
if self._curEl is None:
try:
self._curEl = self._tvIter.next()
except StopIteration:
self._exhausted = True
return None
if self._curPos == self._curEl.start():
trackEl = self._curEl
genome = self._tv.genomeAnchor.genome
chr = self._tv.genomeAnchor.chr
#print 'EL: ',GenomeElement(genome,chr, trackEl.start(), trackEl.end(), trackEl.val(), trackEl.strand())
outEl = GenomeElement(genome, chr, trackEl.start(), trackEl.end(),
trackEl.val(), trackEl.strand())
self._curEl = None
return outEl
else:
#print self._curPos,' AND ', self._curEl.start()
#print 'None'
return None
def _getIter(elList,
valDataType,
valDim,
edgeWeightDataType,
edgeWeightDim,
brList=[]):
geIter = MyGeIter(valDataType, valDim, edgeWeightDataType, edgeWeightDim)
for i in xrange(len(elList)):
ge = GenomeElement(genome=elList[i][0],
chr=elList[i][1],
start=elList[i][2],
end=elList[i][3])
if len(elList[i]) == 5:
for prefix in elList[i][4]:
setattr(ge, prefix, elList[i][4][prefix])
geIter.iter.append(ge)
for i in xrange(len(brList)):
br = GenomeRegion(genome=brList[i][0],
chr=brList[i][1],
start=brList[i][2],
end=brList[i][3])
geIter.boundingRegionTuples.append(
BoundingRegionTuple(br, brList[i][4]))
return geIter
def _parseDeclarationLine(self, line):
returnGE = None
chr, start, step, span = self._getDeclarationLineAttrValues(line)
self._fixedStep = self._checkFixedStep(line, start, step)
chr = self._handleChr(chr)
self._span = self._handleSpan(span)
self._isPoints = self._span == 1
if self._fixedStep:
start = self._handleStart(chr, start)
self._step = self._handleStep(step)
self._isStepFunction = (self._step == self._span
and self._step > 1)
self._isFunction = (self._step == self._span and self._step == 1)
if self._isFunction:
self._genomeElement.chr = chr
if not self._shouldExpandBoundingRegion(chr, start):
if self._chr is not None: #self._chr is still the chromosome of the previous decl. line
self._appendBoundingRegionTuple()
self._start = start
self._curElCountInBoundingRegion = 0
if self._isStepFunction:
returnGE = GenomeElement(genome=self._genome, chr=chr, end=self._start, \
val=numpy.nan, isBlankElement=True)
self._chr = chr
return returnGE
def next(self):
if self._returnedOneElement:
raise StopIteration
self._returnedOneElement = True
return GenomeElement(genome=self._genome,
chr=self._region.chr,
val=self._valSlice)
def testAssignAndRetrieve(self):
e = GenomeElement('TestGenome', start=5, val=1.0, extra={'a':1,'b':2}, orderedExtraKeys=['a','b'])
self.assertEqual(e.genome, 'TestGenome')
self.assertEqual(e.chr, None)
self.assertEqual(e.start, 5)
self.assertEqual(e.end, None)
self.assertEqual(e.val, 1.0)
self.assertEqual(e.strand, None)
self.assertEqual(e.a, 1)
self.assertEqual(e.b, 2)
self.assertEqual(e.extra, {'a':1,'b':2})
self.assertEqual(e.orderedExtraKeys, ['a', 'b'])
e = GenomeElement('TestGenome', a=1)
e.b = 2
self.assertEqual(e.genome, 'TestGenome')
self.assertEqual(e.a, 1)
self.assertEqual(e.b, 2)
self.assertEqual(e.extra, {'a':1,'b':2})
self.assertEqual(e.orderedExtraKeys, ['a', 'b'])
self.assertRaises(AttributeError, lambda : e.nonExisting)
def testAssignAndRetrieve(self):
e = GenomeElement("TestGenome", start=5, val=1.0, extra={"a": 1, "b": 2}, orderedExtraKeys=["a", "b"])
self.assertEqual(e.genome, "TestGenome")
self.assertEqual(e.chr, None)
self.assertEqual(e.start, 5)
self.assertEqual(e.end, None)
self.assertEqual(e.val, 1.0)
self.assertEqual(e.strand, None)
self.assertEqual(e.a, 1)
self.assertEqual(e.b, 2)
self.assertEqual(e.extra, {"a": 1, "b": 2})
self.assertEqual(e.orderedExtraKeys, ["a", "b"])
e = GenomeElement("TestGenome", a=1)
e.b = 2
self.assertEqual(e.genome, "TestGenome")
self.assertEqual(e.a, 1)
self.assertEqual(e.b, 2)
self.assertEqual(e.extra, {"a": 1, "b": 2})
self.assertEqual(e.orderedExtraKeys, ["a", "b"])
self.assertRaises(AttributeError, lambda: e.nonExisting)
def _next(self, line):
if line.startswith('##FASTA'):
raise StopIteration
if len(line)>0 and line[0]=='#':
return None
origCols = line.split('\t')
cols = [unquote(x) for x in origCols]
if len(cols) != 9:
raise InvalidFormatError("Error: GFF files must contain 9 tab-separated columns")
ge = GenomeElement(self._genome)
ge.chr = self._checkValidChr(cols[0])
ge.source = cols[1]
self._parseThirdCol(ge, cols[2])
ge.start = self._checkValidStart(ge.chr, int(cols[3]) - 1)
ge.end = self._checkValidEnd(ge.chr, int(cols[4]), start=ge.start)
self._parseSixthCol(ge, cols[5])
ge.strand = self._getStrandFromString(cols[6])
ge.phase = cols[7]
ge.attributes = cols[8]
for attr in origCols[8].split(';'):
attrSplitted = attr.split('=')
if len(attrSplitted) == 2:
key, val = attrSplitted
if key.lower() == 'id':
ge.id = unquote(val)
elif key.lower() == 'name':
ge.name = unquote(val)
return ge
def __init__(self,
fn,
genome=None,
trackName=None,
external=False,
printWarnings=True,
strToUseInsteadOfFn='',
*args,
**kwArgs): #, depth=0
self._fn = fn
self._genome = genome
self._genomeElement = GenomeElement(genome)
self._trackName = trackName
self._external = external
self._prefixList = None
self._printWarnings = printWarnings
self._strToUseInsteadOfFn = strToUseInsteadOfFn
self._lastWarning = None
def next(self):
self._index += 1
if self._index >= len(self):
raise StopIteration
return GenomeElement(start=self._startList[self._index]
if self._index < len(self._startList) else None,
end=self._endList[self._index]
if self._index < len(self._endList) else None,
strand=self._strandList[self._index]
if self._index < len(self._strandList) else None,
val=self._valList[self._index]
if self._index < len(self._valList) else None,
id=self._idList[self._index]
if self._index < len(self._idList) else None,
edges=self._edgesList[self._index]
if self._index < len(self._edgesList) else None,
weights=self._weightsList[self._index]
if self._index < len(self._weightsList) else None,
extra=self._extraList[self._index]
if self._index < len(self._extraList) else None)
def _next(self, line):
if self._isDeclarationLine(line):
ge = self._parseDeclarationLine(line)
if ge is not None:
return ge
else:
if line.startswith('#'):
return None
cols = line.split()
self._checkDataLineCols(cols)
if self._fixedStep:
self._curElCountInBoundingRegion += 1
val = numpy.float(self._handleNan(cols[0]))
if self._isFunction:
self._genomeElement.val = val
return self._genomeElement
else:
start = self._checkValidStart(
self._chr, self._getFixedStepCurElStart())
else:
start = self._checkValidStart(self._chr, int(cols[0]) - 1)
val = numpy.float(self._handleNan(cols[1]))
end = None
if not self._isPoints:
end = self._checkValidEnd(self._chr, self._getEnd(start),
start)
if self._isStepFunction:
start = None
return GenomeElement(genome=self._genome,
chr=self._chr,
start=start,
end=end,
val=val)
def _wrappedTrackElsGenerator(self):
track = self._getTrack()
for region,tv in ((region, self._getTrackView(track, region)) for region in self._boundingRegions):
for te in tv:
yield GenomeElement.createGeFromTrackEl(te, tv.trackFormat, globalCoords=self._globalCoords)
def testExclude(self):
self.assertEqual([GenomeElement('TestGenome','chr21',100,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',90,100) ))
self.assertEqual([GenomeElement('TestGenome','chr21',100,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',200,210) ))
self.assertEqual([GenomeElement('TestGenome','chr21',100,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chrM',100,110) ))
self.assertEqual([GenomeElement('TestGenome','chr21',110,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',100,110) ))
self.assertEqual([GenomeElement('TestGenome','chr21',110,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',90,110) ))
self.assertEqual([GenomeElement('TestGenome','chr21',100,190)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',190,200) ))
self.assertEqual([GenomeElement('TestGenome','chr21',100,190)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',190,210) ))
self.assertEqual([],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',90,210) ))
self.assertEqual([GenomeElement('TestGenome','chr21',100,140), GenomeElement('TestGenome','chr21',160,200)],\
GenomeElement('TestGenome','chr21',100,200).exclude( GenomeElement('TestGenome','chr21',140,160) ))
def testExtend(self):
self.assertEqual(GenomeElement('TestGenome','chr21',100,200),\
GenomeElement('TestGenome','chr21',100,200).extend( 0 ))
self.assertEqual(GenomeElement('TestGenome','chr21',0,200),\
GenomeElement('TestGenome','chr21',100,200).extend( -100 ))
self.assertEqual(GenomeElement('TestGenome','chr21',-100,200),\
GenomeElement('TestGenome','chr21',100,200).extend( -200, ensureValidity=False ))
self.assertEqual(GenomeElement('TestGenome','chr21',0,200),\
GenomeElement('TestGenome','chr21',100,200).extend( -200, ensureValidity=True ))
self.assertEqual(GenomeElement('TestGenome','chr21',100,300),\
GenomeElement('TestGenome','chr21',100,200).extend( 100 ))
self.assertEqual(GenomeElement('TestGenome','chr21',100,50000200),\
GenomeElement('TestGenome','chr21',100,200).extend( 50000000, ensureValidity=False ))
self.assertEqual(GenomeElement('TestGenome','chr21',100,46944323),\
GenomeElement('TestGenome','chr21',100,200).extend( 50000000, ensureValidity=True ))
def _wrappedTrackElsGenerator(self):
track = self._getTrack()
for region,tv in ((region, self._getTrackView(track, region)) for region in self._boundingRegions):
for te in tv:
yield GenomeElement.createGeFromTrackEl(te, tv.trackFormat, globalCoords=self._globalCoords)
def testEqual(self):
self.assertEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('NCBI46','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chrM',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',20,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,110,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,6,True,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,False,'id',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,True,'id4',['id2','id3'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id4'],[5,6],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,7],extra={'source':'source'}))
self.assertNotEqual(GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source'}),
GenomeElement('TestGenome','chr21',10,100,5,True,'id',['id2','id3'],[5,6],extra={'source':'source', 'other':'value'}))
def next(self):
trackEl = self._tvIter.next()
ge = GenomeElement.createGeFromTrackEl(trackEl, self._tv.trackFormat)
return ge
def testContains(self):
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',10,100)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',20,80)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',10,101)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',9,100)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',9,101)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chr21',0,10)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).contains( \
GenomeElement('TestGenome','chrM',20,80)))
def testWriteElement(self):
s = SetupDir(self.path, ['start', 'end', 'val', 'strand', 'id', 'edges', 'weights', 'cat'])
ge = GenomeElement()
s.od.writeElement(ge)
for f in s.od._files.values():
self.assertEqual(ge, f.ge)
def testOverlaps(self):
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',10,100)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',20,80)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',10,101)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',9,100)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',9,101)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',0,10)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chr21',100,110)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).overlaps( \
GenomeElement('TestGenome','chrM',20,80)))
def testTouches(self):
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',10,100)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',20,80)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',10,101)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',9,100)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',9,101)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',0,10)))
self.assertTrue(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',100,110)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',0,9)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chr21',101,110)))
self.assertFalse(GenomeElement('TestGenome','chr21',10,100).touches( \
GenomeElement('TestGenome','chrM',20,80)))
def next(self):
trackEl = self._tvIter.next()
ge = GenomeElement.createGeFromTrackEl(trackEl, self._tv.trackFormat)
return ge
def _next(self, line):
if line.startswith('##FASTA'):
raise StopIteration
if len(line) > 0 and line[0] == '#':
return None
origCols = line.split('\t')
cols = [unquote(x) for x in origCols]
if len(cols) != 9:
raise InvalidFormatError(
"Error: GFF files must contain 9 tab-separated columns")
ge = GenomeElement(self._genome)
ge.chr = self._checkValidChr(cols[0])
ge.source = cols[1]
self._parseThirdCol(ge, cols[2])
ge.start = self._checkValidStart(ge.chr, int(cols[3]) - 1)
ge.end = self._checkValidEnd(ge.chr, int(cols[4]), start=ge.start)
self._parseSixthCol(ge, cols[5])
ge.strand = self._getStrandFromString(cols[6])
ge.phase = cols[7]
ge.attributes = cols[8]
for attr in origCols[8].split(';'):
attrSplitted = attr.split('=')
if len(attrSplitted) == 2:
key, val = attrSplitted
if key.lower() == 'id':
ge.id = unquote(val)
elif key.lower() == 'name':
ge.name = unquote(val)
return ge