def createNmerChains(self, n):
for chr in GenomeInfo.getChrList(self._genome):
print 'Creating chains of nmers of length ', n, ' for chromosome ', chr
chrLen = GenomeInfo.getChrLen(self._genome,chr)
chrReg = GenomeRegion( self._genome, chr, 0, chrLen )
seqTV = PlainTrack( GenomeInfo.getSequenceTrackName(self._genome) ).getTrackView(chrReg)
#nmersAsInts = NmerAsIntSlidingWindow(n, FuncValTvWrapper(seqTV))
nmersAsInts = NmerAsIntSlidingWindow(n, seqTV.valsAsNumpyArray())
SameValueIndexChainsFactory.generate( nmersAsInts, chrLen, 4**n, self._createPath(n), chr )
def createNmerChains(self, n):
for chr in GenomeInfo.getChrList(self._genome):
print 'Creating chains of nmers of length ', n, ' for chromosome ', chr
chrLen = GenomeInfo.getChrLen(self._genome, chr)
chrReg = GenomeRegion(self._genome, chr, 0, chrLen)
seqTV = PlainTrack(GenomeInfo.getSequenceTrackName(
self._genome)).getTrackView(chrReg)
#nmersAsInts = NmerAsIntSlidingWindow(n, FuncValTvWrapper(seqTV))
nmersAsInts = NmerAsIntSlidingWindow(n, seqTV.valsAsNumpyArray())
SameValueIndexChainsFactory.generate(nmersAsInts, chrLen, 4**n,
self._createPath(n), chr)
def _createRandomizedNumpyArrays(self, binLen, starts, ends, vals, strands,
ids, edges, weights, extras, region):
intensityTV = PlainTrack(self._trackNameIntensity).getTrackView(region)
if len(intensityTV.valsAsNumpyArray()) == 0:
raise InvalidRunSpecException('Error: No intensity data available for sampling randomized locations in region' + \
str(region) + \
'. Please check that the intensity track was created with the same main track that is being randomized in this analysis.')
#intensityTV = PlainTrack(self._trackNameIntensity).getTrackView(self._origRegion) #Dependence on origRegion is not nice, but not a big problem..
if intensityTV.trackFormat.isDense():
assert intensityTV.trackFormat.isValued('number')
return self._createRandomizedNumpyArraysFromIntensityFunction(binLen, starts, ends, vals, strands, ids, edges, \
weights, extras, intensityTV)
else:
raise NotImplementedError
def _createRandomizedNumpyArrays(self, binLen, starts, ends, vals, strands, ids, edges, weights, extras, origTrackFormat, region):
if self._minimal:
return numpy.array([]), None, None, None, None, None, None, OrderedDict()
intensityTV = PlainTrack(self._trackNameIntensity).getTrackView(region)
if len(intensityTV.valsAsNumpyArray())==0:
raise InvalidRunSpecException('Error: No intensity data available for sampling randomized locations in region' + \
str(region) + \
'. Please check that the intensity track was created with the same main track that is being randomized in this analysis.')
#intensityTV = PlainTrack(self._trackNameIntensity).getTrackView(self._origRegion) #Dependence on origRegion is not nice, but not a big problem..
if intensityTV.trackFormat.isDense():
assert intensityTV.trackFormat.isValued('number')
return self._createRandomizedNumpyArraysFromIntensityFunction(binLen, starts, ends, vals, strands, ids, edges, weights, extras, intensityTV)
else:
raise NotImplementedError
def _createRandomizedNumpyArrays(self, binLen, starts, ends, vals, strands,
ids, edges, weights, extras, region):
referenceTV = PlainTrack(self._trackNameIntensity).getTrackView(
region
) #self._trackNameIntensity based on naming convenience wrt. inheritance
if len(referenceTV.valsAsNumpyArray()) == 0:
raise InvalidRunSpecException('Error: No reference data available for sampling randomized locations in region' + \
str(region) + \
'. Please check that the reference track was created with the same main track that is being randomized in this analysis.')
if referenceTV.trackFormat.isDense():
raise InvalidRunSpecException(
'Error: Cannot sample by distance to reference if reference is a dense track'
)
else:
return self._createRandomizedNumpyArraysFromDistanceToReference(
binLen, starts, ends, vals, strands, ids, edges, weights,
extras, referenceTV)
