def merge(genome, trackName, allowOverlaps):
path = createDirPath(trackName, genome, allowOverlaps=allowOverlaps)
chrList = TrackInfoDataCollector(genome, trackName).getPreProcessedChrs(allowOverlaps)
existingChrList = [chr for chr in ChrMemmapFolderMerger._existingChrIter(path, chrList)]
if len(existingChrList) == 0:
raise EmptyGESourceError('No data lines has been read from source file (probably because it is empty).')
firstChrTrackData = TrackSource().getTrackData(trackName, genome, existingChrList[0], allowOverlaps, forceChrFolders=True)
arrayList = firstChrTrackData.keys()
for arrayName in arrayList:
mergedArray = firstChrTrackData[arrayName][:]
elementDim, dtypeDim = parseMemmapFileFn(firstChrTrackData[arrayName].filename)[1:3]
del firstChrTrackData[arrayName]
for chr in existingChrList[1:]:
chrTrackData = TrackSource().getTrackData(trackName, genome, chr, allowOverlaps, forceChrFolders=True)
mergedArray = ChrMemmapFolderMerger.mergeArrays(mergedArray, np.array(chrTrackData[arrayName][:]))
elementDimNew, dtypeDimNew = parseMemmapFileFn(chrTrackData[arrayName].filename)[1:3]
elementDim = max(elementDim, elementDimNew)
dtypeDim = max(dtypeDim, dtypeDimNew)
del chrTrackData[arrayName]
mergedFn = createMemmapFileFn(path, arrayName, elementDim, dtypeDim, str(mergedArray.dtype))
f = np.memmap(mergedFn, dtype=mergedArray.dtype, mode='w+', shape=mergedArray.shape)
f[:] = mergedArray
f.flush()
del f
del mergedArray
def merge(genome, trackName, allowOverlaps):
path = createDirPath(trackName, genome, allowOverlaps=allowOverlaps)
collector = PreProcMetaDataCollector(genome, trackName)
chrList = collector.getPreProcessedChrs(allowOverlaps)
if not collector.getTrackFormat().reprIsDense():
chrList = sorted(chrList)
existingChrList = [
chr
for chr in ChrMemmapFolderMerger._existingChrIter(path, chrList)
]
if len(existingChrList) == 0:
raise EmptyGESourceError(
'No data lines has been read from source file (probably because it is empty).'
)
firstChrTrackData = TrackSource().getTrackData(trackName,
genome,
existingChrList[0],
allowOverlaps,
forceChrFolders=True)
arrayList = firstChrTrackData.keys()
for arrayName in arrayList:
mergedArray = firstChrTrackData[arrayName][:]
elementDim, dtypeDim = parseMemmapFileFn(
firstChrTrackData[arrayName].filename)[1:3]
del firstChrTrackData[arrayName]
for chr in existingChrList[1:]:
chrTrackData = TrackSource().getTrackData(trackName,
genome,
chr,
allowOverlaps,
forceChrFolders=True)
mergedArray = ChrMemmapFolderMerger.mergeArrays(
mergedArray, np.array(chrTrackData[arrayName][:]))
elementDimNew, dtypeDimNew = parseMemmapFileFn(
chrTrackData[arrayName].filename)[1:3]
elementDim = max(elementDim, elementDimNew)
dtypeDim = max(dtypeDim, dtypeDimNew)
del chrTrackData[arrayName]
mergedFn = createMemmapFileFn(path, arrayName, elementDim,
dtypeDim, str(mergedArray.dtype))
f = np.memmap(mergedFn,
dtype=mergedArray.dtype,
mode='w+',
shape=mergedArray.shape)
f[:] = mergedArray
f.flush()
del f
del mergedArray
def testCreateIntensityTrack(self):
regions = [GenomeRegion(self._genome, self._chr, 1000, 5000),\
GenomeRegion(self._genome, self._chr, 6000, 7000),\
GenomeRegion(self._genome, self._chr, 10000, 16000)]
job = AnalysisDefJob('[dataStat=SimpleBpIntensityStat] [outTrackName=' + '^'.join(self._trackName) + '] [numDiscreteVals=10] -> CreateFunctionTrackStat', \
['nums'], ['points'], regions, genome=self._genome)
for x in range(2):
job.run()
brShelve = BoundingRegionShelve(self._genome,
self._trackName,
allowOverlaps=False)
self.assertRaises(OutsideBoundingRegionError, \
brShelve.getBoundingRegionInfo, GenomeRegion(self._genome, self._chr, 0, 1))
#self.assertEquals(BoundingRegionInfo(0, 1, 0, 0, 0, 0),
# brShelve.getBoundingRegionInfo(GenomeRegion(self._genome, self._chr, 0, 1)))
self.assertEquals(
BoundingRegionInfo(1000, 5000, 0, 4000, 0, 0),
brShelve.getBoundingRegionInfo(
GenomeRegion(self._genome, self._chr, 2000, 2001)))
self.assertRaises(OutsideBoundingRegionError, \
brShelve.getBoundingRegionInfo, GenomeRegion(self._genome, self._chr, 5500, 5501))
#self.assertEquals(BoundingRegionInfo(5500, 5501, 0, 0, 0, 0),
# brShelve.getBoundingRegionInfo(GenomeRegion(self._genome, self._chr, 5500, 5501)))
self.assertEquals(
BoundingRegionInfo(6000, 7000, 4000, 5000, 0, 0),
brShelve.getBoundingRegionInfo(
GenomeRegion(self._genome, self._chr, 6500, 6501)))
self.assertRaises(OutsideBoundingRegionError, \
brShelve.getBoundingRegionInfo, GenomeRegion(self._genome, self._chr, 8000, 8001))
#self.assertEquals(BoundingRegionInfo(8000, 8001, 0, 0, 0, 0),
# brShelve.getBoundingRegionInfo(GenomeRegion(self._genome, self._chr, 8000, 8001)))
self.assertEquals(
BoundingRegionInfo(10000, 16000, 5000, 11000, 0, 0),
brShelve.getBoundingRegionInfo(
GenomeRegion(self._genome, self._chr, 11000, 11001)))
self.assertRaises(OutsideBoundingRegionError, \
brShelve.getBoundingRegionInfo, GenomeRegion(self._genome, self._chr, 16500, 16501))
#self.assertEquals(BoundingRegionInfo(16500, 16501, 0, 0, 0, 0),
# brShelve.getBoundingRegionInfo(GenomeRegion(self._genome, self._chr, 16500, 16501)))
trackData = TrackSource().getTrackData(self._trackName, self._genome,
None, False)
self.assertListsOrDicts(['val'], trackData.keys())
self.assertListsOrDicts((11000, ), trackData['val'].shape)