def plainRun(analysisDef, genome, track1Fn, track2Fn):
'''
Currently under development, not yet functioning.
Note that file names (track1Fn, track2Fn) must either be a valid .dat-file from a Galaxy system,
or be located in a path where the two deepest levels are numerical values above 1000, and unique between different runs by this tool'''
#GalaxyInterface.runManual(trackNames, analysisDef, regSpec, binSpec, genome, galaxyFn=None, trackNameIntensity=None, username='', \
#printResults=True, printProgress=True, printHtmlWarningMsgs=True, applyBoundaryFilter=False, printRunDescription=True, **kwArgs):
from quick.application.ExternalTrackManager import ExternalTrackManager
#fixme: is this correct?
assert all(
idPart > 1000
for idPart in ExternalTrackManager.extractIdFromGalaxyFn(track1Fn))
tn1 = ExternalTrackManager.constructGalaxyTnFromSuitedFn(track1Fn)
assert all(
idPart > 1000
for idPart in ExternalTrackManager.extractIdFromGalaxyFn(track2Fn))
tn2 = ExternalTrackManager.constructGalaxyTnFromSuitedFn(track2Fn)
from quick.application.GalaxyInterface import GalaxyInterface
userBinSource = GalaxyInterface._getUserBinSource('chrs', '*', genome, tn1,
tn2)
from gold.application.StatRunner import AnalysisDefJob
job = AnalysisDefJob(analysisDef, tn1, tn2, userBinSource)
result = job.run(printProgress=False)
return result
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)
def doAnalysisFromDefTwo(cls, tracks, analysisDef, regSpec, binSpec,
genome, flag):
userBinSource = UserBinSource(regSpec, binSpec, genome)
job = AnalysisDefJob(analysisDef.getDefAfterChoices(),
tracks[0].trackName,
tracks[1].trackName,
userBinSource,
galaxyFn=None)
result = job.run()
unorderedResults = [(str(key), result[key])
for key in result.getAllRegionKeys()]
return list(cls.sortDictionaries(unorderedResults, flag))
def doAnalysis(analysisSpec, analysisBins, tracks):
'''Performs an analysis,
as specified by analysisSpec object,
in each bin specified by analysisBins,
on data sets specified in tracks.
Typical usage:
analysisSpec = AnalysisSpec(AvgSegLenStat)
analysisSpec.addParameter("withOverlaps","no")
analysisBins = GlobalBinSource('hg18')
tracks = [ Track(['Genes and gene subsets','Genes','Refseq']) ]
results = doAnalysis(analysisSpec, analysisBins, tracks)
'''
# TODO: handle multiple tracks analysis
# assert len(tracks) in [1,2] #for now..
# in an API setting, exceptions should not generally be hidden.
# Maybe this should be optional.
# setupDebugModeAndLogging()
silenceRWarnings()
silenceNumpyWarnings()
if len(tracks) > 2:
from gold.util.CommonConstants import MULTIPLE_EXTRA_TRACKS_SEPARATOR
analysisSpec.addParameter(
'extraTracks',
MULTIPLE_EXTRA_TRACKS_SEPARATOR.join([
'^'.join([quote(part) for part in x.trackName])
for x in tracks[2:]
]))
job = AnalysisDefJob(analysisSpec.getDefAfterChoices(),
tracks[0].trackName,
tracks[1].trackName if len(tracks) > 1 else None,
analysisBins,
galaxyFn=None)
res = job.run(printProgress=False) # printProgress should be optional?
return res
