def _buildTestTrees(self):
# inputTree splitOnA splitOnB pairwise (A vs B)
# / \ | / \ / \
# A B C D E t1_t2 t1_t3
# | / \ / \ /\ /\ / \ / \
# C D E A B A B A B Q R Q R
# | | | | / \ | | | | | | | |
# t1 t2 t3 t1 D E C t2 C t3 t1 t2 t1 t2
# | | | |
# t2 t3 t1 t1
self.t1 = SingleTrackTS(Track(['t1']), {'field 1': 'value 1'})
self.t2 = SingleTrackTS(Track(['t2']), {
'field 1': 'value 2',
'field 2': '6'
})
self.t3 = SingleTrackTS(Track(['t3']), {
'field 1': 'value 2',
'field 3': 'None'
})
self.inputTree = TrackStructureV2()
self.inputTree['A'] = TrackStructureV2()
self.inputTree['A']['C'] = self.t1
self.inputTree['B'] = TrackStructureV2()
self.inputTree['B']['D'] = self.t2
self.inputTree['B']['E'] = self.t3
# correct result of the input tree splitted on node A
self.splittedOnNodeA = TrackStructureV2()
self.splittedOnNodeA['C'] = TrackStructureV2()
self.splittedOnNodeA['C']['A'] = self.t1
self.splittedOnNodeA['C']['B'] = TrackStructureV2()
self.splittedOnNodeA['C']['B']['D'] = self.t2
self.splittedOnNodeA['C']['B']['E'] = self.t3
# correct result of the input tree splitted on node B
self.splittedOnNodeB = TrackStructureV2()
self.splittedOnNodeB['D'] = TrackStructureV2()
self.splittedOnNodeB['D']['A'] = TrackStructureV2()
self.splittedOnNodeB['D']['A']['C'] = self.t1
self.splittedOnNodeB['D']['B'] = self.t2
self.splittedOnNodeB['E'] = TrackStructureV2()
self.splittedOnNodeB['E']['A'] = TrackStructureV2()
self.splittedOnNodeB['E']['A']['C'] = self.t1
self.splittedOnNodeB['E']['B'] = self.t3
self.pairwiseCombinations = TrackStructureV2()
self.pairwiseCombinations["['t1']_['t2']"] = TrackStructureV2()
self.pairwiseCombinations["['t1']_['t2']"]['query'] = self.t1
self.pairwiseCombinations["['t1']_['t2']"]['reference'] = self.t2
self.pairwiseCombinations["['t1']_['t3']"] = TrackStructureV2()
self.pairwiseCombinations["['t1']_['t3']"]['query'] = self.t1
self.pairwiseCombinations["['t1']_['t3']"]['reference'] = self.t3
self.flatTrackStructure = FlatTracksTS()
self.flatTrackStructure['A'] = self.t1
self.flatTrackStructure['B'] = self.t2
self.flatTrackStructure['C'] = self.t3
def setUp(self):
stsRef1 = SingleTrackTS(Mock(spec=Track), dict(title="trackA"))
ts1 = FlatTracksTS()
ts1["reference"] = stsRef1
ts1.result = 0.5
ts1["query"] = Mock(spec=SingleTrackTS)
stsRef2 = SingleTrackTS(Mock(spec=Track), dict(title="trackB"))
ts2 = FlatTracksTS()
ts2["reference"] = stsRef2
ts2.result = 1.2
ts2["query"] = Mock(spec=SingleTrackTS)
self.resultTS = TrackStructureV2()
self.resultTS['0'] = ts1
self.resultTS['1'] = ts2
def setUp(self):
genomeAnchor = [0, 101]
self.bins = [
GenomeRegion('TestGenome', 'chr21', 0, 50),
GenomeRegion('TestGenome', 'chr21', 50, 101)
]
self.ts = defaultdict(TrackStructureV2)
allSegmentSetNoOverlap = [[[10, 20]],
[[5, 10], [25, 50], [55, 70], [75, 80],
[90, 95]],
[[5, 15], [20, 30], [45, 55], [75, 80]],
[[0, 20], [60, 65], [85, 100]]]
allSegmentSetOverlap = [[[10, 25]],
[[5, 10], [20, 50], [55, 70], [75, 80],
[90, 95]],
[[5, 20], [15, 30], [45, 55], [75, 80]],
[[0, 20], [60, 65], [85, 100]]]
excludedSegments = [[45, 55], [70, 75]]
for allowOverlaps, allSegmentSet in \
zip([False, True], [allSegmentSetNoOverlap, allSegmentSetOverlap]):
for i, segmentSet in enumerate(allSegmentSet):
track = SampleTrack(
SampleTV(segments=segmentSet,
anchor=genomeAnchor,
allowOverlaps=allowOverlaps))
track.getUniqueKey = MagicMock(return_value=i)
self.ts[allowOverlaps]['dummy' + str(i)] = SingleTrackTS(
track, {})
track = SampleTrack(
SampleTV(segments=[],
anchor=genomeAnchor,
allowOverlaps=allowOverlaps))
track.getUniqueKey = MagicMock(return_value=i + 1)
self.ts[allowOverlaps]['emptyTrack'] = SingleTrackTS(track, {})
exclusionTrack = SampleTrack(
SampleTV(segments=excludedSegments,
anchor=genomeAnchor,
allowOverlaps=False))
self.exclusionTs = SingleTrackTS(exclusionTrack, {})
def getFlatTracksTS(genome, guiSelectedGSuite):
ts = FlatTracksTS()
gsuite = getGSuiteFromGalaxyTN(guiSelectedGSuite)
for gsTrack in gsuite.allTracks():
assert gsTrack.trackName is not None, "Gstrack name is None %s" % gsTrack
track = PlainTrack(gsTrack.trackName)
metadata = OrderedDict(title=gsTrack.title, genome=str(genome))
metadata.update(gsTrack.attributes)
assert track is not None
assert metadata is not None
ts[gsTrack.title] = SingleTrackTS(track, metadata)
return ts
def setUp(self):
"""
Create a fake trackstructure containing empty tracks,
this is only used to ensure the randomization pool knows there should be 3 input tracks.
"""
genomeAnchor = [0, 101]
self.amountTracks = 3
self.region = GenomeRegion('TestGenome', 'chr21', genomeAnchor[0],
genomeAnchor[1])
self.ts = TrackStructureV2()
for i, segmentSet in enumerate([[]] * self.amountTracks):
track = SampleTrack(
SampleTV(segments=segmentSet, anchor=genomeAnchor))
track.getUniqueKey = MagicMock(return_value=i)
self.ts[str(i)] = SingleTrackTS(track, {})
self.shufflePool = ShuffleElementsBetweenTracksPool(
self.ts, self.region, None)
def getSingleTrackTS(genome, guiSelectedTrack, title='Dummy'):
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, guiSelectedTrack)
return SingleTrackTS(PlainTrack(trackName), {'title': title})
def create_track(file_name, trackName):
from gtrackcore.core.Api import importFile
importFile(file_name, genome="hg18", trackName=trackName)
t = PlainTrack([trackName])
single_track_ts = SingleTrackTS(t, {"title": trackName})
return single_track_ts
from gold.track.TrackStructure import FlatTracksTS, SingleTrackTS, TrackStructureV2
class MockTrack(Track):
def __new__(cls, *args):
return object.__new__(cls)
def __init__(self, name):
object.__init__(self)
self.trackName = [
name
] #because needed by hash in makeTreeSegregatedByCategory
self.trackTitle = name
q = SingleTrackTS(MockTrack('t1'), OrderedDict(title='t1', cat='q'))
flat = FlatTracksTS()
flat['t2'] = SingleTrackTS(MockTrack('t2'), OrderedDict(title='t2', cat='C1'))
flat['t3'] = SingleTrackTS(MockTrack('t3'), OrderedDict(title='t3', cat='C1'))
flat['t4'] = SingleTrackTS(MockTrack('t4'), OrderedDict(title='t4', cat='C2'))
r = flat.getSplittedByCategoryTS('cat')
orig = TrackStructureV2()
orig['q'] = q
orig['r'] = r
def allInOne():
print 'Orig: ', orig