def execute(choices, galaxyFn=None, username=''):
'''Is called when execute-button is pushed by web-user.
Should print output as HTML to standard out, which will be directed to a results page in Galaxy history.
If getOutputFormat is anything else than HTML, the output should be written to the file with path galaxyFn.
If needed, StaticFile can be used to get a path where additional files can be put (e.g. generated image files).
choices is a list of selections made by web-user in each options box.
'''
genome = choices[0]
trackName = choices[1].split(':')
galaxyOutTrackName = 'galaxy:hbfunction:%s:Create function track of distance to nearest segment' % galaxyFn
outTrackName = ExternalTrackManager.getStdTrackNameFromGalaxyTN(galaxyOutTrackName.split(':'))
if choices[2] == 'No transformation':
valTransformation = 'None'
elif choices[2] =='Logarithmic (log10(x))':
valTransformation = 'log10'
elif choices[2] == 'Fifth square root (x**0.2)':
valTransformation = 'power0.2'
analysisDef ='[dataStat=MakeDistanceToNearestSegmentStat] [valTransformation=%s][outTrackName=' % valTransformation \
+ '^'.join(outTrackName) + '] -> CreateFunctionTrackStat'
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(trackName, None, analysisDef, '*', '*', genome)
#
#for el in userBinSource:
# print el.chr, el.start, el.end
from quick.application.GalaxyInterface import GalaxyInterface
print GalaxyInterface.getHbFunctionOutputBegin(galaxyFn, withDebug=False)
GalaxyInterface.runManual([trackName], analysisDef, '*', '*', genome, username=username, printResults=False, printHtmlWarningMsgs=False)
#job = AnalysisDefJob(analysisDef, trackName, None, userBinSource).run()
print GalaxyInterface.getHbFunctionOutputEnd('A custom track has been created by finding the bp-distance to the nearest segment', withDebug=False)
def execute(choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
genome = choices.genome
seqTrackName = choices.seqTrack.split(':')
outBaseTrackName = choices.outTrackStr.split(':')
regSpec = '__brs__' if choices.regions == 'Bounding regions of the sequence track' else '*'
for pwm, chosen in choices.pwms.iteritems():
if chosen:
pfmFileName = os.sep.join([CreatePwmScoreTracksTool.JASPAR_DIR, pwm + '.pfm'])
outTrackName = outBaseTrackName + [pwm]
analysisDef ='[dataStat=MaxPwmScoreInWindowArrayStat] [pfmFileName=%s] [outTrackName=%s] [valDataType=float32]-> CreateFunctionTrackStat' \
% (pfmFileName.replace('/','^'), '^'.join(outTrackName))
from quick.application.GalaxyInterface import GalaxyInterface
GalaxyInterface.runManual([seqTrackName], analysisDef, regSpec, '*', genome, username=username, printResults=False, printHtmlWarningMsgs=False)
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
genome = choices.genome
seqTrackName = choices.seqTrack.split(':')
outBaseTrackName = choices.outTrackStr.split(':')
regSpec = '__brs__' if choices.regions == 'Bounding regions of the sequence track' else '*'
for pwm, chosen in choices.pwms.iteritems():
if chosen:
pfmFileName = os.sep.join([cls.JASPAR_DIR, pwm + '.pfm'])
outTrackName = outBaseTrackName + [pwm]
analysisDef ='[dataStat=MaxPwmScoreInWindowArrayStat] [pfmFileName=%s] [outTrackName=%s] [valDataType=float32]-> CreateFunctionTrackStat' \
% (pfmFileName.replace('/','^'), '^'.join(outTrackName))
from quick.application.GalaxyInterface import GalaxyInterface
GalaxyInterface.runManual([seqTrackName],
analysisDef,
regSpec,
'*',
genome,
username=username,
printResults=False,
printHtmlWarningMsgs=False)
def findOverrepresentedTFsFromGeneSet(genome, tfSource, ensembleGeneIdList,upFlankSize, downFlankSize, geneSource, galaxyFn):
#galaxyFn = '/usit/insilico/web/lookalike/galaxy_dist-20090924-dev/database/files/003/dataset_3347.dat'
#print 'overriding galaxyFN!: ', galaxyFn
galaxyId = extractIdFromGalaxyFn(galaxyFn)
uniqueWebPath = getUniqueWebPath(extractIdFromGalaxyFn(galaxyFn))
assert genome == 'hg18'
tfTrackNameMappings = TfInfo.getTfTrackNameMappings(genome)
tfTrackName = tfTrackNameMappings[tfSource]
#Get gene track
assert geneSource == 'Ensembl'
targetGeneRegsTempFn = uniqueWebPath + os.sep + 'geneRegs.bed'
geneRegsTrackName = GenomeInfo.getStdGeneRegsTn(genome)
geneRegsFn = getOrigFn(genome, geneRegsTrackName, '.category.bed')
GalaxyInterface.getGeneTrackFromGeneList(genome, geneRegsTrackName, ensembleGeneIdList, targetGeneRegsTempFn )
assert upFlankSize == downFlankSize == 0 #Should instead extend regions to include flanks
tcGeneRegsTempFn = uniqueWebPath + os.sep + 'tcGeneRegs.targetcontrol.bedgraph'
#Think this will be okay, subtraction not necessary as targets are put first:
controlGeneRegsTempFn = geneRegsFn
#print targetGeneRegsTempFn, controlGeneRegsTempFn, tcGeneRegsTempFn
GalaxyInterface.combineToTargetControl(targetGeneRegsTempFn, controlGeneRegsTempFn, tcGeneRegsTempFn)
#tcGeneRegsExternalTN = ['external'] +galaxyId + [tcGeneRegsTempFn]
tcGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(galaxyId, 'tempTc')
#tcGeneRegsExternalTN = ['external'] +targetGalaxyId + [tcGeneRegsTempFn]
#tcGeneRegsExternalTN = ['galaxy', externalId, tcGeneRegsTempFn]
targetGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(galaxyId, 'tempTc', '1')
controlGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(galaxyId, 'tempTc', '0')
#pre-process
print 'Pre-processing file: %s, with trackname: %s ' % (tcGeneRegsTempFn, tcGeneRegsExternalTN)
ExternalTrackManager.preProcess(tcGeneRegsTempFn, tcGeneRegsExternalTN, 'targetcontrol.bedgraph',genome)
print 'Pre-processing TN: ', targetGeneRegsExternalTN
ExternalTrackManager.preProcess(targetGeneRegsTempFn, targetGeneRegsExternalTN, 'bed',genome)
print 'Pre-processing TN: ', controlGeneRegsExternalTN
ExternalTrackManager.preProcess(controlGeneRegsTempFn, controlGeneRegsExternalTN, 'bed',genome)
#print tcGeneRegsExternalTN
trackName1, trackName2 = tfTrackName, tcGeneRegsExternalTN
analysisDef = 'Categories differentially located in targets?: Which categories of track1-points fall more inside case than control track2-segments? [rawStatistic:=PointCountInsideSegsStat:]' +\
'[tf1:=SegmentToStartPointFormatConverter:] [tf2:=TrivialFormatConverter:]' +\
'-> DivergentRowsInCategoryMatrixStat'
regSpec, binSpec = '*','*'
#print 'skipping preproc!!'
#ExternalTrackManager.preProcess(tcGeneRegsExternalTN[-1], tcGeneRegsExternalTN, 'targetcontrol.bedgraph', genome)
#ExternalTrackManager.preProcess(targetGeneRegsTempFn, targetGeneRegsExternalTN, 'bed', genome)
GalaxyInterface.runManual([trackName1, trackName2], analysisDef, regSpec, binSpec, genome, printResults=True, printHtmlWarningMsgs=False)
def execute(cls, choices, galaxyFn=None, username=''):
'''Is called when execute-button is pushed by web-user.
Should print output as HTML to standard out, which will be directed to a results page in Galaxy history.
If getOutputFormat is anything else than HTML, the output should be written to the file with path galaxyFn.
If needed, StaticFile can be used to get a path where additional files can be put (e.g. generated image files).
choices is a list of selections made by web-user in each options box.
'''
# from gold.application.LogSetup import setupDebugModeAndLogging
# setupDebugModeAndLogging()
# Hack to not have to refactor tool to use keyword-based choices
class MockChoices(object):
debugMode = choices[10]
mockChoices = MockChoices()
cls._setDebugModeIfSelected(mockChoices)
analysisDef = choices[2].replace(os.linesep, '')
#from gold.application.StatRunner import AnalysisDefJob
# print 'Preparing arguments'
genome = choices[3]
prefixTN1 = cls.STD_PREFIX_TN if choices[4] == 'yes' else []
tn1 = prefixTN1 + choices[5].split(':')
prefixTN2 = cls.STD_PREFIX_TN if choices[6] == 'yes' else []
tn2 = prefixTN2 + choices[7].split(':')
if tn2[-1] == cls.NO_TRACK_SHORTNAME:
tn2 = None
#from gold.track.GenomeRegion import GenomeRegion
#region = GenomeRegion(genome, 'chr1',1000,2000)
#region2 = GenomeRegion(genome, 'chr1',5000,6000)
#kwArgs = {}
regVal = choices[8]
binSpecVal = choices[9]
#ubSource = UserBinSource(regVal, binSpecVal, genome=genome)
from quick.application.GalaxyInterface import GalaxyInterface
# print tn1, tn2
GalaxyInterface.runManual([tn1, tn2],
analysisDef,
regVal,
binSpecVal,
genome,
galaxyFn,
username=username)
def computeDistance(
cls,
genome,
track1,
track2,
feature,
regSpec,
binSpec,
galaxyFn=None): #direct distance between track1, track2
'''
track1 and track2 are two lists like : ['Sequence','Repeating elements','LINE']
feature specifies how the distance between track1 and track2 is defined
'''
validFeature = DirectDistanceCatalog.getValidAnalyses(
genome, track1, track2)[feature]
analysisDef = validFeature[
0] #'bla bla -> PropFreqOfTr1VsTr2Stat' #or any other statistic from the HB collection
#userBinSource = GalaxyInterface._getUserBinSource(regSpec,binSpec,genome)
result = GalaxyInterface.runManual([track1, track2],
analysisDef,
regSpec,
binSpec,
genome,
galaxyFn=galaxyFn)
#result = AnalysisDefJob(analysisDef, track1, track2, userBinSource).run()
mainResultDict = result.getGlobalResult()
return mainResultDict[validFeature[1]]
def _runSingleStatistic(self, regionTrackName, analysisDef, predictionTrackName, answerTrackName=None):
splittedPredictionTrackName = predictionTrackName.split(':')
regionFileName = regionTrackName.split(':')[2]
if answerTrackName == None:
resultObject = GalaxyInterface.runManual([splittedPredictionTrackName],
analysisDef, 'gtrack', regionFileName, self._genome, username=self._username,
printResults=False, printHtmlWarningMsgs=False)
else:
splittedAnswerTrackName = answerTrackName.split(':')
resultObject = GalaxyInterface.runManual([splittedPredictionTrackName, splittedAnswerTrackName],
analysisDef, 'gtrack', regionFileName, self._genome, username=self._username,
printResults=False, printHtmlWarningMsgs=False)
return resultObject.getGlobalResult()
def extract_feature(cls, genome, track, ref, option, regSpec, binSpec,
trackFormat):
#print 'genome, track, ref, option, regSpec, binSpec, trackFormat: ', genome, track, ref, option, regSpec, binSpec, trackFormat
#featureKeys = FeatureCatalog.getFeaturesFromTracks(genome,track,ref).keys()
#logMessage('ClusterExecution (extract_feature) track og refTrack = ('+ repr(track)+' ##### '+repr(ref))
#logMessage('ClusterExecution (extract_feature) featureKeys = '+ repr(featureKeys))
validFeature = FeatureCatalog.getFeaturesFromTracks(
genome, track, ref
)[option] #validFeature contains analysisDef and the key to get the needed number from the global result
if option == 'Prop. of tr1-points falling inside segments of tr2' and trackFormat in [
'Segments', 'Valued segments'
]:
analysisDef = 'dummy [tf1=SegmentToMidPointFormatConverter] -> DerivedPointCountsVsSegsStat'
else:
analysisDef = validFeature[
0] #or any other statistic from the HB collection
#userBinSource = GalaxyInterface._getUserBinSource(regSpec,binSpec,genome)
result = GalaxyInterface.runManual([ref, track],
analysisDef,
regSpec,
binSpec,
genome,
printResults=False,
printProgress=False,
printHtmlWarningMsgs=False,
printRunDescription=False)
#result = AnalysisDefJob(analysisDef, ref, track, userBinSource).run() if option == 'Prop. of tr2 covered by tr1' else AnalysisDefJob(analysisDef, track, ref, userBinSource).run()
validAnalysisDef = validFeature[1]
assert result.getGlobalResult(
) is not None, 'Did not get any global result for analysisDef: ' + validAnalysisDef
return result.getGlobalResult()[validAnalysisDef]
def execute(cls, choices, galaxyFn=None, username=''):
start = time.time()
genome = choices[0]
trackName = choices[1].split(':')
outFn = galaxyFn
if choices[5] == 'Write to Standardised file':
outFn = createOrigPath(genome, choices[-1].split(':'),
'collapsed_result.bedgraph')
ensurePathExists(outFn[:outFn.rfind('/') + 1])
combineMethod = choices[2]
category = choices[3] if choices[3] else ''
numSamples = choices[4] if choices[4] else '1'
analysisDef = 'dummy [combineMethod=%s] %s [numSamples=%s] -> ConvertToNonOverlappingCategorySegmentsPythonStat' % \
(combineMethod, '[category=%s]' % category if category != '' else '', numSamples) #'Python'
for regSpec in GenomeInfo.getChrList(genome):
res = GalaxyInterface.runManual([trackName], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
from gold.origdata.TrackGenomeElementSource import TrackViewGenomeElementSource
from gold.origdata.BedComposer import CategoryBedComposer
for resDict in res.values():
trackView = resDict['Result']
tvGeSource = TrackViewGenomeElementSource(
genome, trackView, trackName)
CategoryBedComposer(tvGeSource).composeToFile(outFn)
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
from quick.application.ExternalTrackManager import ExternalTrackManager
from quick.application.GalaxyInterface import GalaxyInterface
from gold.application.LogSetup import setupDebugModeAndLogging
#setupDebugModeAndLogging()
genome = choices[0]
tn1, tn2 = choices[1:]
fnTn1 = ExternalTrackManager.extractFnFromGalaxyTN(tn1)
track1 = [
'Sequence', 'DNA'
] #ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, tn1.split(':'))
track2 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, tn2.split(':'))
analysisDef = 'dummy -> GetMutatedSequenceStat'
res = GalaxyInterface.runManual([track1, track2], analysisDef, 'bed',
fnTn1, genome)
with open(galaxyFn, 'w') as utfil:
for region, sequence in res.items():
print >> utfil, '>' + region.chr + ' ' + str(region.start +
1) + '-' + str(
region.end)
print >> utfil, sequence['Result']
def execute(cls, choices, galaxyFn=None, username=''):
start = time.time()
genome = choices[0]
trackName = choices[1].split(':')
outFn = galaxyFn
if choices[5] == 'Write to Standardised file':
outFn = createOrigPath(genome, choices[-1].split(':'), 'collapsed_result.bedgraph')
ensurePathExists(outFn[:outFn.rfind('/')+1])
combineMethod = choices[2]
category = choices[3] if choices[3] else ''
numSamples = choices[4] if choices[4] else '1'
analysisDef = 'dummy [combineMethod=%s] %s [numSamples=%s] -> ConvertToNonOverlappingCategorySegmentsPythonStat' % \
(combineMethod, '[category=%s]' % category if category != '' else '', numSamples) #'Python'
for regSpec in GenomeInfo.getChrList(genome):
res = GalaxyInterface.runManual([trackName], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
from gold.origdata.TrackGenomeElementSource import TrackViewGenomeElementSource
from gold.origdata.BedComposer import CategoryBedComposer
for resDict in res.values():
tvGeSource = TrackViewGenomeElementSource(genome, resDict['Result'], trackName)
CategoryBedComposer(tvGeSource).composeToFile(outFn)
def execute(choices, galaxyFn=None, username=''):
genome = choices.targetTrackGenome
queryTrackName = choices.targetTrack.split(':')
from quick.multitrack.MultiTrackCommon import getGSuiteDataFromGalaxyTN
trackTitles, refTrackNameList, genome = getGSuiteDataFromGalaxyTN(choices.refTrackCollection)
regSpec = 'track'
binSpec = choices.targetTrack
#future iter1
#regSpec, binSpec = UserBinSelector.getRegsAndBinsSpec(choices)
results = []
for refTrack in refTrackNameList:
analysisDef = '-> ProportionCountStat'
res = GalaxyInterface.runManual([refTrack], analysisDef, regSpec, binSpec, genome, username=username, galaxyFn=galaxyFn) # res is of class Results
segCoverageProp = [res[seg]['Result'] for seg in res.getAllRegionKeys()]#to heatmap
results.append(segCoverageProp)
geSource = FullTrackGenomeElementSource(genome, queryTrackName)
modGeSource = MyElementModifier(geSource, results, genome)
composer = BedComposer(modGeSource, extraTrackLineAttributes={'itemRgb': '"On"'})
composer.composeToFile(galaxyFn)
def run(self):
regSpec, binSpec = 'file', self._referenceTrackFn
trackName1 = self._queryTrackName
trackName2 = None
from gold.description.TrackInfo import TrackInfo
formatName = TrackInfo(self._genome, trackName1).trackFormatName
formatConv = ''
if 'segments' in formatName:
formatConv = '[tf1:=SegmentToStartPointFormatConverter:]'
analysisDef = formatConv + '-> CountPointStat'
print '<div class="debug">'
#trackName1, trackName2, analysisDef = GalaxyInterface._cleanUpAnalysisDef(trackName1, trackName2, analysisDef)
#trackName1, trackName2 = GalaxyInterface._cleanUpTracks([trackName1, trackName2], genome, realPreProc=True)
#
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(trackName1, trackName2, analysisDef, regSpec, binSpec, self._genome)
#res = AnalysisDefJob(analysisDef, trackName1, trackName2, userBinSource, **fullRunArgs).run()
res = GalaxyInterface.runManual([trackName1, trackName2], analysisDef, regSpec, binSpec, self._genome, printResults=False, printHtmlWarningMsgs=False)
#print res
print '</div>'
resDictKeys = res.getResDictKeys()
assert len(resDictKeys)==1, resDictKeys
resDictKey = resDictKeys[0]
targetBins = [bin for bin in res.keys() if res[bin][resDictKey]>0]
self._result = res
self._intersectedReferenceBins = targetBins
def execute(cls, choices, galaxyFn=None, username=''):
from gold.origdata.TrackGenomeElementSource import TrackViewListGenomeElementSource
from gold.origdata.GtrackComposer import StdGtrackComposer
genome = choices[0]
if choices[1] == 'track':
trackName = choices[2].split(':')
else:
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, choices[2].split(':'))
outFn = galaxyFn
if choices[4] == 'Write to Standardised file':
outFn = createOrigPath(genome, choices[-1].split(':'),
'collapsed_result.bedgraph')
ensurePathExists(outFn[:outFn.rfind('/') + 1])
threshold = choices[3]
analysisDef = 'dummy [threshold=%s] -> ForEachSegmentDistToNearestInSameTrackStat' % threshold #'Python'
res = GalaxyInterface.runManual([trackName], analysisDef, '*', '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
tvGeSource = TrackViewListGenomeElementSource(
genome, [x['Result'] for x in res.values()], trackName)
StdGtrackComposer(tvGeSource).composeToFile(outFn)
def runAllAnalysises(self, genome, tn, regSpec, binSpec):
for analysisKey, analysisVals in self.analysisDict.items():
trackName, analysisDef, resKey = analysisVals
trackNames = [tn] if trackName is None else [tn, trackName]
try:
self.resultDict[analysisKey] = (resKey, GalaxyInterface.runManual(trackNames, analysisDef, regSpec, binSpec, genome, galaxyFn=None, printResults=False, printProgress=False))
except:
pass
def analyseTrackData(genome, baseTn, username, galaxyFn):
from quick.application.GalaxyInterface import GalaxyInterface
print GalaxyInterface.getSubTrackNames(genome,
baseTn,
deep=False,
username=username)
allTns = [
baseTn + [leafTn[0]]
for leafTn in GalaxyInterface.getSubTrackNames(
genome, baseTn, deep=False, username=username)[0]
if leafTn != None
]
mainTns = allTns[:2]
extraTracks = '&'.join(['^'.join(tn) for tn in allTns[2:]])
analysisDef = '[extraTracks=%s] -> ThreeWayCoverageDepthStat' % extraTracks
print allTns, mainTns
GalaxyInterface.runManual(mainTns, analysisDef, 'chr1:1-150m', '*', genome, galaxyFn=galaxyFn, username=username, \
printResults=True, printProgress=True, printHtmlWarningMsgs=True, applyBoundaryFilter=False, printRunDescription=True)
def build_feature_vector(genome, ctrack, feature, regSpec, binSpec):
'''
this function create a feature vector for ctrack
feature specifies how the vector is constructed
'''
#print 'TEMP1:', ctrack, '<br>', LocalResultsAsFeaturesCatalog.getValidAnalyses(genome, ctrack, []),'<br>'
#validFeature = LocalResultsAsFeaturesCatalog.getValidAnalyses(genome, ctrack, [])[feature]
validFeature = LocalResultsAsFeaturesCatalog.getAllFeatures()[feature]
analysisDef = validFeature[0]
#regSpec = self.params.get("region")
#binSpec = self.params.get("binsize")
#userBinSource = GalaxyInterface._getUserBinSource(regSpec,binSpec,genome)
#result = GalaxyInterface.runManual([ctrack], analysisDef, regSpec, binSpec, genome, printResults=False, printProgress=False)
#result = AnalysisDefJob(analysisDef, ctrack, [], userBinSource).run()
if validFeature[1] == 'Microbins':
microBinConfig = 'extraDummy [microBin=%i] ' % int(
binSpec.replace('k', '000').replace('m', '000000'))
analysisDef = microBinConfig + analysisDef
result = GalaxyInterface.runManual([ctrack],
analysisDef,
regSpec,
'*',
genome,
printResults=False,
printProgress=False)
res = result.getGlobalResult()['Result']
return res
else:
result = GalaxyInterface.runManual([ctrack],
analysisDef,
regSpec,
binSpec,
genome,
printResults=False,
printProgress=False)
return [
result[localKey][validFeature[1]]
for localKey in sorted(result.keys())
]
def execute(cls, choices, galaxyFn=None, username=''):
'''Is called when execute-button is pushed by web-user.
Should print output as HTML to standard out, which will be directed to a results page in Galaxy history.
If getOutputFormat is anything else than HTML, the output should be written to the file with path galaxyFn.
If needed, StaticFile can be used to get a path where additional files can be put (e.g. generated image files).
choices is a list of selections made by web-user in each options box.
'''
genome = choices.genome
trackChoice = choices.history if choices.source == 'history' else choices.track
trackName = trackChoice.split(':')
galaxyOutTrackName = 'galaxy:hbfunction:%s:Create function track of distance to nearest segment' % galaxyFn
outTrackName = ExternalTrackManager.getStdTrackNameFromGalaxyTN(galaxyOutTrackName.split(':'))
if choices.transform == 'no transformation':
valTransformation = 'None'
elif choices.transform =='logarithmic (log10(x))':
valTransformation = 'log10'
elif choices.transform == 'fifth square root (x**0.2)':
valTransformation = 'power0.2'
else:
raise ShouldNotOccurError
analysisDef ='[dataStat=MakeDistanceToNearestSegmentStat] [valTransformation=%s][outTrackName=' % valTransformation \
+ '^'.join(outTrackName) + '] -> CreateFunctionTrackStat'
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(trackName, None, analysisDef, '*', '*', genome)
#
#for el in userBinSource:
# print el.chr, el.start, el.end
from quick.application.GalaxyInterface import GalaxyInterface
print GalaxyInterface.getHbFunctionOutputBegin(galaxyFn, withDebug=False)
GalaxyInterface.runManual([trackName], analysisDef, '*', '*', genome, username=username, printResults=False, printHtmlWarningMsgs=False)
#job = AnalysisDefJob(analysisDef, trackName, None, userBinSource).run()
print GalaxyInterface.getHbFunctionOutputEnd('A custom track has been created by finding the bp-distance to the nearest segment', withDebug=False)
def execute(cls, choices, galaxyFn=None, username=''):
'''Is called when execute-button is pushed by web-user.
Should print output as HTML to standard out, which will be directed to a results page in Galaxy history.
If getOutputFormat is anything else than HTML, the output should be written to the file with path galaxyFn.
If needed, StaticFile can be used to get a path where additional files can be put (e.g. generated image files).
choices is a list of selections made by web-user in each options box.
'''
#from config.Config import DebugConfig
#DebugConfig.PASS_ON_VALIDSTAT_EXCEPTIONS = True
#DebugConfig.PASS_ON_COMPUTE_EXCEPTIONS = True
##DebugConfig.VERBOSE = True
#
from quick.webtools.restricted.DebugAnalysisTool import DebugAnalysisTool
DebugAnalysisTool.setupDebugModeAndLogging(verbose=False) #set from selection box..
analysisDef = choices[2].replace(os.linesep,'')
#from gold.application.StatRunner import AnalysisDefJob
print 'Preparing arguments'
genome = choices[3]
prefixTN1 = cls.STD_PREFIX_TN if choices[4] == 'yes' else []
tn1 = prefixTN1 + choices[5].split(':')
prefixTN2 = cls.STD_PREFIX_TN if choices[6] == 'yes' else []
tn2 = prefixTN2 + choices[7].split(':')
if tn2[-1] == cls.NO_TRACK_SHORTNAME:
tn2 = None
#from gold.track.GenomeRegion import GenomeRegion
#region = GenomeRegion(genome, 'chr1',1000,2000)
#region2 = GenomeRegion(genome, 'chr1',5000,6000)
#kwArgs = {}
regVal = choices[8]
binSpecVal = choices[9]
#ubSource = UserBinSource(regVal, binSpecVal, genome=genome)
from quick.application.GalaxyInterface import GalaxyInterface
print tn1, tn2
GalaxyInterface.runManual([tn1, tn2], analysisDef, regVal, binSpecVal, genome, galaxyFn, username=username)
def _runCategoryPointCount(cls, genome, regSpec, binSpec, trackName1, applyBoundaryFilter=True):
analysisDef = 'Category point count: Number of elements each category of track1 (with overlaps)'+\
'[tf1:=SegmentToStartPointFormatConverter:]'+\
'-> FreqByCatStat'
print '<div class="debug">'
res = GalaxyInterface.runManual([trackName1], analysisDef, regSpec, binSpec, genome, \
printResults=False, printHtmlWarningMsgs=False, applyBoundaryFilter=applyBoundaryFilter)
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(trackName1, None, analysisDef, regSpec, binSpec, genome)
#if applyBoundaryFilter:
# userBinSource = RegionBoundaryFilter(userBinSource, GlobalBinSource(genome))
#res = AnalysisDefJob(analysisDef, trackName1, None, userBinSource, **fullRunArgs).run()
print res
print '</div>'
return res
def execute(cls, choices, galaxyFn=None, username=''):
start = time.time()
genome = choices[0]
trackName = choices[1].split(':')
#outFn = open(NONSTANDARD_DATA_PATH+'/hg19/Private/Sigven/resultat.bed','w')
analysisDef = '-> ConvertToNonOverlappingCategorySegmentsPythonStat' #'Python'
for regSpec in GenomeInfo.getChrList(genome):
res = GalaxyInterface.runManual([trackName], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
from gold.origdata.TrackGenomeElementSource import TrackViewGenomeElementSource
from gold.origdata.BedComposer import CategoryBedComposer
for resDict in res.values():
tvGeSource = TrackViewGenomeElementSource(genome, resDict['Result'], trackName)
CategoryBedComposer(tvGeSource).composeToFile(outFn)
def PrintResultToHistItem(cls, outFn, geSource, preProcTN1, genome, username):
analysisDef = 'dummy ->SplitToCorrespondingSubBinsBasedOnBreakpointsStat'
#outFile = open(outFn,'w')
with open(outFn,'w') as outFile:
print>>outFile, '##track type: valued segments\n###seqid\tstart\tend\tvalue\tstrand'
for ge in geSource:
regSpec = '%s:%i-%i' %(ge.chr, ge.start+1, ge.end) #+1 since assumed 1-indexed, end-inclusive
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(preProcTN1, None, analysisDef, regSpec, '*', genome)
#res = AnalysisDefJob(analysisDef, preProcTN1, None, userBinSource, **fullRunArgs).run(printProgress=False)
#trackName
res = GalaxyInterface.runManual([preProcTN1], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printProgress=False, printHtmlWarningMsgs=False)
for region,resDict in res.items():
print>>outFile, resDict['Result']
def execute(cls, choices, galaxyFn=None, username=''):
start = time.time()
genome = choices[0]
trackName = choices[1].split(':')
#outFn = open(NONSTANDARD_DATA_PATH+'/hg19/Private/Sigven/resultat.bed','w')
analysisDef = '-> ConvertToNonOverlappingCategorySegmentsPythonStat' #'Python'
for regSpec in GenomeInfo.getChrList(genome):
res = GalaxyInterface.runManual([trackName], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
from gold.origdata.TrackGenomeElementSource import TrackViewGenomeElementSource
from gold.origdata.BedComposer import CategoryBedComposer
for resDict in res.values():
tvGeSource = TrackViewGenomeElementSource(
genome, resDict['Result'], trackName)
CategoryBedComposer(tvGeSource).composeToFile(outFn)
def PrintResultToHistItem(cls, outFn, geSource, preProcTN1, genome,
username):
analysisDef = 'dummy ->SplitToCorrespondingSubBinsBasedOnBreakpointsStat'
#outFile = open(outFn,'w')
with open(outFn, 'w') as outFile:
print >> outFile, '##track type: valued segments\n###seqid\tstart\tend\tvalue\tstrand'
for ge in geSource:
regSpec = '%s:%i-%i' % (
ge.chr, ge.start + 1, ge.end
) #+1 since assumed 1-indexed, end-inclusive
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(preProcTN1, None, analysisDef, regSpec, '*', genome)
#res = AnalysisDefJob(analysisDef, preProcTN1, None, userBinSource, **fullRunArgs).run(printProgress=False)
#trackName
res = GalaxyInterface.runManual([preProcTN1], analysisDef, regSpec, '*', genome, username=username, \
printResults=False, printProgress=False, printHtmlWarningMsgs=False)
for region, resDict in res.items():
print >> outFile, resDict['Result']
def _runCategoryPointCount(cls,
genome,
regSpec,
binSpec,
trackName1,
applyBoundaryFilter=True):
analysisDef = 'Category point count: Number of elements each category of track1 (with overlaps)'+\
'[tf1:=SegmentToStartPointFormatConverter:]'+\
'-> FreqByCatStat'
print '<div class="debug">'
res = GalaxyInterface.runManual([trackName1], analysisDef, regSpec, binSpec, genome, \
printResults=False, printHtmlWarningMsgs=False, applyBoundaryFilter=applyBoundaryFilter)
#userBinSource, fullRunArgs = GalaxyInterface._prepareRun(trackName1, None, analysisDef, regSpec, binSpec, genome)
#if applyBoundaryFilter:
# userBinSource = GERegionBoundaryFilter(userBinSource, GlobalBinSource(genome))
#res = AnalysisDefJob(analysisDef, trackName1, None, userBinSource, **fullRunArgs).run()
print res
print '</div>'
return res
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
analysisDef = 'dummy -> ProportionCountStat'
genome = choices[0]
tn1 = choices[2].split(':')
binSpec = choices[4]
regSpec = 'track'
if choices[3] == 'from history':
regSpec = ExternalTrackManager.extractFileSuffixFromGalaxyTN(
choices[4].split(':'))
binSpec = ExternalTrackManager.extractFnFromGalaxyTN(
choices[4].split(':'))
numBins = open(binSpec).read().count('\n')
if numBins > 330000:
gold.application.StatRunner.MAX_NUM_USER_BINS = numBins
percent = float(choices[5]) if float(
choices[5]) <= 1.0 else float(choices[5]) / 100.0
GalaxyInterface.ALLOW_OVERLAPPING_USER_BINS = True
resultDict = GalaxyInterface.runManual([tn1],
analysisDef,
regSpec,
binSpec,
genome,
galaxyFn,
printResults=False,
printProgress=True)
overlapRegions = [
k for k, v in resultDict.items() if v['Result'] >= percent
]
with open(galaxyFn, 'w') as utfil:
for i in overlapRegions:
print >> utfil, '\t'.join([i.chr, str(i.start), str(i.end)])
def execute(cls, choices, galaxyFn=None, username=''):
from gold.origdata.TrackGenomeElementSource import TrackViewListGenomeElementSource
from gold.origdata.GtrackComposer import StdGtrackComposer
genome = choices[0]
if choices[1] == 'Track':
trackName = choices[2].split(':')
else:
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices[2].split(':'))
outFn = galaxyFn
if choices[4] == 'Write to Standardised file':
outFn = createOrigPath(genome, choices[-1].split(':'), 'collapsed_result.bedgraph')
ensurePathExists(outFn[:outFn.rfind('/')+1])
threshold = choices[3]
analysisDef = 'dummy [threshold=%s] -> ForEachSegmentDistToNearestInSameTrackStat' % threshold #'Python'
res = GalaxyInterface.runManual([trackName], analysisDef, '*', '*', genome, username=username, \
printResults=False, printHtmlWarningMsgs=False)
tvGeSource = TrackViewListGenomeElementSource(genome, [x['Result'] for x in res.values()], trackName)
StdGtrackComposer(tvGeSource).composeToFile(outFn)
def execute(cls, choices, galaxyFn=None, username=''):
from quick.application.ProcTrackOptions import ProcTrackOptions
from quick.application.GalaxyInterface import GalaxyInterface
import gold.application.StatRunner
analysisDef = 'dummy -> PercentageChangeStat'
genome = choices[0]
binSpec = '*'
regSpec = 'Days_1900_2036:36890-41424'
tnRoot = 'Company stocks:Historical prices:OSE:'
stockList = [k for k, v in choices[1].items() if v]
numStocks = 0
totalPercent = 0.0
for stock in stockList:
tn = tnRoot + stock
if ProcTrackOptions.isValidTrack(genome,
tn.split(':'),
fullAccess=True):
resultDict = GalaxyInterface.runManual(
[tn.split(':'), choices[2].split(':')],
analysisDef,
regSpec,
binSpec,
'days',
galaxyFn,
printResults=False,
printProgress=False)
for k, v in resultDict.items():
print 'increase (from jan. 2001 - jun. 2013) for ', stock, ': ', v
res = v['Result']
if res == 0.0 or res > 10000:
continue
totalPercent += v['Result']
numStocks += 1
else:
print 'this is not a valid track', tn
print 'Average increase (from jan. 2001 - jun. 2013): ', totalPercent / numStocks, ' (number of stocks =', numStocks, ')'
def execute(choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
from gold.application.LogSetup import setupDebugModeAndLogging
setupDebugModeAndLogging()
targetTrackNames, targetTrackCollection, targetTrackGenome = getGSuiteDataFromGalaxyTN(
choices.gSuiteFirst)
targetTracksDict = OrderedDict(
zip(targetTrackNames, targetTrackCollection))
refTrackNames, refTrackCollection, refTrackCollectionGenome = getGSuiteDataFromGalaxyTN(
choices.gSuiteSecond)
assert targetTrackGenome == refTrackCollectionGenome, 'Reference genome must be the same one in both GSuite files.'
refTracksDict = OrderedDict(zip(refTrackNames, refTrackCollection))
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisDef = 'dummy -> RawOverlapStat'
results = OrderedDict()
for targetTrackName, targetTrack in targetTracksDict.iteritems():
for refTrackName, refTrack in refTracksDict.iteritems():
result = GalaxyInterface.runManual([targetTrack, refTrack],
analysisDef,
regSpec,
binSpec,
targetTrackGenome,
galaxyFn,
printRunDescription=False,
printResults=False)
if targetTrackName not in results:
results[targetTrackName] = OrderedDict()
results[targetTrackName][
refTrackName] = result.getGlobalResult()
targetTrackTitles = results.keys()
stat = choices.statistic
statIndex = STAT_LIST_INDEX[stat]
title = stat + ' analysis of track collections'
processedResults = []
headerColumn = []
for targetTrackName in targetTrackTitles:
resultRowDict = processRawResults(results[targetTrackName])
resultColumn = []
headerColumn = []
for refTrackName, statList in resultRowDict.iteritems():
resultColumn.append(statList[statIndex])
headerColumn.append(refTrackName)
processedResults.append(resultColumn)
transposedProcessedResults = [list(x) for x in zip(*processedResults)]
tableHeader = ['Track names'] + targetTrackTitles
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.header(title)
htmlCore.divBegin('resultsDiv')
htmlCore.tableHeader(tableHeader,
sortable=True,
tableId='resultsTable')
for i, row in enumerate(transposedProcessedResults):
line = [headerColumn[i]] + row
htmlCore.tableLine(line)
htmlCore.tableFooter()
htmlCore.divEnd()
addColumnPlotToHtmlCore(htmlCore,
targetTrackNames,
refTrackNames,
stat,
title + ' plot',
processedResults,
xAxisRotation=315)
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
def execute(cls, choices, galaxyFn=None, username=''):
cls._setDebugModeIfSelected(choices)
targetGSuite = getGSuiteFromGalaxyTN(choices.gSuiteFirst)
refGSuite = getGSuiteFromGalaxyTN(choices.gSuiteSecond)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisDef = 'dummy -> RawOverlapStat'
# analysisDef = 'dummy [withOverlaps=yes] -> RawOverlapAllowSingleTrackOverlapsStat'
results = OrderedDict()
for targetTrack in targetGSuite.allTracks():
targetTrackName = targetTrack.title
for refTrack in refGSuite.allTracks():
refTrackName = refTrack.title
if targetTrack.trackName == refTrack.trackName:
# print targetTrack.title
# print targetTrack.trackName
result = DetermineSuiteTracksCoincidingWithAnotherSuite.handleSameTrack(
targetTrack.trackName, regSpec, binSpec,
targetGSuite.genome, galaxyFn)
else:
result = GalaxyInterface.runManual(
[targetTrack.trackName, refTrack.trackName],
analysisDef,
regSpec,
binSpec,
targetGSuite.genome,
galaxyFn,
printRunDescription=False,
printResults=False,
printProgress=False).getGlobalResult()
if targetTrackName not in results:
results[targetTrackName] = OrderedDict()
results[targetTrackName][refTrackName] = result
stat = STAT_OVERLAP_COUNT_BPS
statIndex = STAT_LIST_INDEX[stat]
title = ''
processedResults = []
headerColumn = []
for targetTrackName in targetGSuite.allTrackTitles():
resultRowDict = processRawResults(results[targetTrackName])
resultColumn = []
headerColumn = []
for refTrackName, statList in resultRowDict.iteritems():
resultColumn.append(statList[statIndex])
headerColumn.append(refTrackName)
processedResults.append(resultColumn)
outputTable = {}
for elN in range(0, len(headerColumn)):
outputTable[elN] = {}
outputTable[elN]['id'] = headerColumn[elN]
transposedProcessedResults = [list(x) for x in zip(*processedResults)]
# second question sumSecondgSuite
# first question numSecondgSuite
# fifth question numSecondgSuitePercentage
for i in range(0, len(transposedProcessedResults)):
outputTable[i]['sumSecondgSuite'] = sum(
transposedProcessedResults[i])
if not 'numSecondgSuite' in outputTable[i]:
outputTable[i]['numSecondgSuite'] = 0
for j in range(0, len(transposedProcessedResults[i])):
if transposedProcessedResults[i][j] >= 1:
outputTable[i]['numSecondgSuite'] += 1
else:
outputTable[i]['numSecondgSuite'] += 0
outputTable[i]['numSecondgSuitePercentage'] = float(
outputTable[i]['numSecondgSuite']) / float(
targetGSuite.numTracks()) * 100
from gold.statistic.CountSegmentStat import CountSegmentStat
from gold.statistic.CountPointStat import CountPointStat
from gold.description.TrackInfo import TrackInfo
from gold.statistic.CountStat import CountStat
# third question numPairBpSecondgSuite
# fourth question numFreqBpSecondgSuite
i = 0
for refTrack in refGSuite.allTracks():
formatName = TrackInfo(refTrack.genome,
refTrack.trackName).trackFormatName
analysisDef = CountStat
analysisBins = GalaxyInterface._getUserBinSource(
regSpec, binSpec, refTrack.genome)
results = doAnalysis(AnalysisSpec(analysisDef), analysisBins,
[PlainTrack(refTrack.trackName)])
resultDict = results.getGlobalResult()
if len(resultDict) == 0:
outputTable[i]['numPairBpSecondgSuite'] = None
outputTable[i]['numFreqBpSecondgSuite'] = None
outputTable[i]['numFreqUniqueBpSecondgSuite'] = None
else:
outputTable[i]['numPairBpSecondgSuite'] = resultDict['Result']
if outputTable[i]['numPairBpSecondgSuite'] != 0:
outputTable[i]['numFreqBpSecondgSuite'] = float(
outputTable[i]['sumSecondgSuite']) / float(
outputTable[i]['numPairBpSecondgSuite'])
else:
outputTable[i]['numFreqBpSecondgSuite'] = None
if outputTable[i]['sumSecondgSuite'] != 0:
outputTable[i]['numFreqUniqueBpSecondgSuite'] = float(
outputTable[i]['numPairBpSecondgSuite']) / float(
outputTable[i]['sumSecondgSuite'])
else:
outputTable[i]['numFreqUniqueBpSecondgSuite'] = None
i += 1
# sortTable
outputTableLine = []
for key, item in outputTable.iteritems():
line = [
item['id'], item['numSecondgSuite'], item['sumSecondgSuite'],
item['numPairBpSecondgSuite'], item['numFreqBpSecondgSuite'],
item['numFreqUniqueBpSecondgSuite'],
item['numSecondgSuitePercentage']
]
outputTableLine.append(line)
import operator
outputTableLineSort = sorted(outputTableLine,
key=operator.itemgetter(1),
reverse=True)
tableHeader = [
'Region ID ', 'Number of cases with at least one event ',
'Total number of events', 'Genome coverage (unique bp)',
'Number of events per unique bp', 'Number of unique bp per event',
'Percentage of cases with at least one event'
]
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.line(
"<b>Identification of genomic elements with high event recurrence</b> "
)
htmlCore.header(title)
htmlCore.divBegin('resultsDiv')
htmlCore.tableHeader(tableHeader,
sortable=True,
tableId='resultsTable')
for line in outputTableLineSort:
htmlCore.tableLine(line)
plotRes = []
plotXAxis = []
for lineInx in range(1, len(outputTableLineSort[0])):
plotResPart = []
plotXAxisPart = []
for lineInxO in range(0, len(outputTableLineSort)):
# if outputTableLineSort[lineInxO][lineInx]!=0 and
# if outputTableLineSort[lineInxO][lineInx]!=None:
plotResPart.append(outputTableLineSort[lineInxO][lineInx])
plotXAxisPart.append(outputTableLineSort[lineInxO][0])
plotRes.append(plotResPart)
plotXAxis.append(plotXAxisPart)
htmlCore.tableFooter()
htmlCore.divEnd()
htmlCore.divBegin('plot', style='padding-top:20px;margin-top:20px;')
vg = visualizationGraphs()
res = vg.drawColumnCharts(
plotRes,
titleText=tableHeader[1:],
categories=plotXAxis,
height=500,
xAxisRotation=270,
xAxisTitle='Ragion ID',
yAxisTitle='Number of cases with at least one event',
marginTop=30,
addTable=True,
sortableAccordingToTable=True,
legend=False)
htmlCore.line(res)
htmlCore.divEnd()
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
def execute(choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
# from gold.application.LogSetup import setupDebugModeAndLogging
#setupDebugModeAndLogging()
# targetTrackNames, targetTrackCollection, targetTrackGenome = getGSuiteDataFromGalaxyTN(choices.gSuiteFirst)
# targetTracksDict = OrderedDict(zip(targetTrackNames, targetTrackCollection))
# refTrackNames, refTrackCollection, refTrackCollectionGenome = getGSuiteDataFromGalaxyTN(choices.gSuiteSecond)
# refTracksDict = OrderedDict(zip(refTrackNames, refTrackCollection))
#
targetGSuite = getGSuiteFromGalaxyTN(choices.gSuiteFirst)
refGSuite = getGSuiteFromGalaxyTN(choices.gSuiteSecond)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
if choices.intraOverlap == TrackCollectionsAnalysis.MERGE_INTRA_OVERLAPS:
analysisDef = 'dummy -> RawOverlapStat'
else:
analysisDef = 'dummy [withOverlaps=yes] -> RawOverlapAllowSingleTrackOverlapsStat'
results = OrderedDict()
# for targetTrackName, targetTrack in targetTracksDict.iteritems():
# for refTrackName, refTrack in refTracksDict.iteritems():
for targetTrack in targetGSuite.allTracks():
targetTrackName = targetTrack.title
for refTrack in refGSuite.allTracks():
refTrackName = refTrack.title
if targetTrack.trackName == refTrack.trackName:
result = TrackCollectionsAnalysis.handleSameTrack(targetTrack.trackName, regSpec, binSpec,
choices.genome, galaxyFn)
else:
result = GalaxyInterface.runManual([targetTrack.trackName, refTrack.trackName],
analysisDef, regSpec, binSpec,
choices.genome, galaxyFn,
printRunDescription=False,
printResults=False).getGlobalResult()
if targetTrackName not in results :
results[targetTrackName] = OrderedDict()
results[targetTrackName][refTrackName] = result
stat = choices.statistic
statIndex = STAT_LIST_INDEX[stat]
title = 'Screening track collections (' + stat + ')'
processedResults = []
headerColumn = []
for targetTrackName in targetGSuite.allTrackTitles():
resultRowDict = processRawResults(results[targetTrackName])
resultColumn = []
headerColumn = []
for refTrackName, statList in resultRowDict.iteritems():
resultColumn.append(statList[statIndex])
headerColumn.append(refTrackName)
processedResults.append(resultColumn)
transposedProcessedResults = [list(x) for x in zip(*processedResults)]
tableHeader = ['Track names'] + targetGSuite.allTrackTitles()
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.header(title)
htmlCore.divBegin('resultsDiv')
htmlCore.tableHeader(tableHeader, sortable=True, tableId='resultsTable')
for i, row in enumerate(transposedProcessedResults):
line = [headerColumn[i]] + [strWithStdFormatting(x) for x in row]
htmlCore.tableLine(line)
htmlCore.tableFooter()
htmlCore.divEnd()
# #hicharts can't handle strings that contain ' or " as input for series names
targetTrackNames = [x.replace('\'', '').replace('"','') for x in targetGSuite.allTrackTitles()]
refTrackNames = [x.replace('\'', '').replace('"','') for x in refGSuite.allTrackTitles()]
#
# '''
# addColumnPlotToHtmlCore(htmlCore, targetTrackNames, refTrackNames,
# stat, title + ' plot',
# processedResults, xAxisRotation = -45, height=800)
# '''
# '''
# addPlotToHtmlCore(htmlCore, targetTrackNames, refTrackNames,
# stat, title + ' plot',
# processedResults, xAxisRotation = -45, height=400)
# '''
#
from quick.webtools.restricted.visualization.visualizationGraphs import visualizationGraphs
vg = visualizationGraphs()
result = vg.drawColumnChart(processedResults,
height=600,
yAxisTitle=stat,
categories=refTrackNames,
xAxisRotation=90,
seriesName=targetTrackNames,
shared=False,
titleText=title + ' plot',
overMouseAxisX=True,
overMouseLabelX = ' + this.value.substring(0, 10) +')
htmlCore.line(result)
#htmlCore.line(vg.visualizeResults(result, htmlCore))
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
def makeAnalysis(cls, gSuite1, gSuite2, columnsForStat, galaxyFn, choices):
genomeType = 'single'
#if True then none of column in both gsuite are equal then for example, for gSUite 2 dim and 3 dim we have 5 dim
# in the other case (when some columns are equal we have 4 dim)
checkWhichDimension = True
for el in columnsForStat:
if el == 0:
checkWhichDimension=False
break
if choices.type == 'basic':
stat = choices.statistic
statIndex = CountStatisticForCombinationsFromTwoLevelSuites.STAT_LIST_INDEX
statIndex = statIndex.index(stat)
else:
stat= '0'
statIndex = 0
if choices.intraOverlap == CountStatisticForCombinationsFromTwoLevelSuites.MERGE_INTRA_OVERLAPS:
analysisDef = 'dummy -> RawOverlapStat'
else:
analysisDef = 'dummy [withOverlaps=yes] -> RawOverlapAllowSingleTrackOverlapsStat'
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
listResults = []
for el1 in gSuite1:
for el2 in gSuite2:
if genomeType == 'single':
listPartResults=[]
if (checkWhichDimension == True and el1[columnsForStat[0]] == el2[columnsForStat[1]]) or (checkWhichDimension == False):
gSuite1Path = el1[0]#path for track1
gSuite2Path = el2[0]#path for track2
#print [gSuite1Path, gSuite2Path]
result = GalaxyInterface.runManual([gSuite1Path, gSuite2Path],
analysisDef, regSpec, binSpec, choices.genome, galaxyFn,
printRunDescription=False,
printResults=False)
#print str(result) + '<br \>'
#print str(processResult(result.getGlobalResult())) + '<br \>'
resVal = processResult(result.getGlobalResult())[statIndex]
else:
resVal = 0
if checkWhichDimension == False:
listPartResults = el1[1:] + el2[1:] + [resVal]
else:
shorterEl2=[]
for eN in range(0, len(el2[1:])):
if el2[1:][eN] != el2[columnsForStat[1]]:
shorterEl2.append(el2[1:][eN])
#listPartResults = el1[1:] + shorterEl2 + [resVal]
listPartResults = el1[1:] + el2[1:] + [resVal]
listResults.append(listPartResults)
if genomeType == 'multi':
#count only values with the same genome
pass
return listResults
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
regSpec = '*'
binSpec = '*'
analysisDef = 'dummy -> PropPointCountsAllowOverlapsVsSegsStat'
genome = choices[0]
tn1 = choices[1].split(':')
trInfo1 = TrackInfo(genome, tn1)
trackType = trInfo1.trackFormatName
resultDict = defaultdict(dict)
singleTrackDict = defaultdict(list)
geneSourceList = []
singleRuns = 0
trackType, trackCatObj = cls.getTrackTypeAnalysis(genome, trackType)
trackCatObj.runAllAnalysises(genome, tn1, regSpec, binSpec)
#for index, values in enumerate(analysisList):
# geneSource, AnalysisElements = values
#
# if geneSource:
# geneSourceList +=[geneSource]
# else:
# singleRuns +=1
# if not geneSource and singleRuns>1:
# continue
#
# for analysisKey, analysisVals in AnalysisElements.items():
# trackName, analysisDef, resKey = analysisVals
# trackNames = [tn1] if trackName is None else [tn1, trackName]
# if trackName == None:
# if index>0:
#
# gen, an = singleTrackDict[analysisKey]
# resultDict[geneSource][analysisKey] = resultDict[gen][an]
#
# else:
# singleTrackDict[analysisKey]+=[geneSource, analysisKey]
# resultDict[geneSource][analysisKey] = (resKey, GalaxyInterface.runManual(trackNames, analysisDef, regSpec, binSpec, genome, galaxyFn, printResults=False, printProgress=False))
#
# else:
# resultDict[geneSource][analysisKey] = (resKey, GalaxyInterface.runManual(trackNames, analysisDef, regSpec, binSpec, genome, galaxyFn, printResults=False, printProgress=False))
#
core = HtmlCore()
core.begin(extraJavaScriptFns=['tabber.js','https://www.google.com/jsapi'],\
extraCssFns=['tabber.css'])
stack = []
SummaryTextTemplate = '''<div style="background-color:#FFE899;"><h3>Track report:</h3><br/><b>The track consists of %i elements along the genome (%i after merging overlapping elements), and covers %i base pairs (%s percent ) of the genome.
The distribution of track elements along the genome is *visualized below*. It overlaps %s percent with exons,
%s percent with introns, and %s percent with remaining inter-genic regions (according to the Ensembl gene definition, version <E>).
Corresponding numbers for other gene definitions, as well as local results per chromosome, are given in the *tables below*. </b><br></div>'''# %(resultDict[geneSourceList[0]]['CountPointStat'].getGlobalResult()['Result'])
#if singleTrackDict.has_key('CountPointStat'):
if trackCatObj.resultDict.has_key('CountPointStat'):
#globalResDict, localResDict = cls.makeGlobalAndLocalResDicts(resultDict)
globalResDict, localResDict = trackCatObj.makeGlobalAndLocalResDicts()
geneSourceList = globalResDict.keys()
print 'globalResDict', globalResDict
print 'localResDict', localResDict
if len(geneSourceList)==0:
exonPercent, intronsPercent, interGeneticPercent = '0','0','0'
geneSourceList += [None]
else:
sectionsEnsmblRes = globalResDict[geneSourceList[0]]
totalEnsmbl = float(sum(sectionsEnsmblRes))
exonPercent = str(round(sectionsEnsmblRes[0]*100/totalEnsmbl, 2)) if sectionsEnsmblRes>0 else '0'
intronsPercent = str(round(sectionsEnsmblRes[1]*100/totalEnsmbl, 2)) if totalEnsmbl>0 else '0'
interGeneticPercent = str(round(sectionsEnsmblRes[2]*100/totalEnsmbl, 2) ) if totalEnsmbl>0 else '0'
resKey, result = trackCatObj.resultDict.get('CountPointStat')
bpCoverage = numElems = int(result.getGlobalResult()[resKey])
resKey, result = trackCatObj.resultDict.get('numElAllowOverlap')
numUniqueElems = int(result.getGlobalResult()[resKey])
if trackType.lower().find('segment')>=0:
resKey, result = trackCatObj.resultDict.get('bpCoverage')
bpCoverage = int(result.getGlobalResult()[resKey])
genomeBps = sum( GenomeInfo.getStdChrLengthDict(genome).values())
core.paragraph(SummaryTextTemplate % (numUniqueElems, numElems, bpCoverage, str(round(float(bpCoverage)/genomeBps, 2)) , exonPercent, intronsPercent, interGeneticPercent))
core._str += '<div class="tabber">\n'
stack.append('</div>\n')
analysisDef = ' [centerRows=True] [normalizeRows=True] -> RawVisualizationDataStat'
res = GalaxyInterface.runManual([tn1], analysisDef, regSpec, binSpec, genome, username=username, printResults=False, printHtmlWarningMsgs=False)
cls.MakeGlobalResultsDiv(globalResDict, res, core, galaxyFn)
cls.MakeLocalResultsDiv(localResDict, core)
core._str += stack.pop()
if len(cls.pieList)>0:
core.line(cls.makeAllGooglePieChart(cls.pieList))
print core
def execute(choices, galaxyFn=None, username=''):
#targetTrackNames, targetTrackCollection, targetTrackGenome = getGSuiteDataFromGalaxyTN(choices.gSuiteFirst)
gFirst = choices.gSuiteFirst.split(':')
firstGSuite = ScreenTwoTrackCollectionsAgainstEachOther2LevelDepth.returnGSuiteDict3LevelDept(
gFirst)
gSecond = choices.gSuiteSecond.split(':')
secondGSuite = ScreenTwoTrackCollectionsAgainstEachOther2LevelDepth.returnGSuiteDict2LevelDept(
gSecond)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
if choices.intraOverlap == ScreenTwoTrackCollectionsAgainstEachOther2LevelDepth.MERGE_INTRA_OVERLAPS:
analysisDef = 'dummy -> RawOverlapStat'
else:
analysisDef = 'dummy [withOverlaps=yes] -> RawOverlapAllowSingleTrackOverlapsStat'
if choices.type == 'basic':
results = []
for elFG in firstGSuite:
for elSG in secondGSuite:
if elFG['genome'] == elSG['genome']:
targetTrackGenome = elFG['genome']
resultPartList3 = []
for targetTrackDetailFolder1 in elFG[
'dataFolderValue0']:
resultPartList2 = []
for targetTrackDetail in targetTrackDetailFolder1[
'dataFolderValue1']:
resultPartList1 = []
for el in elSG['dataFolderValue0']:
result = GalaxyInterface.runManual(
[
targetTrackDetail['trackPath'],
el['trackPath']
],
analysisDef,
regSpec,
binSpec,
elFG['genome'].split('-')[0],
galaxyFn,
printRunDescription=False,
printResults=False)
resultPartList1.append({
'refTrackName':
el['trackName'].replace(
targetTrackGenome, ''),
'data':
processResult(result.getGlobalResult())
})
resultPartList2.append({
'folderName2':
targetTrackDetail['folderName2'],
'targetTrackName':
targetTrackDetail['trackName'],
'dataFolderValue2':
resultPartList1
})
resultPartList3.append({
'folderName1':
targetTrackDetailFolder1['folderName1'],
'dataFolderValue1':
resultPartList2
})
results.append({
'genome': targetTrackGenome,
'dataFolderValue0': resultPartList3
})
else:
from quick.statistic.NumT2SegsTouchedByT1SegsStat import NumT2SegsTouchedByT1SegsStat
results = []
for elFG in firstGSuite:
for elSG in secondGSuite:
if elFG['genome'] == elSG['genome']:
if choices.statistic == 'Number of touched segments':
analysisSpec = AnalysisSpec(
NumT2SegsTouchedByT1SegsStat)
#analysisBins = UserBinSource('*', '10m', genome=elFG['genome'].split('-')[0])
analysisBins = GlobalBinSource(
elFG['genome'].split('-')[0])
targetTrackGenome = elFG['genome']
resultPartList3 = []
for targetTrackDetailFolder1 in elFG[
'dataFolderValue0']:
resultPartList2 = []
for targetTrackDetail in targetTrackDetailFolder1[
'dataFolderValue1']:
resultPartList1 = []
for el in elSG['dataFolderValue0']:
res = doAnalysis(
analysisSpec, analysisBins, [
PlainTrack(
targetTrackDetail['trackPath']
),
PlainTrack(el['trackPath'])
])
resultDict = res.getGlobalResult()
resultPartList1.append({
'refTrackName':
el['trackName'].replace(
targetTrackGenome, ''),
'data': [resultDict['Result']]
})
resultPartList2.append({
'folderName2':
targetTrackDetail['folderName2'],
'targetTrackName':
targetTrackDetail['trackName'],
'dataFolderValue2':
resultPartList1
})
resultPartList3.append({
'folderName1':
targetTrackDetailFolder1['folderName1'],
'dataFolderValue1':
resultPartList2
})
results.append({
'genome': targetTrackGenome,
'dataFolderValue0': resultPartList3
})
if choices.type == 'basic':
stat = choices.statistic
#statIndex = STAT_LIST_INDEX[stat]
statIndex = ScreenTwoTrackCollectionsAgainstEachOther2LevelDepth.STAT_LIST_INDEX
statIndex = statIndex.index(stat)
else:
stat = '0'
statIndex = 0
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.line("""
<style type="text/css">
.hidden {
display: none;
{
.visible {
display: block;
}
</style>
""")
folderValue0Unique = []
folderValue1Unique = []
folderValue2Unique = []
targetTrackFeatureTitles = []
for dataDetail0 in results:
if dataDetail0['genome'] not in folderValue0Unique:
folderValue0Unique.append(dataDetail0['genome'])
for dataDetail1 in dataDetail0['dataFolderValue0']:
if dataDetail1['folderName1'] not in folderValue1Unique:
folderValue1Unique.append(dataDetail1['folderName1'])
for dataDetail2 in dataDetail1['dataFolderValue1']:
if dataDetail2['folderName2'] not in folderValue2Unique:
folderValue2Unique.append(dataDetail2['folderName2'])
for dataDetail3 in dataDetail2['dataFolderValue2']:
if dataDetail3[
'refTrackName'] not in targetTrackFeatureTitles:
targetTrackFeatureTitles.append(
dataDetail3['refTrackName'])
#print 'folderValue0Unique=' + str(folderValue0Unique)
#print 'folderValue1Unique=' + str(folderValue1Unique)
#print 'folderValue2Unique=' + str(folderValue2Unique)
#print 'targetTrackFeatureTitles=' + str(targetTrackFeatureTitles)
targetTrackNameList = targetTrackFeatureTitles
htmlCore.line('Statistic: ' + stat)
htmlCore.line(
addJS3levelOptionList(folderValue1Unique, folderValue2Unique,
targetTrackFeatureTitles,
targetTrackNameList, folderValue0Unique))
htmlCore.divBegin('results')
#htmlCore.paragraph(preporcessResults(results, folderValue1Unique, folderValue2Unique, targetTrackFeatureTitles, statIndex))
htmlCore.paragraph(
preporcessResults3(results, folderValue1Unique, folderValue2Unique,
targetTrackFeatureTitles, folderValue0Unique,
statIndex))
htmlCore.divEnd()
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
import gold.gsuite.GSuiteComposer as GSuiteComposer
from gold.gsuite.GSuite import GSuite
from gold.gsuite.GSuiteTrack import GSuiteTrack, HbGSuiteTrack
from gold.origdata.TrackGenomeElementSource import TrackViewListGenomeElementSource
from gold.origdata.FileFormatComposer import getComposerClsFromFileSuffix
from quick.multitrack.MultiTrackCommon import getGSuiteFromGalaxyTN
from quick.application.ExternalTrackManager import ExternalTrackManager
from quick.application.GalaxyInterface import GalaxyInterface
from quick.application.UserBinSource import UserBinSource
from quick.extra.TrackExtractor import TrackExtractor
genome = choices.genome
gSuite = getGSuiteFromGalaxyTN(choices.gSuite)
if choices.withOverlaps == cls.NO_OVERLAPS:
if choices.trackSource == cls.FROM_HISTORY_TEXT:
filterTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices.trackHistory)
else:
filterTrackName = choices.track.split(':')
else:
if choices.trackSource == cls.FROM_HISTORY_TEXT:
regSpec = ExternalTrackManager.extractFileSuffixFromGalaxyTN(choices.trackHistory)
binSpec = ExternalTrackManager.extractFnFromGalaxyTN(choices.trackHistory)
else:
regSpec = 'track'
binSpec = choices.track
userBinSource = UserBinSource(regSpec, binSpec, genome)
desc = cls.OUTPUT_GSUITE_DESCRIPTION
emptyFn = cls.extraGalaxyFn \
[getGSuiteHistoryOutputName('nointersect', description=desc, datasetInfo=choices.gSuite)]
primaryFn = cls.extraGalaxyFn \
[getGSuiteHistoryOutputName('primary', description=desc, datasetInfo=choices.gSuite)]
errorFn = cls.extraGalaxyFn \
[getGSuiteHistoryOutputName('nopreprocessed', description=desc, datasetInfo=choices.gSuite)]
preprocessedFn = cls.extraGalaxyFn \
[getGSuiteHistoryOutputName('preprocessed', description=desc, datasetInfo=choices.gSuite)]
hiddenStorageFn = cls.extraGalaxyFn \
[getGSuiteHistoryOutputName('storage', description=desc, datasetInfo=choices.gSuite)]
analysisDef = '-> TrackIntersectionStat'
# analysisDef = '-> TrackIntersectionWithValStat'
numTracks = gSuite.numTracks()
progressViewer = ProgressViewer([(cls.PROGRESS_INTERSECT_MSG, numTracks),
(cls.PROGRESS_PREPROCESS_MSG, numTracks)], galaxyFn)
emptyGSuite = GSuite()
primaryGSuite = GSuite()
for track in gSuite.allTracks():
newSuffix = cls.OUTPUT_TRACKS_SUFFIX
extraFileName = os.path.sep.join(track.trackName)
extraFileName = changeSuffixIfPresent(extraFileName, newSuffix=newSuffix)
title = getTitleWithSuffixReplaced(track.title, newSuffix)
primaryTrackUri = GalaxyGSuiteTrack.generateURI(
galaxyFn=hiddenStorageFn, extraFileName=extraFileName,
suffix=newSuffix if not extraFileName.endswith(newSuffix) else '')
primaryTrack = GSuiteTrack(primaryTrackUri, title=title,
genome=track.genome, attributes=track.attributes)
if choices.withOverlaps == cls.NO_OVERLAPS:
res = GalaxyInterface.runManual([track.trackName, filterTrackName], analysisDef, '*', '*',
genome=genome, galaxyFn=galaxyFn, username=username)
trackViewList = [res[key]['Result'] for key in sorted(res.keys())]
tvGeSource = TrackViewListGenomeElementSource(genome, trackViewList)
composerCls = getComposerClsFromFileSuffix(cls.OUTPUT_TRACKS_SUFFIX)
composerCls(tvGeSource).composeToFile(primaryTrack.path)
else:
TrackExtractor.extractOneTrackManyRegsToOneFile( \
track.trackName, userBinSource, primaryTrack.path, fileFormatName=cls.OUTPUT_TRACKS_SUFFIX, \
globalCoords=True, asOriginal=False, allowOverlaps=True)
# Temporary hack until better solution for empty result tracks have been implemented
from gold.origdata.GenomeElementSource import GenomeElementSource
geSource = GenomeElementSource(primaryTrack.path, genome=genome, suffix=cls.OUTPUT_TRACKS_SUFFIX)
try:
geSource.parseFirstDataLine()
primaryGSuite.addTrack(primaryTrack)
except Exception, e: # Most likely empty file
primaryTrack.comment = e.message
emptyGSuite.addTrack(primaryTrack)
numTracks -= 1
progressViewer.updateProgressObjectElementCount(
cls.PROGRESS_PREPROCESS_MSG, numTracks)
#
progressViewer.update()
def execute(cls, choices, galaxyFn=None, username=''):
path = str(URL_PREFIX)
dataset = choices.dataset
genome = choices.genome
text = choices.newtrack
secondDataset = choices.newdataset
inputFile = open(ExternalTrackManager.extractFnFromGalaxyTN(dataset),
'r')
with inputFile as f:
data = [x for x in f.readlines()]
silenceRWarnings()
binSourceParam = '*'
regSourceParam = '*'
trackNamePrep = cls.preprocessTrack(genome, dataset)
if text == 'No':
figUrl = ''
if (len(data) > 30000):
core = HtmlCore()
core.styleInfoBegin(styleClass='debug')
figImage = GalaxyRunSpecificFile(['VizTrackOnGenome.png'],
galaxyFn)
analysisDef = ' [normalizeRows=%s] [centerRows=%s] -> RawVisualizationDataStat'
res = GalaxyInterface.runManual([trackNamePrep],
analysisDef,
regSourceParam,
binSourceParam,
genome,
username=username,
printResults=False,
printHtmlWarningMsgs=False)
core.styleInfoEnd()
core.line('')
core.tableHeader(None)
rScript = VisualizeTrackPresenceOnGenome.customRExecution(
res, figImage.getDiskPath(ensurePath=True), '')
figUrl = figImage.getURL()
print GalaxyInterface.getHtmlEndForRuns()
binSourceParam = '10m'
regSourceParam = '*'
cls.resultPrintGeneric(genome, binSourceParam, regSourceParam,
figUrl, path, trackNamePrep)
else:
if isinstance(trackNamePrep[0], (list, )):
numTracks = len(trackNamePrep[0])
firstTrack = cls.prepareTracknameForURL(trackNamePrep[0])
trackTitle = json.dumps(trackNamePrep[1])
cls.resultPrintGSuite(genome, binSourceParam,
regSourceParam, figUrl, path,
firstTrack, trackTitle, numTracks)
else:
firstTrack = cls.prepareTracknameForURL(trackNamePrep)
cls.resultPrintGeneric(genome, binSourceParam,
regSourceParam, figUrl, path,
firstTrack)
else:
trackName2 = cls.preprocessTrack(genome, secondDataset)
firstTrack = cls.prepareTracknameForURL(trackNamePrep)
secondTrack = cls.prepareTracknameForURL(trackName2)
cls.resultPrintOverlap(genome, binSourceParam, regSourceParam,
path, firstTrack, secondTrack)
def findOverrepresentedTFsFromGeneSet(genome, tfSource, ensembleGeneIdList,
upFlankSize, downFlankSize,
geneSource, galaxyFn):
#galaxyFn = '/usit/insilico/web/lookalike/galaxy_dist-20090924-dev/database/files/003/dataset_3347.dat'
#print 'overriding galaxyFN!: ', galaxyFn
galaxyId = extractIdFromGalaxyFn(galaxyFn)
uniqueWebPath = GalaxyRunSpecificFile([], galaxyFn).getDiskPath()
assert genome == 'hg18'
tfTrackNameMappings = TfInfo.getTfTrackNameMappings(genome)
tfTrackName = tfTrackNameMappings[tfSource]
#Get gene track
assert geneSource == 'Ensembl'
targetGeneRegsTempFn = uniqueWebPath + os.sep + 'geneRegs.bed'
geneRegsTrackName = GenomeInfo.getStdGeneRegsTn(genome)
geneRegsFn = getOrigFn(genome, geneRegsTrackName, '.category.bed')
GalaxyInterface.getGeneTrackFromGeneList(genome, geneRegsTrackName,
ensembleGeneIdList,
targetGeneRegsTempFn)
assert upFlankSize == downFlankSize == 0 #Should instead extend regions to include flanks
tcGeneRegsTempFn = uniqueWebPath + os.sep + 'tcGeneRegs.targetcontrol.bedgraph'
#Think this will be okay, subtraction not necessary as targets are put first:
controlGeneRegsTempFn = geneRegsFn
#print targetGeneRegsTempFn, controlGeneRegsTempFn, tcGeneRegsTempFn
GalaxyInterface.combineToTargetControl(targetGeneRegsTempFn,
controlGeneRegsTempFn,
tcGeneRegsTempFn)
#tcGeneRegsExternalTN = ['external'] +galaxyId + [tcGeneRegsTempFn]
tcGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(
galaxyId + ['tempTc'])
#tcGeneRegsExternalTN = ['external'] +targetGalaxyId + [tcGeneRegsTempFn]
#tcGeneRegsExternalTN = ['galaxy', externalId, tcGeneRegsTempFn]
targetGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(
galaxyId + ['tempTc', '1'])
controlGeneRegsExternalTN = ExternalTrackManager.createStdTrackName(
galaxyId + ['tempTc', '0'])
#pre-process
print 'Pre-processing file: %s, with trackname: %s ' % (
tcGeneRegsTempFn, tcGeneRegsExternalTN)
ExternalTrackManager.preProcess(tcGeneRegsTempFn, tcGeneRegsExternalTN,
'targetcontrol.bedgraph', genome)
print 'Pre-processing TN: ', targetGeneRegsExternalTN
ExternalTrackManager.preProcess(targetGeneRegsTempFn,
targetGeneRegsExternalTN, 'bed',
genome)
print 'Pre-processing TN: ', controlGeneRegsExternalTN
ExternalTrackManager.preProcess(controlGeneRegsTempFn,
controlGeneRegsExternalTN, 'bed',
genome)
#print tcGeneRegsExternalTN
trackName1, trackName2 = tfTrackName, tcGeneRegsExternalTN
analysisDef = 'Categories differentially located in targets?: Which categories of track1-points fall more inside case than control track2-segments? [rawStatistic:=PointCountInsideSegsStat:]' +\
'[tf1:=SegmentToStartPointFormatConverter:] [tf2:=TrivialFormatConverter:]' +\
'-> DivergentRowsInCategoryMatrixStat'
regSpec, binSpec = '*', '*'
#print 'skipping preproc!!'
#ExternalTrackManager.preProcess(tcGeneRegsExternalTN[-1], tcGeneRegsExternalTN, 'targetcontrol.bedgraph', genome)
#ExternalTrackManager.preProcess(targetGeneRegsTempFn, targetGeneRegsExternalTN, 'bed', genome)
GalaxyInterface.runManual([trackName1, trackName2],
analysisDef,
regSpec,
binSpec,
genome,
printResults=True,
printHtmlWarningMsgs=False)
def execute(cls, choices, galaxyFn=None, username=''):
'''
Is called when execute-button is pushed by web-user. Should print
output as HTML to standard out, which will be directed to a results page
in Galaxy history. If getOutputFormat is anything else than HTML, the
output should be written to the file with path galaxyFn. If needed,
StaticFile can be used to get a path where additional files can be put
(e.g. generated image files). choices is a list of selections made by
web-user in each options box.
'''
genome = choices[0]
regSpec = '__chrs__'
binSpec = '*'
if choices[6] == 'Chromosome arms':
regSpec = '__chrArms__'
elif choices[6] == 'Track from history...':
#put in history bins support here
#print choices[4:]
regSpec = ExternalTrackManager.extractFileSuffixFromGalaxyTN(choices[7].split(':'))
binSpec = ExternalTrackManager.extractFnFromGalaxyTN(choices[7].split(':'))
#print 'regSpec, binSpec,', regSpec, binSpec
lineList, counter, tooManyBins = [], 0, False
for line in open(binSpec):
if line.strip() !='':
if counter == cls.MAX_NUM_ROWS:
tooManyBins = True
break
lineList.append(line)
counter+= 1 if line.strip()[0] !='#' else 0
if tooManyBins:
newHist = GalaxyRunSpecificFile(['newHistFile.%s' % regSpec], galaxyFn)
binSpec = newHist.getDiskPath(ensurePath=True)
open(binSpec, 'w').write(''.join(lineList))
print GalaxyInterface.getHtmlBeginForRuns(galaxyFn)
print GalaxyInterface.getHtmlForToggles(withRunDescription=False)
core = HtmlCore()
core.styleInfoBegin(styleClass='debug')
figImage = GalaxyRunSpecificFile(['VizTrackOnGenome.png'], galaxyFn)
#StaticImage(['VizTrackOnGenome.png'])
analysisDef = ' [normalizeRows=%s] [centerRows=%s] -> RawVisualizationDataStat' % \
(choices[4] == 'Scale to same size', choices[5] == 'Center')
if choices[1] == 'HyperBrowser repository':
trackName = choices[2].split(':')
else:
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices[3].split(':'))
res = GalaxyInterface.runManual([trackName], analysisDef, regSpec, binSpec, genome, username=username, printResults=False, printHtmlWarningMsgs=False)
core.styleInfoEnd()
core.line('')
core.tableHeader(None)
#visPresenter = RawVisualizationPresenter(res, galaxyFn,'')#os.path.split()[0]
#htmlStreng = visPresenter.getReference('Result', fullImage=True)
rScript = cls.customRExecution(res, figImage.getDiskPath(ensurePath=True), '')
figUrl = figImage.getURL()
figLinkText ='<img src="%s" alt="Figure" height="%i" width="800"/>' % (figUrl, 20 *min(cls.MAX_NUM_ROWS, len(res)))
core.tableLine([figImage.getLink(figLinkText)])
rScriptGalaxyFile = GalaxyRunSpecificFile(['RScript.R'], galaxyFn)
with open(rScriptGalaxyFile.getDiskPath(ensurePath=True), 'w') as rScriptFile:
rScriptFile.write(rScript)
core.tableLine([rScriptGalaxyFile.getLink('R script')])
core.tableFooter()
print core
print GalaxyInterface.getHtmlEndForRuns()
def execute(cls, choices, galaxyFn=None, username=''):
genome = choices.genome
from quick.multitrack.MultiTrackCommon import getGSuiteDataFromGalaxyTN
trackTitles, refTrackNameList, genome = getGSuiteDataFromGalaxyTN(choices.gsuite)
queryTrackName = ExternalTrackManager.extractFnFromGalaxyTN(choices.targetTrack)
if choices.isBasic:
suffix = ExternalTrackManager.extractFileSuffixFromGalaxyTN(choices.targetTrack, False)
regSpec = suffix
binSpec = queryTrackName
else:
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
#targetTrack = choices.targetTrack.split(':')
#targetTrackTitle = targetTrack[-1]
#print targetTrackTitle
#
#binSpec = targetTrackTitle
#Phenotype and disease associations:Assorted experiments:Virus integration, HPV specific, Kraus and Schmitz, including 50kb flanks
from gold.gsuite.GSuiteConstants import TITLE_COL
from gold.gsuite.GSuite import GSuite
from proto.hyperbrowser.StaticFile import GalaxyRunSpecificFile
from gold.gsuite.GSuiteEditor import selectColumnsFromGSuite
staticFile=[]
results = []
for refTrack in refTrackNameList:
analysisDef = '-> ProportionCountStat' #ProportionCountStat #CountStat
res = GalaxyInterface.runManual([refTrack], analysisDef, regSpec, binSpec, genome, username=username, galaxyFn=galaxyFn, printRunDescription=False, printResults=False, printProgress=False)
segCoverageProp = [res[seg]['Result'] for seg in res.getAllRegionKeys()]
results.append(segCoverageProp)
regFileNamer = GalaxyRunSpecificFile(refTrack, galaxyFn)
staticFile.append([regFileNamer.getLink('Download bed-file'), regFileNamer.getLoadToHistoryLink('Download bed-file to History')])
refGSuite = getGSuiteFromGalaxyTN(choices.gsuite)
if TITLE_COL == choices.selectColumns:
selected = trackTitles
else:
selected = refGSuite.getAttributeValueList(choices.selectColumns)
yAxisNameOverMouse=[]
metadataAll =[]
for x in range(0, len(selected)):
if selected[x] == None:
yAxisNameOverMouse.append(str(trackTitles[x]) + ' --- ' + 'None')
else:
if TITLE_COL == choices.selectColumns:
yAxisNameOverMouse.append(selected[x].replace('\'', '').replace('"', ''))
else:
metadata = str(selected[x].replace('\'', '').replace('"', ''))
yAxisNameOverMouse.append(str(trackTitles[x]) + ' --- ' + metadata)
metadataAll.append(metadata)
colorListForYAxisNameOverMouse = []
if len(metadataAll) > 0:
import quick.webtools.restricted.visualization.visualizationGraphs as vg
cList = vg.colorList().fullColorList()
uniqueCList = list(set(metadataAll))
for m in metadataAll:
colorListForYAxisNameOverMouse.append(cList[uniqueCList.index(m)])
#startEnd - order in res
startEndInterval = []
startEnd = []
i=0
extraX=[]
rowLabel = []
for ch in res.getAllRegionKeys():
rowLabel.append(str(ch.chr) + ":" + str(ch.start) + "-" + str(ch.end) + str(' (Pos)' if ch.strand else ' (Neg)'))
if not i==0 and not i==len(res.getAllRegionKeys())-1:
start = ch.start
if start-end > 0:
startEnd.append(start-end)
else:
startEnd.append('null')
extraX.append("""{ color: 'orange', width: 5, value: '""" + str(i-0.5) + """' }""")
startEndInterval.append(ch.end - ch.start)
else:
startEndInterval.append(ch.end - ch.start)
end = ch.end
i+=1
extraXAxis='plotLines: [ '
extraXAxis = extraXAxis + ",".join(extraX)
extraXAxis = extraXAxis + """ ], """
#rowLabel = res.getAllRegionKeys()
#rowLabel = [str(x) for x in rowLabel]
import quick.webtools.restricted.visualization.visualizationPlots as vp
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.divBegin(divId='results-page')
htmlCore.divBegin(divClass='results-section')
htmlCore.divBegin('plotDiv')
htmlCore.line(vp.addJSlibs())
htmlCore.line(vp.useThemePlot())
htmlCore.line(vp.addJSlibsExport())
htmlCore.line(vp.axaddJSlibsOverMouseAxisisPopup())
#vp.addGuideline(htmlCore)
htmlCore.line(vp._addGuidelineV1())
htmlCore.line(vp.addJSlibsHeatmap())
from config.Config import DATA_FILES_PATH
from proto.StaticFile import StaticFile, GalaxyRunSpecificFile
#sf = GalaxyRunSpecificFile(['result.txt'], galaxyFn)
#outFile = sf.getDiskPath(ensurePath=True)
htmlCore.divBegin()
writeFile = open(
cls.makeHistElement(galaxyExt='tabular',
title='result'), 'w')
# htmlCore.link('Get all results', sf.getURL())
htmlCore.divEnd()
i = 0
writeFile.write('Track' + '\t' + '\t'.join(rowLabel)+ '\n')
for rList in results:
writeFile.write(str(yAxisNameOverMouse[i]) + '\t' + '\t'.join([str(r) for r in rList]) + '\n')
i+=1
fileOutput = GalaxyRunSpecificFile(['heatmap.png'],
galaxyFn)
ensurePathExists(fileOutput.getDiskPath())
fileOutputPdf = GalaxyRunSpecificFile(['heatmap.pdf'],
galaxyFn)
ensurePathExists(fileOutputPdf.getDiskPath())
cls.generateStaticRPlot(results, colorListForYAxisNameOverMouse, rowLabel, yAxisNameOverMouse,
colorMaps[choices.colorMapSelectList],
fileOutput.getDiskPath(), fileOutputPdf.getDiskPath())
htmlCore.divBegin(divId='heatmap', style="padding: 10px 0 px 10 px 0px;margin: 10px 0 px 10 px 0px")
htmlCore.link('Download heatmap image', fileOutputPdf.getURL())
htmlCore.divEnd()
if len(results) * len(results[1]) >= 10000:
htmlCore.image(fileOutput.getURL())
else:
min = 1000000000
max = -1000000000
for rList in results:
for r in rList:
if min > r:
min = r
if max < r:
max = r
if max-min != 0:
resultNormalised = []
for rList in results:
resultNormalisedPart = []
for r in rList:
resultNormalisedPart.append((r-min)/(max-min))
resultNormalised.append(resultNormalisedPart)
addText = '(normalised to [0, 1])'
else:
resultNormalised = results
addText = ''
hm, heatmapPlotNumber, heatmapPlot = vp.drawHeatMap(
resultNormalised,
colorMaps[choices.colorMapSelectList],
label='this.series.xAxis.categories[this.point.x] + ' + "'<br >'" + ' + yAxisNameOverMouse[this.point.y] + ' + "'<br>Overlap proportion" + str(addText) + ": <b>'" + ' + this.point.value + ' + "'</b>'",
yAxisTitle= 'Reference tracks',
categories=rowLabel,
tickInterval=1,
plotNumber=3,
interaction=True,
otherPlotNumber=1,
titleText='Overlap with reference tracks for each local region',
otherPlotData=[startEnd, startEndInterval],
overMouseAxisX=True,
overMouseAxisY=True,
yAxisNameOverMouse=yAxisNameOverMouse,
overMouseLabelY=" + 'Track: '" + ' + this.value + ' + "' '" + ' + yAxisNameOverMouse[this.value] + ',
overMouseLabelX = ' + this.value.substring(0, 20) +',
extrOp = staticFile
)
htmlCore.line(hm)
htmlCore.line(vp.drawChartInteractionWithHeatmap(
[startEndInterval, startEnd],
tickInterval=1,
type='line',
categories=[rowLabel, rowLabel],
seriesType=['line', 'column'],
minWidth=300,
height=500,
lineWidth=3,
titleText=['Lengths of segments (local regions)','Gaps between consecutive segments'],
label=['<b>Length: </b>{point.y}<br/>', '<b>Gap length: </b>{point.y}<br/>'],
subtitleText=['',''],
yAxisTitle=['Lengths','Gap lengths'],
seriesName=['Lengths','Gap lengths'],
xAxisRotation=90,
legend=False,
extraXAxis=extraXAxis,
heatmapPlot=heatmapPlot,
heatmapPlotNumber=heatmapPlotNumber,
overMouseAxisX=True,
overMouseLabelX = ' + this.value.substring(0, 20) +'
))
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.end()
htmlCore.hideToggle(styleClass='debug')
print htmlCore
