def _createPreprocessedGsuiteTrack(self,
gSuiteTrack,
baseFileName,
doEncodeId,
urlPrefix=None):
from quick.application.ExternalTrackManager import ExternalTrackManager as ETM
from gold.description.TrackInfo import TrackInfo
self.genericVisit(gSuiteTrack)
galaxyTN = ETM.constructGalaxyTnFromSuitedFn(
gSuiteTrack.path, fileEnding=gSuiteTrack.suffix, name=baseFileName)
trackName = ETM.getPreProcessedTrackFromGalaxyTN(
gSuiteTrack.genome,
galaxyTN,
printErrors=False,
printProgress=False,
renameExistingTracksIfNeeded=False,
doEncodeId=doEncodeId,
urlPrefix=urlPrefix)
trackType = TrackInfo(gSuiteTrack.genome,
trackName).trackFormatName.lower()
hbUri = HbGSuiteTrack.generateURI(trackName=trackName)
return GSuiteTrack(hbUri,
title=gSuiteTrack.title,
trackType=trackType,
genome=gSuiteTrack.genome,
attributes=gSuiteTrack.attributes,
comment=gSuiteTrack.comment)
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 _collectTracks(self):
tracks = [self._track, self._track2]
if 'trackNameIntensity' in self._kwArgs:
assert not 'extraTracks' in self._kwArgs
self._kwArgs['extraTracks'] = self._kwArgs['trackNameIntensity']
if 'extraTracks' in self._kwArgs:
from gold.track.Track import PlainTrack
import re
from config.Config import MULTIPLE_EXTRA_TRACKS_SEPARATOR
extraTracks = self._kwArgs['extraTracks']
if type(extraTracks) == str:
extraTracks = extraTracks.split(MULTIPLE_EXTRA_TRACKS_SEPARATOR)
for extraT in extraTracks:
if type(extraT) == str:
#extraT = extraT.split('|')
#extraT = re.split('\^|\|',extraT)
extraT = convertTNstrToTNListFormat(extraT)
if type(extraT) == list:
#print 'TEMP1: ', extraT
from urllib import unquote
extraT = [unquote(part) for part in extraT]
from quick.application.ExternalTrackManager import ExternalTrackManager
if ExternalTrackManager.isGalaxyTrack(extraT):
extraT = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(self.getGenome(), extraT)
extraT = PlainTrack(extraT)
tracks.append(extraT)
return tracks
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.ExternalTrackManager import ExternalTrackManager
genome = choices[0]
preProcTN1 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, choices[2].split(
':')) if choices[1] == 'history' else choices[2].split(':')
chrSizeDict = dict([(chrom, GenomeInfo.getChrLen(genome, chrom))
for chrom in GenomeInfo.getChrList(genome)])
trackType = choices[3].split(':')[1]
fnSource = ExternalTrackManager.extractFnFromGalaxyTN(
choices[3].split(':'))
if trackType in ['valued.bed', 'category.bed', 'bed']:
geSource = GenomeElementSorter(
BedGenomeElementSource(fnSource, genome=genome)).__iter__()
elif trackType == 'gtrack':
geSource = GenomeElementSorter(
GtrackGenomeElementSource(fnSource, genome=genome)).__iter__()
#headLinesStr = geSource.getHeaderLines().replace('##','\n##')
else:
raise InvalidFormatError(
'The Binning must be of the following formats: gtrack, valued.bed, category.bed ,bed ...'
)
cls.PrintResultToHistItem(galaxyFn, geSource, preProcTN1, genome,
username)
def visitGalaxyGSuiteTrack(self, gSuiteTrack):
self.genericVisit(gSuiteTrack)
from quick.application.ExternalTrackManager import ExternalTrackManager
from gold.description.TrackInfo import TrackInfo
if gSuiteTrack.hasExtraFileName():
baseFileName = os.path.basename(gSuiteTrack.uriWithoutSuffix)
else:
baseFileName = gSuiteTrack.title
galaxyTN = ExternalTrackManager.constructGalaxyTnFromSuitedFn(
gSuiteTrack.path, fileEnding=gSuiteTrack.suffix, name=baseFileName)
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
gSuiteTrack.genome,
galaxyTN,
printErrors=False,
printProgress=False,
renameExistingTracksIfNeeded=False)
trackType = TrackInfo(gSuiteTrack.genome,
trackName).trackFormatName.lower()
hbUri = HbGSuiteTrack.generateURI(trackName=trackName)
return GSuiteTrack(hbUri,
title=gSuiteTrack.title,
trackType=trackType,
genome=gSuiteTrack.genome,
attributes=gSuiteTrack.attributes,
comment=gSuiteTrack.comment)
def getOptionsBoxTfTracks(cls, prevChoices):
if prevChoices.sourceTfsDetails != cls.SELECT:
genome = prevChoices.genome
sourceTfs = prevChoices.sourceTfs
sourceTfsDetails = prevChoices.sourceTfsDetails
if sourceTfs == cls.SELECT:
return
elif sourceTfs == 'Hyperbrowser repository':
tfSourceTN = TfTrackNameMappings.getTfTrackNameMappings(
prevChoices.genome)[sourceTfsDetails]
subtypes = ProcTrackOptions.getSubtypes(
prevChoices.genome, tfSourceTN, True)
falses = ['False'] * len(subtypes)
return OrderedDict(zip(subtypes, falses))
elif sourceTfs == cls.REGIONS_FROM_HISTORY:
if isinstance(sourceTfsDetails, basestring):
galaxyTN = sourceTfsDetails.split(':')
if galaxyTN[
1] == "gsuite": #ExternalTrackManager.extractFileSuffixFromGalaxyTN(prevChoices.sourceTfsDetails, allowUnsupportedSuffixes=True) == "gsuite"
errorString = GeneralGuiTool._checkGSuiteFile(
sourceTfsDetails)
if not errorString:
gSuite = getGSuiteFromGalaxyTN(sourceTfsDetails)
sizeErrorString = GeneralGuiTool._checkGSuiteTrackListSize(
gSuite, 1, 1000)
if not sizeErrorString:
reqErrorString = GeneralGuiTool._checkGSuiteRequirements \
(gSuite,
AllTfsOfRegions.GSUITE_ALLOWED_FILE_FORMATS,
AllTfsOfRegions.GSUITE_ALLOWED_LOCATIONS,
AllTfsOfRegions.GSUITE_ALLOWED_TRACK_TYPES,
AllTfsOfRegions.GSUITE_DISALLOWED_GENOMES)
if not reqErrorString:
validity = 'Valid'
else:
return
else:
return
else:
return
if validity == 'Valid':
selectedTrackNames = []
gSuite = getGSuiteFromGalaxyTN(sourceTfsDetails)
for track in gSuite.allTracks():
selectedTrackNames.append(':'.join(
track.trackName))
falses = ['False'] * len(selectedTrackNames)
return OrderedDict(zip(selectedTrackNames, falses))
else:
tfTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, galaxyTN)
return [':'.join(tfTrackName)]
else:
return
else:
return
else:
return
def execute(cls, choices, galaxyFn=None, username=''):
genome = choices.genome
sourceHic = choices.sourceHic
hicdic = choices.hicdic
annotation = choices.annotation
# Get Genomic Regions track names:
selectedTrackNames = []
if isinstance(hicdic, dict):
selectedHiCRegions = [
key for key, val in hicdic.iteritems() if val == 'True'
]
else:
selectedHiCRegions = hicdic
if sourceHic == 'Hyperbrowser repository':
selectedTrackNames = []
for i in selectedHiCRegions:
hicTrackName = HiCNameMappings.getHiCNameMappings(genome)[i]
selectedTrackNames.append(hicTrackName)
print '<p>', i, '</p>'
print getTrackRelevantInfo.getNumberElements(
genome, hicTrackName)
print getTrackRelevantInfo.getSegmentSizes(
genome, hicTrackName)
#print getTrackRelevantInfo.getAnchor(genome, hicTrackName)
targetBins = getTrackRelevantInfo.getGenomicElements(
genome, hicTrackName)
print targetBins
print '<p>==============================================</p>'
analysis = AntoniosTool.getHiCFileFromTargetBins(
targetBins, galaxyFn)
print '<p>', analysis.getLoadToHistoryLink(
'Send file to History'), '</p>'
print '<p>==============================================</p>'
elif sourceHic == cls.REGIONS_FROM_HISTORY:
galaxyTN = selectedHiCRegions.split(':')
hicTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, galaxyTN)
print '<p>', selectedHiCRegions, '</p>'
print '<p>', hicTrackName, '</p>'
numElements = getTrackRelevantInfo.getNumberElements(
genome, hicTrackName)
print 'numElements=', sum(numElements), numElements
print getTrackRelevantInfo.getSegmentSizes(genome, hicTrackName)
trackElements = getTrackRelevantInfo.getGenomicElements(
genome, hicTrackName)
print len(trackElements), trackElements
print '<p>==============================================</p>'
analysis = AntoniosTool.getHiCFileFromTargetBins(
trackElements, galaxyFn)
print '<p>', analysis.getLoadToHistoryLink(
'Send file to History'), '</p>'
print '<p>==============================================</p>'
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]
trackList = choices[2].values()
print trackList
from quick.application.ExternalTrackManager import ExternalTrackManager
for track in trackList:
ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, track.split(':'))
print 'Executing...'
def _collectTracks(self):
tracks = [self._track, self._track2]
#Map trackNameIntensity over to extraTrack, if extraTrack does not exist in itself (to allow selection of third track in GUI and through some existing code interfaces)
if 'trackNameIntensity' in self._kwArgs:
if not 'extraTracks' in self._kwArgs:
self._kwArgs['extraTracks'] = self._kwArgs[
'trackNameIntensity']
if 'extraTracks' in self._kwArgs:
from gold.util.CommonConstants import MULTIPLE_EXTRA_TRACKS_SEPARATOR
extraTracks = self._kwArgs['extraTracks']
if isinstance(extraTracks, basestring):
extraTracks = extraTracks.split(
MULTIPLE_EXTRA_TRACKS_SEPARATOR)
for extraT in extraTracks:
#HACK for tests
if isinstance(
extraT,
(Track, SampleTrack)): #SampleTrack is used by tests
if str(extraT) not in self._trackDict:
self._trackDict[str(extraT)] = extraT
tracks.append(extraT)
else:
tracks.append(self._trackDict[str(extraT)])
#####################################
else:
if isinstance(extraT, basestring):
#extraT = extraT.split('|')
#extraT = re.split('\^|\|',extraT)
extraT = convertTNstrToTNListFormat(extraT,
doUnquoting=False)
if type(extraT) == list:
from urllib import unquote
extraT = [unquote(part) for part in extraT]
from quick.application.ExternalTrackManager import ExternalTrackManager
if ExternalTrackManager.isGalaxyTrack(extraT):
extraT = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
self.getGenome(), extraT)
#extraT = PlainTrack(extraT)
#extraT = Track(extraT)
#tracks.append(extraT)
if str(extraT) not in self._trackDict:
extraTrack = Track(extraT)
self._trackDict[str(extraT)] = extraTrack
tracks.append(extraTrack)
else:
tracks.append(self._trackDict[str(extraT)])
return tracks
def execute(cls, choices, galaxyFn=None, username=''):
import time
start = time.clock()
# HTML settings
from gold.result.HtmlCore import HtmlCore
htmlCore = HtmlCore()
htmlCore.divBegin(style=cls.HTML_STYLE)
# Set debug environment
cls._setDebugModeIfSelected(choices)
# Print tool information
cls.htmlClusterTitle(cls.getToolName(), htmlCore)
cls.htmlClusterSubtext(choices.distanceMeasure, cls.CLUSTER_LIST, choices.linkageCriterion, htmlCore)
htmlCore.line('Threshold of r<sup>2</sup>: ' + choices.rSquare)
# Analysis environment
gSuite = getGSuiteFromGalaxyTN(choices.gSuite)
analysisBins = GlobalBinSource(gSuite.genome)
analysisSpec = AnalysisSpec(ExpandTrackAndMatchStat)
splitName = choices.ldTrack.split(":")
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(gSuite.genome, splitName)
linkedPointTrack = Track(trackName)
# Find distance/correlation matrix
labels = []
distDict = cls.createDistDict(cls.CLUSTER_LIST)
size = gSuite.numTracks()
for i in range(0, size):
gSuiteTrack1 = gSuite.getTrackFromIndex(i)
labels.append(gSuiteTrack1.title)
for j in range(i + 1, size):
gSuiteTrack2 = gSuite.getTrackFromIndex(j)
track1 = Track(gSuiteTrack1.trackName)
track2 = Track(gSuiteTrack2.trackName)
count = doAnalysis(analysisSpec, analysisBins, [track1, track2, linkedPointTrack]).getGlobalResult()
cls.updateDistDict(distDict, count)
# Cluster and print plots
cls.printDistPlots(distDict, labels, choices.distanceMeasure, choices.linkageCriterion, galaxyFn, htmlCore)
cls.htmlClusterTime(str(time.clock() - start), htmlCore)
htmlCore.divEnd()
print htmlCore
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.ExternalTrackManager import ExternalTrackManager
genome = choices[0]
preProcTN1 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices[2].split(':')) if choices[1] == 'History' else choices[2].split(':')
chrSizeDict = dict([ ( chrom, GenomeInfo.getChrLen(genome, chrom)) for chrom in GenomeInfo.getChrList(genome)])
trackType = choices[3].split(':')[1]
fnSource = ExternalTrackManager.extractFnFromGalaxyTN(choices[3].split(':'))
if trackType in ['marked.bed', 'category.bed', 'bed']:
geSource = GenomeElementSorter(BedGenomeElementSource(fnSource, genome=genome)).__iter__()
elif trackType == 'gtrack':
geSource = GenomeElementSorter(GtrackGenomeElementSource(fnSource, genome=genome)).__iter__()
#headLinesStr = geSource.getHeaderLines().replace('##','\n##')
else:
raise InvalidFormatError('The Binning must be of the following formats: gtrack, marked.bed, category.bed ,bed ...')
cls.PrintResultToHistItem( galaxyFn, geSource, preProcTN1, genome, username)
def preprocessTrack(genome, dataset):
suffixForFile = extractFileSuffixFromDatasetInfo(dataset)
if (suffixForFile == 'bed'):
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, dataset, printErrors=False, printProgress=False)
return trackName
elif (suffixForFile == 'gsuite'):
trackName = []
trackTitle = []
# l = []
gSuite = getGSuiteFromGalaxyTN(dataset)
for i, iTrack in enumerate(gSuite.allTracks()):
# l = dir(iTrack)
# if 'trackName' in l:
# print(iTrack.trackName)
# else:
# print('not found')
#print iTrack, iTrack.trackName, iTrack.title, '<br>'
trackName.append(iTrack.trackName)
trackTitle.append(iTrack.title)
return trackName, trackTitle
def execute(cls, choices, galaxyFn=None, username=""):
from quick.application.GalaxyInterface import GalaxyInterface
fileformat = choices[6]
outputFile = open(galaxyFn, "w")
if fileformat == "html":
print GalaxyInterface.getHtmlBeginForRuns(galaxyFn)
print GalaxyInterface.getHtmlForToggles(withRunDescription=False)
t = calendar.timegm(time.gmtime())
htmlfile = GalaxyRunSpecificFile(["fet", str(t)], galaxyFn)
genome = choices[0]
track1 = choices[1].split(":")
track2 = choices[2].split(":")
tn1 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, track1)
tn2 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, track2)
windowSize = int(choices[3])
windowStep = int(choices[4])
percentile = float(choices[5])
# results = {}
# TODO: why this?
# tr = Track(tn1)
# tr.addFormatReq(TrackFormatReq(dense=False, allowOverlaps=True))
outputFile.write("#seqid\tstart\tscore\tstddev\n")
if fileformat == "html":
text = "#seqid\tstart\tscore\tstddev\n"
print "chrs:", str(GenomeInfo.getChrList(genome))
reg = "*"
bins = "*"
analysisDef = "Dummy: dummy name ([wStep=%g] [wSize=%g] [percentile=%g])-> FisherExactScoreStat" % (
windowStep,
windowSize,
percentile,
)
userBinSource = GalaxyInterface._getUserBinSource(reg, bins, genome)
result = GalaxyInterface.runManual([tn1, tn2], analysisDef, reg, bins, genome, galaxyFn=galaxyFn)
for key in result.getAllRegionKeys():
chrom = str(key).split(":")[0]
r = result[key]
if "Result" not in r.keys():
print "skipping chr:", chrom, r
continue
r = r["Result"]
scores = r[0]
stddev = r[1]
for i in range(len(scores)):
if scores[i] != 0:
pos = i * windowStep
# if choices[5] == "html":
# print "%s\t%s\t%s\t%s\n" % (str(chrom), pos, str(scores[i]), str(stddev[i]))
if fileformat == "tabular":
outputFile.write("%s\t%s\t%s\t%s\n" % (str(chrom), pos, str(scores[i]), str(stddev[i])))
else:
text += "%s\t%s\t%s\t%s\n" % (str(chrom), pos, str(scores[i]), str(stddev[i]))
if fileformat == "html":
htmlfile.writeTextToFile(text)
print htmlfile.getLink("Result file")
print GalaxyInterface.getHtmlEndForRuns()
outputFile.close()
def execute(cls, choices, galaxyFn=None, username=''):
cls._setDebugModeIfSelected(choices)
genome = choices.genome
genomicRegions = choices.genomicRegions
genomicRegionsTracks = choices.genomicRegionsTracks
sourceTfs = choices.sourceTfs
sourceTfsDetails = choices.sourceTfsDetails
tfTracks = choices.tfTracks
# Get Genomic Region track name:
if genomicRegions == cls.REGIONS_FROM_HISTORY:
galaxyTN = genomicRegionsTracks.split(':')
genElementTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, galaxyTN)
#queryGSuite = getGSuiteFromGalaxyTN(genomicRegionsTracks)
#queryTrackList = [Track(x.trackName, x.title) for x in queryGSuite.allTracks()]
elif genomicRegions == 'Hyperbrowser repository':
selectedGenRegTrack = TfbsTrackNameMappings.getTfbsTrackNameMappings(
genome)[genomicRegionsTracks]
if isinstance(selectedGenRegTrack, dict):
genElementTrackName = selectedGenRegTrack.values()
else:
genElementTrackName = selectedGenRegTrack
elif genomicRegions == 'Hyperbrowser repository (cell-type-specific)':
genElementTrackName = ['Private', 'Antonio'
] + genomicRegionsTracks.split(':')
else:
return
# Get TF track names:
if isinstance(tfTracks, dict):
selectedTfTracks = [
key for key, val in tfTracks.iteritems() if val == 'True'
]
else:
selectedTfTracks = [tfTracks]
queryTrackTitle = '--'.join(genElementTrackName)
trackTitles = [queryTrackTitle]
tracks = [Track(genElementTrackName, trackTitle=queryTrackTitle)]
for i in selectedTfTracks:
if sourceTfs == 'Hyperbrowser repository':
tfTrackName = TfTrackNameMappings.getTfTrackNameMappings(
genome)[sourceTfsDetails] + [i]
tracks.append(
Track(tfTrackName,
trackTitle=tfTrackName[len(tfTrackName) - 1]))
trackTitles.append(tfTrackName[len(tfTrackName) - 1])
else:
tfTrackName = i.split(':')
queryGSuite = getGSuiteFromGalaxyTN(sourceTfsDetails)
for x in queryGSuite.allTracks():
selectedTrackNames = (':'.join(x.trackName))
if i == selectedTrackNames:
tracks.append(Track(x.trackName, x.title))
trackTitles.append(x.trackName[-1])
# queryGSuite = getGSuiteFromGalaxyTN(sourceTfsDetails)
# tfTrackName = [x.trackName for x in queryGSuite.allTracks()] + [i]
# tracks += [Track(x.trackName, x.title) for x in queryGSuite.allTracks()]
# trackTitles += tfTrackName
# print tfTrackName
# print tracks
# print trackTitles
trackTitlesForStat = trackTitles
trackTitles = CommonConstants.TRACK_TITLES_SEPARATOR.join(trackTitles)
##first statistic for Q2
resultsForStatistics = OrderedDict()
similarityFunc = [ #GSuiteStatUtils.T7_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP,
GSuiteStatUtils.T5_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP
]
for similarityStatClassName in similarityFunc:
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisBins = GalaxyInterface._getUserBinSource(regSpec,
binSpec,
genome=genome)
mcfdrDepth = AnalysisDefHandler(
REPLACE_TEMPLATES['$MCFDR$']).getOptionsAsText().values()[0][0]
analysisDefString = REPLACE_TEMPLATES[
'$MCFDR$'] + ' -> GSuiteSimilarityToQueryTrackRankingsAndPValuesWrapperStat'
analysisSpec = AnalysisDefHandler(analysisDefString)
analysisSpec.setChoice('MCFDR sampling depth', mcfdrDepth)
analysisSpec.addParameter('assumptions',
'PermutedSegsAndIntersegsTrack_')
analysisSpec.addParameter(
'rawStatistic', GSuiteStatUtils.
PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similarityStatClassName])
analysisSpec.addParameter(
'pairwiseStatistic', GSuiteStatUtils.
PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similarityStatClassName]
) #needed for call of non randomized stat for assertion
analysisSpec.addParameter('tail', 'more')
analysisSpec.addParameter('trackTitles',
trackTitles) #that need to be string
analysisSpec.addParameter('queryTracksNum', str(len(tracks)))
results = doAnalysis(analysisSpec, analysisBins,
tracks).getGlobalResult()
if not similarityStatClassName in resultsForStatistics:
resultsForStatistics[similarityStatClassName] = {}
resultsForStatistics[similarityStatClassName] = results
keyTitle = [
#'Normalized ratio of observed to expected overlap (normalized Forbes similarity measure)',
'Ratio of observed to expected overlap (Forbes similarity measure)'
]
# 'Normalized Forbes coefficient: ratio of observed to expected overlap normalized in relation to the reference GSuite',
# 'Forbes coefficient: ratio of observed to expected overlap'
keyTitle = [
#GSuiteStatUtils.T7_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP,
GSuiteStatUtils.T5_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP
]
resultDict = AllTfsOfRegions.countStatistics(similarityFunc, choices,
genome, tracks,
trackTitlesForStat)
resultDictShow = AllTfsOfRegions.countStatisticResults(
resultDict, keyTitle, trackTitlesForStat)
# print resultsForStatistics
'''selectedTrackNames = []
if sourceTfs == 'History (user-defined)':
if selectedTfTracks.split(":")[1] == "gsuite":
gSuite = getGSuiteFromGalaxyTN(selectedTfTracks)
for track in gSuite.allTracks():
selectedTrackNames.append(track.trackName)
else:
galaxyTN = selectedTfTracks.split(':')
gRegTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, galaxyTN)
selectedTrackNames.append(gRegTrackName)
else:'''
tfNameList = []
#Intersection between TF Tracks and selected region (Table 1):
n = 0
allTargetBins = []
alltfNames = []
table1 = []
for i in selectedTfTracks:
n = n + 1
#newGalaxyFn = galaxyFn.split(".")[0] + str(n) + "." + "dat"
if sourceTfs == 'Hyperbrowser repository':
tfTrackName = TfTrackNameMappings.getTfTrackNameMappings(
genome)[sourceTfsDetails] + [i]
else:
tfTrackName = i.split(':')
tfTrackName.pop(0)
#tfIntersection.expandReferenceTrack(upFlankSize, downFlankSize)
tfIntersection = TrackIntersection(genome, genElementTrackName,
tfTrackName, galaxyFn, str(n))
regFileNamer = tfIntersection.getIntersectedRegionsStaticFileWithContent(
)
targetBins = tfIntersection.getIntersectedReferenceBins()
#regSpec, targetBins = UserBinSelector.getRegsAndBinsSpec(choices)
tfHits = [i] * len(targetBins)
fixedTargetBins = [str(a).split(" ")[0] for a in targetBins]
extendedTargetBins = [
list(a) for a in zip(fixedTargetBins, tfHits)
]
allTargetBins = allTargetBins + extendedTargetBins
tfName = i
alltfNames = alltfNames + [tfName]
# Save output table:
tfNameList.append(tfName)
line = [tfName] + [len(targetBins)] + [
regFileNamer.getLink('Download bed-file')
] + [
regFileNamer.getLoadToHistoryLink('Send bed-file to History')
]
table1 = table1 + [line]
# Computing totals:
fullCase = ','.join(alltfNames)
firstColumn = [item[0] for item in allTargetBins]
uniqueAllTargetBins = list(set(firstColumn))
# Group TFs by bound region:
d1 = defaultdict(list)
for k, v in allTargetBins:
d1[k].append(v)
allTFTargetBins = dict((k, ','.join(v)) for k, v in d1.iteritems())
allTFTargetList = []
fullCaseTFTargetList = []
for key, value in allTFTargetBins.iteritems():
allTFTargetList = allTFTargetList + [[key, value]]
if value == fullCase:
fullCaseTFTargetList = fullCaseTFTargetList + [[key, value]]
analysis3 = TrackIntersection.getFileFromTargetBins(
allTFTargetList, galaxyFn, str(3))
analysis4 = TrackIntersection.getFileFromTargetBins(
fullCaseTFTargetList, galaxyFn, str(4))
# Print output to table:
title = 'TF targets and co-occupancy of ' + genElementTrackName[
-1] + ' genomic regions'
htmlCore = HtmlCore()
pf = plotFunction(tableId='resultsTable')
htmlCore.begin()
htmlCore.header(title)
htmlCore.divBegin('resultsDiv')
htmlCore.line(pf.createButton(bText='Show/Hide more results'))
# htmlCore.tableHeader(['Transcription Factor', 'Normalized ratio of observed to expected overlap (normalized Forbes similarity measure) -- Similarity to genomic regions track', 'Normalized ratio of observed to expected overlap (normalized Forbes similarity measure) -- p-value','Ratio of observed to expected overlap (Forbes similarity measure) -- Similarity to genomic regions track', 'Ratio of observed to expected overlap (Forbes similarity measure) -- p-value', 'Number of TF-Target Track Regions', 'File of TF Target Regions', 'File of TF Target Regions', 'Number of TF-co-occupied Regions', 'File of TF co-occupied Regions', 'File of TF co-occupied Regions', 'Rank of TF co-occupancy motifs', 'Rank of TF co-occupancy motifs'], sortable=True, tableId='resultsTable')
#previous ordering
# htmlCore.tableHeader(['Transcription Factor', 'Normalized Forbes index --overlap score',
# 'Normalized Forbes index --p-value',
# 'Forbes index --overlap score', 'Forbes index --p-value',
# 'Number of TF-Target Track Regions', 'File of TF Target Regions',
# 'File of TF Target Regions', 'Number of target track regions occupied by this TF',
# 'File of TF co-occupied Regions', 'File of TF co-occupied Regions',
# 'Rank of TF co-occupancy motifs', 'Rank of TF co-occupancy motifs'],
# sortable=True, tableId='resultsTable')
htmlCore.tableHeader(
[
'Transcription Factor',
'Number of TF-Target Track Regions',
'File of TF Track Regions',
'Number of target track regions occupied by this TF',
'File of TF Target Regions',
'Forbes index --overlap score',
'Forbes index --p-value',
#'Normalized Forbes index --overlap score', 'Normalized Forbes index --p-value',
'File of TF co-occupied Regions',
'Rank of TF co-occupancy motifs'
],
sortable=True,
tableId='resultsTable')
# Adding co-occupancy results to table:
n = 1000
genRegionNumElements = [
int(x) for x in getTrackRelevantInfo.getNumberElements(
genome, genElementTrackName)
]
for key0, it0 in resultsForStatistics.iteritems():
for el in tfNameList:
if el not in it0:
resultsForStatistics[key0][el] = [None, None]
resultsPlotDict = {}
resultPlotCat = []
resultsPlot = []
resultsForStatisticsProper = {}
for key0, it0 in resultsForStatistics.iteritems():
if not key0 in resultsPlotDict:
resultsPlotDict[key0] = {}
resultsPlotPart = []
for key1, it1 in it0.iteritems():
resultsPlotPart.append(it1[0])
if not key1 in resultsForStatisticsProper:
resultsForStatisticsProper[key1] = []
if not key1 in resultsPlotDict[key0]:
resultsPlotDict[key0][key1] = None
for el in it1:
resultsForStatisticsProper[key1].append(el)
resultsPlotDict[key0][key1] = it1[0]
resultPlotCat.append(tfNameList)
resultPlotCat.append(tfNameList)
#resultPlotCatPart = tfNameList
# print resultPlotCatPart
for key0, it0 in resultsPlotDict.iteritems():
resultsPlotPart = []
for el in tfNameList:
if el in it0:
resultsPlotPart.append(it0[el])
else:
resultsPlotPart.append(None)
resultsPlot.append(resultsPlotPart)
for i in table1:
thisCaseTFTargetList = []
for key, value in allTFTargetList:
if i[0] in value and ',' in value:
thisCaseTFTargetList = thisCaseTFTargetList + [[
key, value
]]
n = n + 1
thisAnalysis = TrackIntersection.getFileFromTargetBins(
thisCaseTFTargetList, galaxyFn, str(n))
thisCaseCoCountsList = []
thing = [x[1] for x in thisCaseTFTargetList]
for k in list(set(thing)):
thisCount = thing.count(k)
thisCaseCoCountsList = thisCaseCoCountsList + \
[[k, thisCount, 100*float(thisCount)/float(sum(genRegionNumElements)), 100*float(thisCount)/float(len(thisCaseTFTargetList))]]
thisCaseCoCountsList.sort(key=lambda x: x[2], reverse=True)
n = n + 1
thisCoCountsAnalysis = TrackIntersection.getOccupancySummaryFile(
thisCaseCoCountsList, galaxyFn, str(n))
thisLine = [len(thisCaseTFTargetList)] + \
[thisAnalysis.getLink('Download file')] + [thisAnalysis.getLoadToHistoryLink('Send file to History')] + \
[thisCoCountsAnalysis.getLink('Download file')] + [thisCoCountsAnalysis.getLoadToHistoryLink('Send file to History')]
newLineI = []
tfName = i[0]
newLineI.append(tfName)
for el in resultsForStatisticsProper[tfName]:
newLineI.append(el)
for elN in range(1, len(i)):
newLineI.append(i[elN])
# htmlCore.tableLine(i + thisLine)
# htmlCore.tableHeader(['Transcription Factor', 'Normalized Forbes index --overlap score',
# 'Normalized Forbes index --p-value',
# 'Forbes index --overlap score', 'Forbes index --p-value',
# 'Number of TF-Target Track Regions', 'File of TF Target Regions',
# 'File of TF Target Regions', 'Number of target track regions occupied by this TF',
# 'File of TF co-occupied Regions', 'File of TF co-occupied Regions',
# 'Rank of TF co-occupancy motifs', 'Rank of TF co-occupancy motifs'],
# sortable=True, tableId='resultsTable')
# htmlCore.tableHeader(['Transcription Factor', 'Number of TF-Target Track Regions', 'File of TF Track Regions',
# 'Number of target track regions occupied by this TF', 'File of TF Target Regions',
# 'Forbes index --overlap score', 'Forbes index --p-value',
# 'Normalized Forbes index --overlap score', 'Normalized Forbes index --p-value',
# 'File of TF co-occupied Regions', 'Rank of TF co-occupancy motifs'],
# sortable=True, tableId='resultsTable')
tl = newLineI + thisLine
# previous ordering tl - 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
# actual ordering - 0, 5, 7, 8, 7, 3, 4, 1, 2, 9, 11
#ordering = [0, 5, 7, 8, 10, 3, 4, 1, 2, 10, 12]
ordering = [0, 3, 5, 6, 8, 1, 2, 8, 10]
#1, 2, => delete
eoList = []
for eo in ordering:
eoList.append(tl[eo])
htmlCore.tableLine(eoList)
totalCoOccupancyTargetList = []
n = 2000
for key, value in allTFTargetList:
n = n + 1
if ',' in value:
totalCoOccupancyTargetList = totalCoOccupancyTargetList + [[
key, value
]]
#newGalaxyFn = galaxyFn.split(".")[0] + str(n) + "." + "dat"
totalCoOccupancyAnalysis = TrackIntersection.getFileFromTargetBins(
totalCoOccupancyTargetList, galaxyFn, str(n))
#line = ['Total reported regions'] + [len(allTargetBins)] + [''] + [''] + [''] + [''] + ['']
#line = ['Full co-occupancy of ' + fullCase] + ['-'] + ['-'] + ['-'] + ['-'] + ['-'] + ['-'] + ['-'] + [len(fullCaseTFTargetList)] + [analysis4.getLink('Download file')] + [analysis4.getLoadToHistoryLink('Send file to History')] + ['-'] + ['-']
line = ['Full co-occupancy of ' + fullCase] + \
['-'] + \
['-'] + \
[len(fullCaseTFTargetList)] + \
['-'] + \
['-'] + \
['-'] + \
[analysis4.getLoadToHistoryLink('Send file to History')] + \
['-']
htmlCore.tableLine(line)
#line = ['Total unique regions'] + ['-'] + ['-'] + ['-'] + ['-'] + [len(allTFTargetList)] + [analysis3.getLink('Download bed-file')] + [analysis3.getLoadToHistoryLink('Send bed-file to History')] + [len(totalCoOccupancyTargetList)] + [totalCoOccupancyAnalysis.getLink('Download file')] + [totalCoOccupancyAnalysis.getLoadToHistoryLink('Send file to History')] + ['-'] + ['-']
line = ['Total unique regions'] + \
[len(allTFTargetList)] + \
['-'] + \
[len(totalCoOccupancyTargetList)] + \
[analysis3.getLoadToHistoryLink('Send bed-file to History')] + \
['-'] +\
['-'] + \
[totalCoOccupancyAnalysis.getLoadToHistoryLink('Send file to History')] + \
['-']
htmlCore.tableLine(line)
htmlCore.tableFooter()
htmlCore.divEnd()
# htmlCore.line(pf.hideColumns(indexList=[2, 4]))
#
sumRes = 0
for r in resultsPlot[0]:
if r != None:
sumRes += r
if sumRes != 0:
vg = visualizationGraphs()
result = vg.drawColumnCharts(
[resultsPlot[0]],
height=300,
categories=resultPlotCat,
legend=False,
addOptions='width: 90%; float:left; margin: 0 4%;',
#titleText=['Overlap between TFs and genomic region using normalized Forbes', 'Overlap between TFs and genomic region using Forbes'],
titleText=[
'Overlap between TFs and genomic region using Forbes'
],
xAxisRotation=90,
xAxisTitle='TF',
yAxisTitle='value')
htmlCore.line(result)
for key0, it0 in resultDictShow.iteritems():
htmlCore.divBegin('resultsDiv' + str(key0))
htmlCore.header(key0)
htmlCore.tableHeader(it0[0],
sortable=True,
tableId='resultsTable' + str(key0))
for elN in range(1, len(it0)):
htmlCore.tableLine(it0[elN])
htmlCore.tableFooter()
htmlCore.divEnd()
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
def execute(cls, choices, galaxyFn=None, username=''):
cls._setDebugModeIfSelected(choices)
genome = choices.genome
genomicRegionsSource = choices.genomicRegionsSource
genomicRegions = choices.genomicRegions
#upFlankSize = int(choices.upFlankSize)
#downFlankSize = int(choices.downFlankSize)
sourceTfs = choices.sourceTfs
tfTracks = choices.tfTracks
# Get TF track name:
if sourceTfs == cls.REGIONS_FROM_HISTORY:
galaxyTN = tfTracks.split(':')
tfTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, galaxyTN)
else:
tfTrackName = TfTrackNameMappings.getTfTrackNameMappings(
genome)[sourceTfs] + [tfTracks]
# Get Genomic Regions track names:
selectedTrackNames = []
if isinstance(genomicRegions, dict):
selectedGenRegions = [
key for key, val in genomicRegions.iteritems() if val == 'True'
]
else:
selectedGenRegions = genomicRegions
if genomicRegionsSource == 'Hyperbrowser repository (single tracks)':
for i in selectedGenRegions:
selectedTrackNames.append(
TfbsTrackNameMappings.getTfbsTrackNameMappings(genome)[i])
elif genomicRegionsSource == 'Hyperbrowser repository (cell-specific multi-tracks)':
for i in selectedGenRegions:
genElementGSuiteName = TfbsGSuiteNameMappings.getTfbsGSuiteNameMappings(
genome)[i]
gSuite = getGSuiteFromGSuiteFile(genElementGSuiteName)
for track in gSuite.allTracks():
selectedTrackNames.append(track.trackName)
elif genomicRegionsSource == 'History (user-defined)':
if genomicRegions.split(":")[1] == "gsuite":
gSuite = getGSuiteFromGalaxyTN(selectedGenRegions)
for track in gSuite.allTracks():
selectedTrackNames.append(track.trackName)
else:
galaxyTN = selectedGenRegions.split(':')
gRegTrackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome, galaxyTN)
selectedTrackNames.append(gRegTrackName)
else:
return
#Intersection:
title = 'Targets of ' + tfTrackName[-1] + ' TF track'
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.header(title)
htmlCore.divBegin('resultsDiv')
htmlCore.tableHeader([
'Genomic Region', 'Number of Target Regions',
'Download bed file of Target Regions', 'Send bed file to history'
],
sortable=True,
tableId='resultsTable')
n = 0
allTargetBins = []
dataY = []
allRefSetNames = []
#print 'all:', selectedTrackNames, '<p>'
#print 'tf:', tfTrackName, '<p>'
for i in selectedTrackNames:
n = n + 1
#newGalaxyFn = galaxyFn.split(".")[0] + str(n) + "." + "dat"
tfIntersection = TrackIntersection(genome, i, tfTrackName,
galaxyFn, str(n))
#tfIntersection.expandReferenceTrack(upFlankSize, downFlankSize)
regFileNamer = tfIntersection.getIntersectedRegionsStaticFileWithContent(
)
targetBins = tfIntersection.getIntersectedReferenceBins()
'''print 'Target Bins = ', targetBins, '<p>'
if genomicRegionsSource=='Hyperbrowser repository (single tracks)':
print '\"', tfTracks, '\" in \"', ":".join((i[len(i)-2],i[len(i)-1])), '":<p>'
elif genomicRegionsSource=='History (user-defined)':
print '\"', tfTracks, '\" in \"', i[len(i)-1], '":<p>'
else:
listGenRegion = i[0].split(":")
maxIndex = len(listGenRegion)-1
print '\"', tfTracks, '\" in \"', ":".join((listGenRegion[maxIndex-1],listGenRegion[maxIndex])), '":<p>'
print '<p>Number of Targets = ', len(targetBins), 'regions.</p>'
print '<p>', regFileNamer.getLink('Download bed-file'), ' of all regions with 1 or more hits.</p>'
print '<p>', regFileNamer.getLoadToHistoryLink('Download bed-file to History'), ' of all regions with 1 or more hits.</p>'
print '<p>==============================================</p>'
#with open(galaxyFn, 'w') as outFile:
#print>>outFile, 'TargetBins=', targetBins, '<p>'
#print >>outFile, selectedGenRegions, '<p>' '''
# Collect all target bins and data to plot:
allTargetBins = allTargetBins + targetBins
dataY = dataY + [
TrackIntersection.prepareDataForPlot(genome, targetBins)
]
refSetName = i[len(i) - 1]
allRefSetNames = allRefSetNames + [refSetName]
# Print output to table:
line = [refSetName] + [len(targetBins)] + [
regFileNamer.getLink('Download bed-file')
] + [
regFileNamer.getLoadToHistoryLink(
'Download bed-file to History')
]
#print line, '<p>'
htmlCore.tableLine(line)
line = ['Total'] + [len(allTargetBins)] + [''] + ['']
dataY = dataY + [
TrackIntersection.prepareDataForPlot(genome, allTargetBins)
]
allRefSetNames = allRefSetNames + ['Total']
htmlCore.tableLine(line)
htmlCore.tableFooter()
htmlCore.divEnd()
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
#print 'ALL Target Bins = ', allTargetBins, '<p>'
#print 'dataY = ', dataY, '<p>'
# Plot:
if genome == 'hg19':
chrNames = [
'chr1', 'chr2', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8',
'chr9', 'chr10', 'chr11', 'chr12', 'chr13', 'chr14', 'chr15',
'chr16', 'chr17', 'chr18', 'chr19', 'chr20', 'chr21', 'chr22',
'chrX', 'chrY'
]
if genome == 'mm9':
chrNames = [
'chr1', 'chr2', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8',
'chr9', 'chr10', 'chr11', 'chr12', 'chr13', 'chr14', 'chr15',
'chr16', 'chr17', 'chr18', 'chr19', 'chrX', 'chrY'
]
titleText = 'Targets per Chromosome'
dataX = [[dataY[i][j] for i in range(len(dataY))]
for j in range(len(dataY[0]))]
seriesType = ['column'] * len(dataX)
categories = allRefSetNames
yAxisTitle = 'Number of Targets'
seriesName = chrNames
shared = False
legend = True
xAxisRotation = 0
#print 'dataX = ', dataX, '<p>'
htmlCore = HtmlCore()
htmlCore.begin()
title = 'Targets of ' + tfTrackName[-1] + ' TF track per chromosome'
htmlCore.header(title)
htmlCore.line('<a href="#" id="linkContainer1">Click to see plot</a>')
htmlCore.divBegin(divId='plotDiv', style=' margin: 0 auto')
htmlCore.line(vp.addJSlibs())
htmlCore.line(vp.useThemePlot())
htmlCore.line(vp.addJSlibsExport())
plot = vp.drawChart(dataX,
type='column',
legend=legend,
height=600,
xAxisRotation=xAxisRotation,
seriesType=seriesType,
seriesName=seriesName,
shared=shared,
titleText=titleText,
overMouseAxisX=True,
categories=categories,
showChartClickOnLink=True)
htmlCore.line(plot)
htmlCore.divEnd()
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.
:param choices: Dict holding all current selections
:param galaxyFn:
:param username:
"""
cls._setDebugModeIfSelected(choices)
genome = choices.genome
queryTrackNameAsList = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices.queryTrack,
printErrors=False,
printProgress=False)
if choices.intensityTrack:
intensityTrackNameAsList = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, choices.intensityTrack,
printErrors=False,
printProgress=False)
else:
intensityTrackNameAsList = None
analysisQuestion = choices.analysisQName
similarityStatClassName = choices.similarityFunc if choices.similarityFunc else GSuiteStatUtils.T5_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP
summaryFunc = choices.summaryFunc if choices.summaryFunc else 'average'
reverse = 'Yes' if choices.reversed else 'No'
if analysisQuestion in [cls.Q2, cls.Q3]:
randStrat = 'PermutedSegsAndIntersegsTrack_' if choices.isBasic else GSuiteStatUtils.PAIRWISE_RAND_CLS_MAPPING[choices.randStrat]
gsuite = getGSuiteFromGalaxyTN(choices.gsuite)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisBins = GalaxyInterface._getUserBinSource(regSpec, binSpec, genome=genome)
queryTrack = Track(queryTrackNameAsList)
tracks = [queryTrack] + [Track(x.trackName, trackTitle=x.title) for x in gsuite.allTracks()]
queryTrackTitle = prettyPrintTrackName(queryTrack.trackName).replace('/', '_')
trackTitles = CommonConstants.TRACK_TITLES_SEPARATOR.join(
[quote(queryTrackTitle)] + [quote(x.title, safe='') for x in gsuite.allTracks()])
additionalResultsDict = OrderedDict()
additionalAttributesDict = OrderedDict()
if analysisQuestion in [cls.Q1, cls.Q2]:
additionalAttributesDict = cls.getSelectedAttributesForEachTrackDict(choices.additionalAttributes, gsuite)
# additional analysis
stats = [SingleValueOverlapStat, CountStat,
CountElementStat] # + [CountSegmentsOverlappingWithT2Stat] #takes long time
additionalResultsDict = runMultipleSingleValStatsOnTracks(gsuite, stats, analysisBins,
queryTrack=queryTrack)
core = HtmlCore()
if analysisQuestion == cls.Q1:
analysisSpec = cls.prepareQ1(reverse, similarityStatClassName, trackTitles)
results = doAnalysis(analysisSpec, analysisBins, tracks).getGlobalResult()
gsPerTrackResultsModel = GSuitePerTrackResultModel(results, ['Similarity to query track'],
additionalResultsDict=additionalResultsDict,
additionalAttributesDict=additionalAttributesDict)
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
gsPerTrackResults = gsPerTrackResultsModel.generateColumnTitlesAndResultsDict(choices.leadAttribute)
else:
gsPerTrackResults = gsPerTrackResultsModel.generateColumnTitlesAndResultsDict()
core = cls.generateQ1output(additionalResultsDict, analysisQuestion, choices, galaxyFn, gsPerTrackResults,
queryTrackTitle, gsuite, results, similarityStatClassName)
elif analysisQuestion == cls.Q2:
analysisSpec = cls.prepareQ2(choices, similarityStatClassName, trackTitles, randStrat,
intensityTrackNameAsList)
results = doAnalysis(analysisSpec, analysisBins, tracks).getGlobalResult()
core = cls.generateQ2Output(additionalAttributesDict, additionalResultsDict,
analysisQuestion, choices, galaxyFn, queryTrackTitle,
gsuite, results, similarityStatClassName)
else: # Q3
analysisSpec = cls.prepareQ3(choices, similarityStatClassName, summaryFunc, randStrat)
results = doAnalysis(analysisSpec, analysisBins, tracks).getGlobalResult()
core = cls.generateQ3output(analysisQuestion, queryTrackTitle, results, similarityStatClassName)
print str(core)
def getSingleTrackTS(genome, guiSelectedTrack, title='Dummy'):
trackName = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, guiSelectedTrack)
return SingleTrackTS(PlainTrack(trackName), {'title': title})
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=''):
'''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 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.
'''
choices = list(choices)
if choices[1] == 'history':
choices[2] = ':'.join(
ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
choices[0], choices[2].split(':')))
genome, inTrackName, tf = cls._getBasicTrackFormat(choices,
tnChoiceIndex=2)
statistic = choices[3]
winSize = int(choices[4])
#print 'Format: ', tf
if tf.split()[-1] == 'points':
if statistic == 'Point density in window':
func = lambda tv: len(tv.startsAsNumpyArray()) * 1.0 / len(tv)
elif statistic == 'Number of points in window':
func = lambda tv: len(tv.startsAsNumpyArray())
elif tf.split()[-1] == 'segments':
if statistic == 'Coverage proportion in window':
func = lambda tv: sum(len(x) for x in tv) * 1.0 / len(tv)
elif statistic == 'Density of segments in window (based on number of segments)':
func = lambda tv: len(tv.startsAsNumpyArray()) * 1.0 / len(tv)
elif statistic == 'Number of segments in window':
func = lambda tv: len(tv.startsAsNumpyArray())
elif tf == 'function':
if statistic == 'Mean function value in window':
func = lambda tv: tv.valsAsNumpyArray().sum(dtype='float64'
) / len(tv)
elif statistic == 'Maximum function value in window':
func = lambda tv: tv.valsAsNumpyArray().max()
elif statistic == 'Minimum function value in window':
func = lambda tv: tv.valsAsNumpyArray().min()
else:
raise
#if choices[4] == 'HyperBrowser track':
# outTrackName = choices[5]
#else:
outTrackName = 'galaxy:hbfunction:%s:Create smoothing or density track' % galaxyFn
from quick.application.GalaxyInterface import GalaxyInterface
print GalaxyInterface.getHbFunctionOutputBegin(galaxyFn,
withDebug=True)
#print 'cleaning..'
#print 'inTrackName: ', inTrackName
#inTrackName = GalaxyInterface._cleanUpTracks([inTrackName], genome, True)[0]
#print 'creating..'
GalaxyInterface.createCustomTrack(genome, inTrackName, outTrackName,
winSize, func, username)
infoMsg = 'A custom track has been created by applying the function "%s" on elements of the track "%s" in a sliding window of size %i along the genome "%s".' % (
statistic, ':'.join(inTrackName), winSize, genome)
print GalaxyInterface.getHbFunctionOutputEnd(infoMsg, withDebug=True)
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=''):
from quick.application.GalaxyInterface import GalaxyInterface
fileformat = choices[9];
outputFile = open(galaxyFn, "w")
if fileformat == "html":
print GalaxyInterface.getHtmlBeginForRuns(galaxyFn)
print GalaxyInterface.getHtmlForToggles(withRunDescription=False)
t = calendar.timegm(time.gmtime())
htmlfile = GalaxyRunSpecificFile(["css", str(t)], galaxyFn);
genome = choices[0]
track1 = choices[1].split(":")
track2 = choices[2].split(":")
tn1 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, track1)
tn2 = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(genome, track2)
compare = choices[3] != "Count individual SNP-differences in window"
if choices[4] == "Classical MDS":
mds = 0;
elif choices[4] == "SMACOF":
mds = 1;
else:
mds = 2;
windowSize = int(choices[5])
windowStep = int(choices[6])
mcTreshold = int(choices[7])
mcRuns = int(choices[8])
outputFile.write("#seqid\tstart\tscore\tp\n")
if fileformat == "html":
text = "#seqid\tstart\tscore\tp\n";
print "chrs:"+str(GenomeInfo.getChrList(genome))
reg = "*"
bins = "*"
analysisDef = "Dummy: dummy name ([wStep=%g] [wSize=%s] [func=%s] [mds=%s] [mcT=%s] [mcR=%s])-> CategoryClusterSeparationStat" % (windowStep, windowSize, compare, mds, mcTreshold, mcRuns)
userBinSource = GalaxyInterface._getUserBinSource(reg, bins, genome)
result = GalaxyInterface.runManual([tn1, tn2], analysisDef, reg, bins, genome, galaxyFn=galaxyFn)
for key in result.getAllRegionKeys():
chrom = str(key).split(":")[0];
r = result[key];
if 'Result' not in r.keys():
print "skipping chr:", chrom, r;
continue;
r = r['Result'];
scores = r[0];
stddev = r[1];
for i in range(len(scores)):
if scores[i] != 0:
pos = i*windowStep;
if fileformat == "tabular":
outputFile.write("%s\t%s\t%s\t%s\n" % (str(chrom), pos, str(scores[i]), str(stddev[i])))
else:
text += "%s\t%s\t%s\t%s\n" % (str(chrom), pos, str(scores[i]), str(stddev[i]));
if fileformat == "html":
htmlfile.writeTextToFile(text);
print htmlfile.getLink("Result file");
print GalaxyInterface.getHtmlEndForRuns()
outputFile.close();
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
targetTrack = ExternalTrackManager.getPreProcessedTrackFromGalaxyTN(
genome,
choices.targetTrack,
printErrors=False,
printProgress=False)
refGSuite = getGSuiteFromGalaxyTN(choices.refTrackCollection)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisBins = GalaxyInterface._getUserBinSource(regSpec,
binSpec,
genome=genome)
results = TrackReportCommon.getOverlapResultsForTrackVsCollection(
genome, targetTrack, refGSuite, analysisBins=analysisBins)
processedResults = TrackReportCommon.processRawResults(results)
targetTrackTitle = prettyPrintTrackName(targetTrack)
title = 'Screening of track ' + targetTrackTitle
sortedProcessedResultsTupleList = sorted(
processedResults.iteritems(),
key=lambda x: x[1][STAT_LIST_INDEX[STAT_FACTOR_OBSERVED_VS_EXPECTED
]],
reverse=True)
refTrackNames = [x[0] for x in sortedProcessedResultsTupleList]
refTrackNames = [
x.replace('\'', '').replace('"', '') for x in refTrackNames
]
plotData = [x[1] for x in sortedProcessedResultsTupleList]
# plotData = zip(*plotData) #invert
plotData = normalizeMatrixData(plotData)
printVals = tuple([str(targetTrackTitle)]) + tuple(
[str(x[0]) for x in sortedProcessedResultsTupleList[0:3]])
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.header(title)
if choices.bmQid and choices.bmQid not in ['None']:
htmlCore.append(
str(
quick.gsuite.GSuiteHbIntegration.
getAnalysisQuestionInfoHtml(choices.bmQid)))
htmlCore.divBegin('resultsDiv')
htmlCore.paragraph('''
The query track <b>%s</b> overlaps most strongly (is most highly enriched) with the tracks <b>%s</b>, <b>%s</b> and <b>%s</b>
from the selected collection. See below for a full (ranked) table of overlap and enrichment.
''' % printVals)
htmlCore.paragraph('''
The coverage of the query track is %s bps.
''' % strWithNatLangFormatting(
TrackReportCommon.getQueryTrackCoverageFromRawOverlapResults(
results)))
htmlCore.tableHeader(TrackReportCommon.HEADER_ROW,
sortable=True,
tableId='resultsTable')
for refTrackName, refTrackResults in sortedProcessedResultsTupleList:
line = [refTrackName
] + [strWithNatLangFormatting(x) for x in refTrackResults]
htmlCore.tableLine(line)
htmlCore.tableFooter()
htmlCore.divEnd()
'''
addColumnPlotToHtmlCore(htmlCore,
refTrackNames,
TrackReportCommon.HEADER_ROW[1:],
'stat', 'Results plot (data is normalized for better visual comparison) ',
plotData, xAxisRotation = 315)
'''
addPlotToHtmlCore(
htmlCore,
refTrackNames,
TrackReportCommon.HEADER_ROW[1:],
'stat',
'Results plot (data is normalized for better visual comparison) ',
plotData,
xAxisRotation=315)
htmlCore.hideToggle(styleClass='debug')
htmlCore.end()
print htmlCore
