def getGenesForGenome(self, genome):
tnExon = ['Genes and gene subsets','Exons']
tnGenes =['Genes and gene subsets','Genes']
multiCountDef = 'dummy -> PointCountInsideSegsStat'
try:
genesList = [ v[0] for v in GalaxyInterface.getSubTrackNames(genome, tnGenes, deep=False)[0]]
exonList = [ v[0] for v in GalaxyInterface.getSubTrackNames(genome, tnExon, deep=False)[0]]
for tn in genesList:
self.analysisDict['Genes_CountPointStat_'+tn] = [tnGenes+[tn], multiCountDef, 'Result']
for tn in exonList:
self.analysisDict['Exons_CountPointStat_'+tn] = [tnExon+[tn], multiCountDef, 'Result']
except:
pass
def countSubTracks(genome, parentTrack, recursioncount):
if recursioncount < 10:
thissubs = GalaxyInterface.getSubTrackNames(genome, parentTrack,
False)[0]
tname = genome + ":" + ":".join(parentTrack)
#ti = TrackInfo.
if thissubs: # has subtrakcs, ie is a folder
thisndirs = 0
thisntracks = 0
for a in thissubs:
nextparenttrack = parentTrack
nextparenttrack.append(a[0])
ndirs, ntracks = countSubTracks(genome, nextparenttrack,
recursioncount + 1)
thisndirs = thisndirs + ndirs
thisntracks = thisntracks + ntracks
nextparenttrack = nextparenttrack.remove(a[0])
print('plan to update trackinfo', tname,
' set number of folders to ', thisndirs,
' and number of tracks to ', thisntracks)
return (thisndirs + 1, thisntracks)
else: # is a track.
return (0, 1)
else:
print "to many recursions"
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 getOptionsBox3(prevChoices):
'''Returns a list of options to be displayed in the second options box, which will be displayed after a selection is made in the first box.
prevChoices is a list of selections made by the web-user in the previous input boxes (that is, list containing only one element for this case)
'''
from quick.application.GalaxyInterface import GalaxyInterface
try:
subTrackList = [v[0] for v in GalaxyInterface.getSubTrackNames(prevChoices[0], prevChoices[1].split(':'), deep=False)[0]]
#return (repr(subTrackList), 4, True)
return ('\n'.join(subTrackList), len(subTrackList), True)
except:
pass
def getOptionsBox3(prevChoices):
'''Returns a list of options to be displayed in the second options box, which will be displayed after a selection is made in the first box.
prevChoices is a list of selections made by the web-user in the previous input boxes (that is, list containing only one element for this case)
'''
from quick.application.GalaxyInterface import GalaxyInterface
from collections import OrderedDict
try:
subTrackList = [
v[0] for v in GalaxyInterface.getSubTrackNames(
prevChoices[0], prevChoices[1].split(':'), deep=False)[0]
]
if len(subTrackList) > 0:
subSubTrackList = GalaxyInterface.getSubTrackNames(
prevChoices[0],
prevChoices[1].split(':') + [subTrackList[0]],
deep=False)[0]
if len(subSubTrackList) == 0:
return OrderedDict([(v, False) for v in subTrackList])
#return ('\n'.join(subTrackList), len(subTrackList), True)
except:
pass
def getSubTrackLeafTerms(genome,
parentTrack,
excludeSelfIfValid=True,
username=''):
from quick.application.GalaxyInterface import GalaxyInterface
subTrackTriples = GalaxyInterface.getSubTrackNames(genome,
parentTrack,
deep=False,
username=username)[0]
filteredSubTrackLeaves = [
x[0] for x in subTrackTriples if x not in [[], None]
and not (excludeSelfIfValid and x[0] == '-- All subtypes --')
]
return filteredSubTrackLeaves
def countSubTracks(genome, parentTrack, recursioncount):
if recursioncount < 10:
thissubs=GalaxyInterface.getSubTrackNames(genome, parentTrack, False)[0]
tname=genome+":"+":".join( parentTrack)
#ti = TrackInfo.
if thissubs: # has subtrakcs, ie is a folder
thisndirs=0
thisntracks=0
for a in thissubs:
nextparenttrack=parentTrack
nextparenttrack.append(a[0])
ndirs, ntracks =countSubTracks(genome, nextparenttrack, recursioncount+1)
thisndirs=thisndirs+ndirs
thisntracks=thisntracks+ntracks
nextparenttrack=nextparenttrack.remove(a[0])
print('plan to update trackinfo', tname,' set number of folders to ', thisndirs, ' and number of tracks to ', thisntracks)
return(thisndirs+1, thisntracks)
else:# is a track.
return (0,1)
else:
print "to many recursions"
def execute(choices, galaxyFn=None, username=''):
#setupDebugModeAndLogging()
from time import time
startTime = time()
print HtmlCore().begin()
print '<pre>'
genome = choices[0]
#assert genome=='hg19'
flankSize = choices[3]
if choices[1] == 'Prepared catalogues':
if choices[2] == 'GiulioNewGwas':
gwasTnBase = 'Private:GK:NewGwasBase'.split(':')
elif choices[2] == 'GiulioAllGwas':
gwasTnBase = 'Private:GK:AllGiulioGwasSnpsAsOf9feb13'.split(
':')
elif choices[2] == 'GiulioMay13Gwas':
gwasTnBase = 'Private:GK:Gwas:GiulioMay13'.split(':')
elif choices[2] == 'SmallTest':
gwasTnBase = 'Private:GK:Gwas'.split(':')
else:
raise
gwasTnBase += [flankSize]
elif choices[1] == 'Custom track':
gwasTnBase = choices[2].split(':')
assert flankSize == 'SNPs'
else:
assert False, choices[1]
referenceTrackSource = choices[4]
normalization = choices[5]
assert normalization == 'CoverageDepth'
analysisType = choices[6]
if analysisType == 'Enrichment':
ResultClass = EnrichmentGwasResults
elif analysisType == 'Testing':
ResultClass = HypothesisTestingGwasResults
nullmodelMapping = dict(
zip([
'Sample disease regions uniformly',
'Sample disease regions with preserved inter-region spacings',
'Sample disease regions with preserved distance to nearest exon'
], [
'PermutedSegsAndSampledIntersegsTrack_',
'PermutedSegsAndIntersegsTrack_',
'SegsSampledByDistanceToReferenceTrack_,trackNameIntensity=Genes and gene subsets^Exons^Ensembl exons'
]))
nullmodel = nullmodelMapping[choices[9]]
assert nullmodel in [
'PermutedSegsAndSampledIntersegsTrack_',
'PermutedSegsAndIntersegsTrack_',
'SegsSampledByDistanceToReferenceTrack_,trackNameIntensity=Genes and gene subsets^Exons^Ensembl exons'
]
else:
raise
kernelType = choices[7]
kernelParam = choices[8]
if choices[10] == 'Include links to full underlying results':
includeDetailedResults = True
elif choices[10] == 'Only produce main result values':
includeDetailedResults = False
else:
raise InvalidRunSpecException('Did not understand option: %s' %
choices[12])
mcDepth = choices[11]
if choices[12] == 'yes':
includeLocalResults = True
elif choices[12] == 'no':
includeLocalResults = False
else:
raise InvalidRunSpecException('Did not understand option: %s' %
choices[12])
if choices[15] == 'yes':
useCache = True
elif choices[15] == 'no':
useCache = False
else:
raise InvalidRunSpecException('Did not understand option: %s' %
choices[15])
if choices[16] == 'yes':
printProgress = True
elif choices[16] == 'no':
printProgress = False
else:
raise InvalidRunSpecException('Did not understand option: %s' %
choices[16])
from quick.application.GalaxyInterface import GalaxyInterface
#print GalaxyInterface.getHtmlForToggles()
#print GalaxyInterface.getHtmlBeginForRuns()
#from quick.webtools.GwasAPI import getEnrichmentValues
print 'Progress: '
#print 'base: ',gwasTnBase
#print 'leaves: ',GalaxyInterface.getSubTrackNames(genome, gwasTnBase,deep=False, username=username)
disRes = MultiGwasResults()
from gold.application.HyperBrowserCLI import getSubTrackLeafTerms
from quick.application.ProcTrackOptions import ProcTrackOptions
#for gwasTrackLeaf in GalaxyInterface.getSubTrackNames(genome, gwasTnBase,deep=False, username=username):
allDiseases = getSubTrackLeafTerms(genome,
gwasTnBase,
username=username)
if len(allDiseases) == 0:
assert ProcTrackOptions.isValidTrack(
genome, gwasTnBase, GalaxyInterface.userHasFullAccess(
username)), 'Genome: %s, TN: %s, Access: %s' % (
genome, gwasTnBase,
GalaxyInterface.userHasFullAccess(username))
allDiseases = gwasTnBase[-1:]
gwasTnBase = gwasTnBase[:-1]
for disease in allDiseases:
#print 'Leaf:',gwasTrackLeaf[0]
#if not gwasTrackLeaf[0] in ['11 - Height.txt']:
#if not disease in ['1 - Alzheimer.txt','10 - Graves.txt']:#['Malaria','UC']:
# print 'IGNORING: ', gwasTrackLeaf[0]
# continue
#if gwasTrackLeaf in [[],None] or gwasTrackLeaf[0]=='-- All subtypes --':
#continue
#gwasTn = ':'.join(gwasTnBase + [gwasTrackLeaf[0]])
gwasTn = ':'.join(gwasTnBase + [disease])
#print 'Running API: ', "$getEnrichmentValues(%s, '%s', '%s')" % ([gwasTn], referenceTrackSource, normalization)
#enrichmentsDict = getEnrichmentValues([gwasTn], referenceTrackSource, normalization)#, ['114 - Brain_Mid_Frontal_Lobe.txt','134 - Rectal_Smooth_Muscle.txt'])
#assert len(enrichmentsDict.values())==1
#enrichments = enrichmentsDict.values()[0]
#if gwasTrackLeaf[0] in ['Malaria','UC']:
#print 'HERE IS WHAT I GOT: ',enrichmentsDict
#print 'ENR: ',enrichments
#print 'One: ', (enrichments.values()[0])['enrichment']['13 - CD4'].getGlobalResult()
#assert 'enrichment' in (enrichments.values()[0]), (enrichments.values()[0])
#disRes[gwasTrackLeaf[0]] = (enrichments.values()[0])['enrichment']
#disRes[gwasTrackLeaf[0]] = (enrichments.values()[0])
#disease = gwasTrackLeaf[0]
#disRes[disease] = [x.getGlobalResult() for x in enrichments]
#print 'DISres: ', disRes[gwasTrackLeaf[0]]
#from quick.util.CommonFunctions import extractIdFromGalaxyFn
res = ResultClass(gwasId=disease, verbose=True, galaxyFn=galaxyFn)
#referenceSubTypes = enrichments.keys()
#referenceSubTypes = [x[0] for x in GalaxyInterface.getSubTrackNames(genome, 'Private:GK:Psych:DHSs'.split(':'), deep=False, username=username) if not x[0] == '-- All subtypes --']
if referenceTrackSource == 'H3K4me3':
refTrackBase = 'Private:GK:Psych:H3K4me3'
refTrackCoverageFunction = 'Private^GK^Psych^H3K4me3CoverageTrack'
elif referenceTrackSource == 'DHS':
refTrackBase = 'Private:GK:Psych:DHSs'
refTrackCoverageFunction = 'Private^GK^Psych^DHSCoverageTrack'
elif referenceTrackSource == 'Chromatin state 1-AP':
refTrackBase = 'Private:Anders:Chromatin State Segmentation:1_Active_Promoter'
refTrackCoverageFunction = 'Private^GWAS^Chromatin^CoverageFunctionTracks^1_Active_PromoterV2'
elif referenceTrackSource == 'Chromatin state 4-SE':
refTrackBase = 'Private:Anders:Chromatin State Segmentation:4_Strong_Enhancer'
refTrackCoverageFunction = 'Private^GWAS^Chromatin^CoverageFunctionTracks^4_Strong_Enhancer'
elif referenceTrackSource == 'Chromatin state 5-SE':
refTrackBase = 'Private:Anders:Chromatin State Segmentation:5_Strong_Enhancer'
refTrackCoverageFunction = 'Private^GWAS^Chromatin^CoverageFunctionTracks^5_Strong_Enhancer'
else:
raise
refTrackSelectType = choices[13]
allReferenceTracks = [
x[0] for x in GalaxyInterface.getSubTrackNames(
genome,
refTrackBase.split(':'),
deep=False,
username=username) if not x[0] == '-- All subtypes --'
]
if refTrackSelectType == 'Use all reference tracks':
referenceSubTypes = allReferenceTracks
elif refTrackSelectType == 'Select single reference track':
referenceSubTypes = [choices[14]]
assert referenceSubTypes[0] in allReferenceTracks
elif refTrackSelectType == 'Select a range among all reference tracks':
try:
firstRefTrack, lastRefTrack = choices[14].split('-')
referenceSubTypes = allReferenceTracks[
int(firstRefTrack):int(lastRefTrack) + 1]
print 'Analyzing %s among a total of %s reference tracks' % (
choices[14], len(allReferenceTracks))
except Exception:
print 'Range format should be e.g. "15-18".'
raise
else:
raise
for referenceSubType in referenceSubTypes:
#if not referenceSubType in ['107 - Adult_Kidney.txt','106 - Adipose_Nuclei.txt']:
# #print 'IGNORING: ',referenceSubType
# continue
#
if analysisType == 'Enrichment':
res[referenceSubType] = directGetEnrichment(
gwasTn, referenceSubType, refTrackBase, kernelType,
kernelParam, useCache, printProgress)
elif analysisType == 'Testing':
res[referenceSubType] = directGetTestResults(
gwasTn, referenceSubType, refTrackBase, kernelType,
kernelParam, refTrackCoverageFunction, nullmodel,
mcDepth, useCache, printProgress)
else:
raise
#print disease, referenceSubType, res[referenceSubType]
#print "ENR: ",enrichments
#res[referenceSubType] = enrichments[referenceSubType]
disRes[disease] = res
#for disease in disRes:
# print 'D FULL %s:' %disease, disRes[disease]
# print 'D DICTS %s:'%disease, disRes[disease].getAllGlobalResultDicts()
# print 'DISEASE %s:'%disease, disRes[disease].getAllGlobalResults()
print 'Total run time (excluding figure generation): %i seconds.' % (
time() - startTime)
print '</pre>'
#print GalaxyInterface.getHtmlBeginForRuns()
print '<h1>Results</h1>'
if len(allDiseases) > 1:
try:
heatMapLink = disRes.getLinkToClusteredHeatmap(
'Heatmap', galaxyFn)
print '<h3>Heatmap</h3>', heatMapLink #, '<br>'
except:
print '<p>Creation of heatmap failed</p>'
tableOutput = disRes.getHtmlResultsTable(includeDetailedResults)
print '<h3>Results table</h3>', tableOutput
if choices[-1]:
print '<h3>Prior coloring table</h3>'
colorFn = ExternalTrackManager.extractFnFromGalaxyTN(
choices[-1].split(':'))
print disRes.getColoredSortedReferencesTable(colorFn)
if includeLocalResults:
print '<h3>Local results</h3>'
print disRes.getLinksToAllLocalHtmlResultsTables(galaxyFn)
