def generateQ1output(cls, additionalResultsDict, analysisQuestion, choices, galaxyFn,
gsPerTrackResults, queryTrackTitle, gsuite, results,
similarityStatClassName):
core = HtmlCore()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divClass='results-section')
core.header(analysisQuestion)
topTrackTitle = results.keys()[0]
core.paragraph('''
The track "%s" in the GSuite is the one most similar to the query track %s, with a similarity score of %s
as measured by the "%s" track similarity measure.
''' % (
topTrackTitle, queryTrackTitle, strWithNatLangFormatting(results[topTrackTitle]),
similarityStatClassName))
core.divBegin()
addTableWithTabularAndGsuiteImportButtons(
core, choices, galaxyFn, cls.Q1_SHORT, tableDict=gsPerTrackResults[1],
columnNames=gsPerTrackResults[0], gsuite=gsuite, results=results,
gsuiteAppendAttrs=['similarity_score'], sortable=True)
core.divEnd()
columnInd = 0
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
columnInd = 1
res = GSuiteTracksCoincidingWithQueryTrackTool.drawPlot(
results, additionalResultsDict,
'Similarity to query track', columnInd=columnInd)
core.line(res)
core.divEnd()
core.divEnd()
core.end()
return core
def execute(choices, galaxyFn=None, username=''):
gsuite = getGSuiteFromGalaxyTN(choices.gsuite)
bedRegions = choices.bedRegions
rp = RP(gsuite)
bedData = rp.openBedFile(bedRegions)
chrOrder, chrLength = rp.sortChrDict()
dataDict, dataDictLine, elementOrder, listResCopy, newDictRegions = rp.countMutations(
chrLength, bedData)
vg = visualizationGraphs()
tName = gsuite.allTrackTitles()
uniformDictList = OrderedDict() # expected values
observedDictList = OrderedDict() # observed values
seriesNameRegionUDL = OrderedDict()
seriesNameRegionODL = OrderedDict()
seriesNameRegion = OrderedDict()
GenerateDistributionOfPointsOfInterestTool.countRegionsForDistribution(
newDictRegions, observedDictList, rp, seriesNameRegion,
seriesNameRegionODL, seriesNameRegionUDL, tName, uniformDictList)
res = ''
for elK in uniformDictList.keys():
res += GenerateDistributionOfPointsOfInterestTool.drawDistribution(
elK, observedDictList, seriesNameRegion, seriesNameRegionODL,
seriesNameRegionUDL, uniformDictList, vg)
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.line(res)
htmlCore.end()
htmlCore.hideToggle(styleClass='debug')
print htmlCore
def generateQ2Output(cls, additionalAttributesDict, additionalResultsDict, analysisQuestion, choices,
galaxyFn, queryTrackTitle, gsuite, results, similarityStatClassName):
gsPerTrackResultsModel = GSuitePerTrackResultModel(results, ['Similarity to query track', 'P-value'],
additionalResultsDict=additionalResultsDict,
additionalAttributesDict=additionalAttributesDict)
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
gsPerTrackResults = gsPerTrackResultsModel.generateColumnTitlesAndResultsDict(choices.leadAttribute)
else:
gsPerTrackResults = gsPerTrackResultsModel.generateColumnTitlesAndResultsDict()
core = HtmlCore()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divClass='results-section')
core.header(analysisQuestion)
topTrackTitle = results.keys()[0]
core.paragraph('''
The track "%s" has the lowest P-value of %s corresponding to %s similarity to the query track "%s"
as measured by "%s" track similarity measure.
''' % (topTrackTitle, strWithNatLangFormatting(results[topTrackTitle][1]),
strWithNatLangFormatting(results[topTrackTitle][0]), queryTrackTitle, similarityStatClassName))
addTableWithTabularAndGsuiteImportButtons(
core, choices, galaxyFn, cls.Q2_SHORT, tableDict=gsPerTrackResults[1],
columnNames=gsPerTrackResults[0], gsuite=gsuite, results=results,
gsuiteAppendAttrs=['similarity_score', 'p_value'], sortable=True)
columnInd = 0
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
columnInd = 1
resultsSeparateListPart = OrderedDict()
additionalResultsDictIncludePartFromResults = OrderedDict()
for k, v in results.iteritems():
if k not in resultsSeparateListPart.keys():
resultsSeparateListPart[k] = v[0]
if k not in additionalResultsDictIncludePartFromResults.keys():
additionalResultsDictIncludePartFromResults[k] = OrderedDict()
additionalResultsDictIncludePartFromResults[k]['P-Value'] = v[1]
for k1, v1 in additionalResultsDict[k].iteritems():
additionalResultsDictIncludePartFromResults[k][k1] = v1
res = GSuiteTracksCoincidingWithQueryTrackTool.drawPlot(
resultsSeparateListPart, additionalResultsDictIncludePartFromResults,
'Similarity to query track', columnInd=columnInd)
core.line(res)
core.divEnd()
core.divEnd()
core.end()
return core
def execute(cls, choices, galaxyFn=None, username=''):
#gsuite
gsuite = getGSuiteFromGalaxyTN(choices.gsuite)
#all boxes
multiPlot = choices.multiPlot
#scale = choices.scale
#overlap = choices.overlap
overlap = 'no'
bps = int(choices.bps)
rp = RP(gsuite)
#get length of chromosomes and ordering
chrItems = GenomeInfo.getStdChrLengthDict(gsuite.genome)
chrOrder, chrLength = GenerateRainfallPlotTool.sortChrDict(chrItems)
dataDict, dataDictLine, elementOrder, listResCopy, listDataCountPerBin, newResBinSizeListSum, chrList = GenerateRainfallPlotTool.countMutations(
gsuite, chrLength, bps)
seriesType, newSeriesNameRes, newSeriesNameResOE, yAxisMultiVal = rp.getOptionsForPlot(
elementOrder, gsuite.allTrackTitles())
newResList, newResBinSizeList, newResBinSizeListSortedList = rp.generateBinSizeList(
elementOrder, listResCopy, listDataCountPerBin,
newResBinSizeListSum)
vg = visualizationGraphs()
res = ''
if multiPlot == 'Single':
res += GenerateRainfallPlotTool.drawSinglePlot(
vg, newResBinSizeListSortedList, chrLength, newResList,
newSeriesNameRes, newResBinSizeList, overlap, seriesType,
yAxisMultiVal)
else:
res += GenerateRainfallPlotTool.drawMultiPlot(
newResList, newSeriesNameRes, newResBinSizeList, vg,
seriesType, yAxisMultiVal)
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.line('Bin size: ' + str(bps))
htmlCore.line(res)
htmlCore.end()
htmlCore.hideToggle(styleClass='debug')
print htmlCore
def getInfoForOptionsBoxOperation(cls, prevChoices):
'''
If not None, defines the string content of an clickable info box beside
the corresponding input box. HTML is allowed.
'''
if prevChoices.operation and prevChoices.operation != cls.NO_OPERATION_TEXT:
from quick.extra.StandardizeTrackFiles import getParserClassDocString
from proto.hyperbrowser.HtmlCore import HtmlCore
docString = getParserClassDocString(cls.ALL_OPERATIONS[prevChoices.operation])
core = HtmlCore()
for line in docString.split(os.linesep):
core.line(line)
return str(core)
def getToolDescription():
htmlCore = HtmlCore()
htmlCore.paragraph(
'The tool is used to generate density of distribution.')
htmlCore.divider()
htmlCore.paragraph('The input for tool is following:')
htmlCore.line('- GSuite')
htmlCore.line(
'- file with regions (bed format), which should be given by user')
htmlCore.divider()
htmlCore.paragraph('The output for tool is a plot.')
return str(htmlCore)
def _buildContent(self):
#iterate through dictionary and for each key create a section (one of [GSuite, Track...])
# each value in the dictionary is a list of GiudeData objects that go into the section defined by the key
htmlCore = HtmlCore()
htmlCore.divBegin('toolGuideInfo')
htmlCore.divBegin(divClass='toolGuideInfoText')
htmlCore.divBegin(divClass='toolGuideInfoTextHeader')
htmlCore.line(TOOL_GUIDE_HELP_HEADER_TEXT)
htmlCore.divEnd()
htmlCore.divBegin(divClass='toolGuideInfoText')
htmlCore.line(TOOL_GUIDE_HELP_HEADER_TEXT_TEXT)
htmlCore.divEnd()
htmlCore.divEnd()
for guideDataKey, guideDataValues in self._guideDataDict.iteritems():
htmlCore.divBegin('toolGuide')
if guideDataKey in TOOL_INPUT_TYPE_TO_TOOL_GUIDE_HELP_HEADER_DICT:
htmlCore.header(TOOL_INPUT_TYPE_TO_TOOL_GUIDE_HELP_HEADER_DICT[
guideDataKey])
for guideDataValue in guideDataValues:
htmlCore.divBegin(divClass='toolGuideData')
htmlCore.divBegin(divClass='toolGuideImgTitle')
if guideDataValue.imgUrl:
htmlCore.image(guideDataValue.imgUrl)
htmlCore.link(text=guideDataValue.toolDisplayName,
url=str(guideDataValue.toolUrl),
args=(' onclick="%s"' % guideDataValue.onclick)
if guideDataValue.onclick else '')
htmlCore.divEnd()
htmlCore.divBegin(divClass='toolGuideDesc')
htmlCore.append(guideDataValue.description)
if guideDataValue.helpPageUrl:
htmlCore.link(text='...read more',
url=str(guideDataValue.helpPageUrl))
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.divEnd()
#raise Exception(str(htmlCore))#to debug
self._guideContent = str(htmlCore)
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
def getToolDescription():
htmlCore = HtmlCore()
htmlCore.paragraph('This tool provides the possibility to generate synthetic dataset with Poisson distribution.')
htmlCore.divider()
htmlCore.paragraph('The input for tool is following:')
htmlCore.line('- genome, which you can select from the given options')
htmlCore.line('- file with parameters (gtrack format), which should be given by user')
htmlCore.paragraph('File with parameters should include information about:')
htmlCore.line('- chromosome')
htmlCore.line('- start position')
htmlCore.line('- end position')
htmlCore.line('- inter-events distance')
htmlCore.line('- intra-events distance')
htmlCore.line('- probability value')
htmlCore.paragraph('The example of file with parameters:')
htmlCore.paragraph('''
##Track type: points <br \>
###seqid start end inter intra prob <br \>
####genome=hg19 <br \>
chr1 0 100000 0.0001 0 1
''')
htmlCore.line(', where first three lines are a header, the fourth line contains the information about'
'region (chromosome, start, end position) and values ' \
'(inter-, intra-mutations, probability) for which will be calculated synthetic dataset.')
htmlCore.divider()
htmlCore.line('IMPORTANT INFORMATION')
htmlCore.line('The file can contains more than one line with parameters, but the calculated simulated datasets ' \
'for every regions are merged together at the end.')
htmlCore.divider()
htmlCore.paragraph('The output for tool is a GSuite containing one simulated dataset.')
htmlCore.divider()
htmlCore.line('IMPORTANT OTHER TOOLS')
htmlCore.line('To upload your own file with parameters (available later as an element in the history) use the tool called:' \
' Upload file ')
return str(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.
'''
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(cls, choices, galaxyFn=None, username=''):
#cls._setDebugModeIfSelected(choices)
# from config.DebugConfig import DebugConfig
# from config.DebugConfig import DebugModes
# DebugConfig.changeMode(DebugModes.RAISE_HIDDEN_EXCEPTIONS_NO_VERBOSE)
# DebugUtil.insertBreakPoint(5678, suspend=False)
choices_gsuite = choices.gsuite
selected_metadata = choices.cat
choices_queryTrack = choices.query
#genome = 'hg19'
genome = choices.genome
queryTS = factory.getSingleTrackTS(genome, choices_queryTrack)
refTS = factory.getFlatTracksTS(genome, choices_gsuite)
categoricalTS = refTS.getSplittedByCategoryTS(selected_metadata)
fullTS = TrackStructureV2()
fullTS['query'] = queryTS
fullTS['reference'] = categoricalTS
spec = AnalysisSpec(SummarizedInteractionPerTsCatV2Stat)
parameter = 'minLqMedUqMax'
spec.addParameter('pairwiseStatistic', ObservedVsExpectedStat.__name__)
spec.addParameter('summaryFunc', parameter)
bins = UserBinSource('chr1', '*', genome=genome)
res = doAnalysis(spec, bins, fullTS)
tsRes = res.getGlobalResult()['Result']
htmlCore = HtmlCore()
htmlCore.begin()
if parameter == 'minAndMax':
htmlCore.tableHeader(['Track', 'min-max'],
sortable=False,
tableId='tab1')
for k, it in tsRes.iteritems():
htmlCore.tableLine([
k,
str("%.2f" % it.getResult()[0]) + '-' +
str("%.2f" % it.getResult()[1])
])
htmlCore.tableFooter()
if parameter == 'minLqMedUqMax':
dataList = []
categories = []
for keyE, itE in tsRes.iteritems():
categories.append(keyE)
dataList.append(list(itE.getResult()))
from quick.webtools.restricted.visualization.visualizationGraphs import \
visualizationGraphs
vg = visualizationGraphs()
res = vg.drawBoxPlotChart(dataList,
categories=categories,
seriesName=selected_metadata)
htmlCore.line(res)
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.
'''
cls._setDebugModeIfSelected(choices)
genome = choices.genome
queryGSuite = getGSuiteFromGalaxyTN(choices.queryGSuite)
refGSuite = getGSuiteFromGalaxyTN(choices.refGSuite)
if choices.similarityFunc:
similarityStatClassNameKey = choices.similarityFunc
else:
similarityStatClassNameKey = GSuiteStatUtils.T5_RATIO_OF_OBSERVED_TO_EXPECTED_OVERLAP
isPointsVsSegments, pointsGSuite, segGSuite = cls.isPointsVsSegmentsAnalysis(queryGSuite, refGSuite)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisBins = GalaxyInterface._getUserBinSource(regSpec, binSpec, genome=genome)
queryTrackList = [Track(x.trackName, x.title) for x in queryGSuite.allTracks()]
refTrackList = [Track(x.trackName, x.title) for x in refGSuite.allTracks()]
queryTrackTitles = CommonConstants.TRACK_TITLES_SEPARATOR.join(
[quote(x.title, safe='') for x in queryGSuite.allTracks()])
refTrackTitles = CommonConstants.TRACK_TITLES_SEPARATOR.join(
[quote(x.title, safe='') for x in refGSuite.allTracks()])
analysisSpec = AnalysisSpec(GSuiteVsGSuiteWrapperStat)
analysisSpec.addParameter('queryTracksNum', str(len(queryTrackList)))
analysisSpec.addParameter('refTracksNum', str(len(refTrackList)))
analysisSpec.addParameter('queryTrackTitleList', queryTrackTitles)
analysisSpec.addParameter('refTrackTitleList', refTrackTitles)
analysisSpec.addParameter('similarityStatClassName',
GSuiteStatUtils.PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similarityStatClassNameKey])
if choices.removeZeroRow:
analysisSpec.addParameter('removeZeroRow', choices.removeZeroRow)
if choices.removeZeroCol:
analysisSpec.addParameter('removeZeroColumn', choices.removeZeroCol)
resultsObj = doAnalysis(analysisSpec, analysisBins, queryTrackList + refTrackList)
results = resultsObj.getGlobalResult()
# baseDir = GalaxyRunSpecificFile([RAW_OVERLAP_TABLE_RESULT_KEY], galaxyFn).getDiskPath()
# rawOverlapHeatmapPresenter = HeatmapFromDictOfDictsPresenter(resultsObj, baseDir,
# 'Overlapping base-pair of tracks from the two suites',
# printDimensions=False)
rawOverlapTableData = results[RAW_OVERLAP_TABLE_RESULT_KEY]
maxRawOverlap, maxROt1, maxROt2 = rawOverlapTableData.getMaxElement()
similarityScoreTableData = results[SIMILARITY_SCORE_TABLE_RESULT_KEY]
maxSimScore, maxSSt1, maxSSt2 = similarityScoreTableData.getMaxElement()
baseDir = GalaxyRunSpecificFile([], galaxyFn=galaxyFn).getDiskPath()
heatmapPresenter = HeatmapFromTableDataPresenter(resultsObj, baseDir=baseDir,
header='Overlapping base-pairs between the tracks of the two suites',
printDimensions=False)
tablePresenter = MatrixGlobalValueFromTableDataPresenter(resultsObj, baseDir=baseDir,
header='Table of overlapping base-pairs between the tracks of the two suites')
core = HtmlCore()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divId='svs-res-main-div', divClass='svs-res-main')
core.divBegin(divId='raw-overlap-div', divClass='results-section')
core.divBegin(divId='raw-overlap-table', divClass='svs-table-div')
core.header('Base-pair overlaps between the tracks of the two GSuites')
core.paragraph("""From the tracks in the two GSuites the highest base-pair overlap <b>(%s bps)</b>
is observed for the pair of <b>'%s'</b> and <b>'%s'</b>.""" % (maxRawOverlap, maxROt1, maxROt2))
core.divBegin(divId='raw-table-result', divClass='result-div')
core.divBegin(divId='raw-table-result', divClass='result-div-left')
core.line('''Follow the links to view the results in an HTML table
or raw tabular form:''')
core.divEnd()
core.divBegin(divId='raw-table-result', divClass='result-div-right')
core.line(tablePresenter.getReference(RAW_OVERLAP_TABLE_RESULT_KEY))
core.divEnd()#rawoverlap table
core.divEnd()
core.divEnd()
core.divBegin(divId='raw-overlap-heatmap', divClass='svs-heatmap-div')
try:
core.header('Heatmap of base-pair overlaps')
core.divBegin(divId='raw-table-result', divClass='result-div-heatmap')
core.divBegin(divId='raw-table-result', divClass='result-div-left')
core.line('''Follow the links to view the heatmap in the desired format:''')
core.divEnd()
core.divBegin(divId='raw-table-result', divClass='result-div-right')
core.line(heatmapPresenter.getReference(RAW_OVERLAP_TABLE_RESULT_KEY))
core.divEnd()
core.divEnd()
except:
core.line('Heatmap for the base-pair overlaps could not be created.')
core.divEnd()
core.divEnd()
core.divEnd()#rawoverlap heatmap
core.divEnd()#rawoverlap
core.divBegin(divId='sim-score-div', divClass='results-section')
core.divBegin(divId='sim-score-table', divClass='svs-table-div')
core.header('Similarity score between the tracks of the two GSuites measured by %s' % choices.similarityFunc)
core.paragraph("""From the tracks in the two GSuites the highest similarity score <b>(%s)</b>
is observed for the pair of <b>'%s'</b> and <b>'%s'</b>.""" % (maxSimScore, maxSSt1, maxSSt2))
core.divBegin(divId='raw-table-result', divClass='result-div')
core.divBegin(divId='raw-table-result', divClass='result-div-left')
core.line("""Follow the links to view the results in an HTML table or raw tabular form:""")
core.divEnd()
core.divBegin(divId='raw-table-result', divClass='result-div-right')
core.line(tablePresenter.getReference(SIMILARITY_SCORE_TABLE_RESULT_KEY))
core.divEnd()
core.divEnd()
core.divEnd()#simscore table
core.divBegin(divId='sim-score-heatmap', divClass='svs-heatmap-div')
try:
core.header('Heatmap of similarity scores')
core.divBegin(divId='raw-table-result', divClass='result-div-heatmap')
core.divBegin(divId='raw-table-result', divClass='result-div-left')
core.line('''Follow the links to view the heatmap in the desired format:''')
core.divEnd()
core.divBegin(divId='raw-table-result', divClass='result-div-right')
core.line(heatmapPresenter.getReference(SIMILARITY_SCORE_TABLE_RESULT_KEY))
core.divEnd()
core.divEnd()
except:
core.line('Heatmap for the similarity score could not be created.')
core.divEnd()
core.divEnd()
core.divEnd()#simscore heatmap
core.divEnd()#simscore
core.divEnd()#results
# core.paragraph(
# '''Table displaying the number of base-pairs overlapping between the tracks in the two suites:''')
# core.tableFromDictOfDicts(rawOverlapTableData, firstColName='Track title')
# # core.paragraph(rawOverlapHeatmapPresenter.getReference(resDictKey=RAW_OVERLAP_TABLE_RESULT_KEY))
# core.paragraph(
# '''Table displaying the similarity score for the tracks in the two suites as measured by %s:''' % similarityStatClassNameKey)
# core.tableFromDictOfDicts(similarityScoreTableData, firstColName='Track title')
#
core.divEnd()
core.end()
print str(core)
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 numpy
numpy.seterr(all='raise')
cls._setDebugModeIfSelected(choices)
# DebugUtil.insertBreakPoint(username=username, currentUser='******')
genome = choices.genome
analysisQuestion = choices.analysisName
similaryStatClassName = 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'
gsuite = getGSuiteFromGalaxyTN(choices.gsuite)
regSpec, binSpec = UserBinMixin.getRegsAndBinsSpec(choices)
analysisBins = GalaxyInterface._getUserBinSource(regSpec,
binSpec,
genome=genome)
tracks = [
Track(x.trackName, trackTitle=x.title) for x in gsuite.allTracks()
]
trackTitles = CommonConstants.TRACK_TITLES_SEPARATOR.join(
[quote(x.title, safe='') for x in gsuite.allTracks()])
additionalResultsDict = OrderedDict()
additionalAttributesDict = OrderedDict()
if analysisQuestion in [cls.Q1, cls.Q2, cls.Q3]:
additionalAttributesDict = cls.getSelectedAttributesForEachTrackDict(
choices.additionalAttributes, gsuite)
#additional analysis
stats = [CountStat, CountElementStat]
additionalResultsDict = runMultipleSingleValStatsOnTracks(
gsuite, stats, analysisBins, queryTrack=None)
if analysisQuestion == cls.Q1:
analysisSpec = AnalysisSpec(
GSuiteRepresentativenessOfTracksRankingsWrapperStat)
analysisSpec.addParameter(
'pairwiseStatistic', GSuiteStatUtils.
PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similaryStatClassName])
analysisSpec.addParameter(
'summaryFunc',
GSuiteStatUtils.SUMMARY_FUNCTIONS_MAPPER[summaryFunc])
analysisSpec.addParameter('reverse', reverse)
analysisSpec.addParameter('ascending', 'No')
analysisSpec.addParameter('trackTitles', trackTitles)
analysisSpec.addParameter('queryTracksNum', len(tracks))
results = doAnalysis(analysisSpec, analysisBins,
tracks).getGlobalResult()
gsPerTrackResultsModel = GSuitePerTrackResultModel(
results,
['Similarity to rest of tracks in suite (%s)' % summaryFunc],
additionalResultsDict=additionalResultsDict,
additionalAttributesDict=additionalAttributesDict)
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
columnTitles, decoratedResultsDict = \
gsPerTrackResultsModel.generateColumnTitlesAndResultsDict(choices.leadAttribute)
else:
columnTitles, decoratedResultsDict = \
gsPerTrackResultsModel.generateColumnTitlesAndResultsDict()
core = HtmlCore()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divClass='results-section')
core.header(analysisQuestion)
topTrackTitle = results.keys()[0]
core.paragraph('''
The track "%s" is the most representative track of the GSuite with %s %s similarity to the rest of the tracks
as measured by "%s" track similarity measure.
''' % (topTrackTitle, results[topTrackTitle], summaryFunc,
similaryStatClassName))
addTableWithTabularAndGsuiteImportButtons(
core,
choices,
galaxyFn,
cls.Q1_SHORT,
decoratedResultsDict,
columnTitles,
gsuite=gsuite,
results=results,
gsuiteAppendAttrs=['similarity_score'],
sortable=True)
# plot
columnInd = 0
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
columnInd = 1
res = GSuiteTracksCoincidingWithQueryTrackTool.drawPlot(
results,
additionalResultsDict,
'Similarity to rest of tracks in suite (%s)' % summaryFunc,
columnInd=columnInd)
core.line(res)
core.divEnd()
core.divEnd()
core.end()
# elif analysisQuestion == cls.Q2:
# analysisSpec = AnalysisSpec(GSuiteRepresentativenessOfTracksRankingsWrapperStat)
# analysisSpec.addParameter('pairwiseStatistic', GSuiteStatUtils.PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similaryStatClassName])
# analysisSpec.addParameter('summaryFunc', GSuiteStatUtils.SUMMARY_FUNCTIONS_MAPPER[summaryFunc])
# analysisSpec.addParameter('reverse', reverse)
# analysisSpec.addParameter('ascending', 'Yes')
# analysisSpec.addParameter('trackTitles', trackTitles)
# results = doAnalysis(analysisSpec, analysisBins, tracks).getGlobalResult()
#
# gsPerTrackResultsModel = GSuitePerTrackResultModel(
# results, ['Similarity to rest of tracks in suite (%s)' % summaryFunc],
# additionalResultsDict=additionalResultsDict,
# additionalAttributesDict=additionalAttributesDict)
# if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
# columnTitles, decoratedResultsDict = \
# gsPerTrackResultsModel.generateColumnTitlesAndResultsDict(choices.leadAttribute)
# else:
# columnTitles, decoratedResultsDict = \
# gsPerTrackResultsModel.generateColumnTitlesAndResultsDict()
#
# core = HtmlCore()
# core.begin()
# core.divBegin(divId='results-page')
# core.divBegin(divClass='results-section')
# core.header(analysisQuestion)
# topTrackTitle = results.keys()[0]
# core.paragraph('''
# The track "%s" is the most atypical track of the GSuite with %s %s similarity to the rest of the tracks
# as measured by the "%s" track similarity measure.
# ''' % (topTrackTitle, strWithNatLangFormatting(results[topTrackTitle]), summaryFunc, similaryStatClassName))
# # core.tableFromDictionary(results, columnNames=['Track title', 'Similarity to rest of tracks in suite (' + summaryFunc+')'], sortable=False)
#
# from quick.util import CommonFunctions
# rawDataURIList = CommonFunctions.getHyperlinksForRawTableData(
# dataDict=decoratedResultsDict, colNames=columnTitles,
# tableId="resultsTable", galaxyFn=galaxyFn)
# core.tableFromDictionary(decoratedResultsDict, columnNames=columnTitles, sortable=True,
# tableId='resultsTable', addInstruction=True,
# addRawDataSelectBox=True, rawDataURIList=rawDataURIList)
# # core.tableFromDictionary(decoratedResultsDict, columnNames=columnTitles, sortable=True, tableId='resultsTable')
#
# columnInd = 0
# if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
# columnInd = 1
# res = GSuiteTracksCoincidingWithQueryTrackTool.drawPlot(
# results, additionalResultsDict,
# 'Similarity to rest of tracks in suite (%s)' % summaryFunc,
# columnInd=columnInd)
# core.line(res)
# core.divEnd()
# core.divEnd()
# core.end()
#
# if choices.addResults == 'Yes':
# GSuiteStatUtils.addResultsToInputGSuite(
# gsuite, results, ['Similarity_score'],
# cls.extraGalaxyFn[GSUITE_EXPANDED_WITH_RESULT_COLUMNS_FILENAME])
elif analysisQuestion == cls.Q3:
mcfdrDepth = choices.mcfdrDepth if choices.mcfdrDepth else \
AnalysisDefHandler(REPLACE_TEMPLATES['$MCFDR$']).getOptionsAsText().values()[0][0]
analysisDefString = REPLACE_TEMPLATES[
'$MCFDRv3$'] + ' -> GSuiteRepresentativenessOfTracksRankingsAndPValuesWrapperStat'
analysisSpec = AnalysisDefHandler(analysisDefString)
analysisSpec.setChoice('MCFDR sampling depth', mcfdrDepth)
analysisSpec.addParameter('assumptions',
'PermutedSegsAndIntersegsTrack')
analysisSpec.addParameter(
'rawStatistic',
SummarizedInteractionWithOtherTracksV2Stat.__name__)
analysisSpec.addParameter(
'pairwiseStatistic', GSuiteStatUtils.
PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similaryStatClassName])
analysisSpec.addParameter(
'summaryFunc',
GSuiteStatUtils.SUMMARY_FUNCTIONS_MAPPER[summaryFunc])
analysisSpec.addParameter('tail', 'right-tail')
analysisSpec.addParameter('trackTitles', trackTitles)
results = doAnalysis(analysisSpec, analysisBins,
tracks).getGlobalResult()
core = HtmlCore()
gsPerTrackResultsModel = GSuitePerTrackResultModel(
results, [
'Similarity to rest of tracks in suite (%s)' % summaryFunc,
'P-value'
],
additionalResultsDict=additionalResultsDict,
additionalAttributesDict=additionalAttributesDict)
if choices.leadAttribute and choices.leadAttribute != GSuiteConstants.TITLE_COL:
columnTitles, decoratedResultsDict = \
gsPerTrackResultsModel.generateColumnTitlesAndResultsDict(choices.leadAttribute)
else:
columnTitles, decoratedResultsDict = \
gsPerTrackResultsModel.generateColumnTitlesAndResultsDict()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divClass='results-section')
core.header(analysisQuestion)
topTrackTitle = results.keys()[0]
core.paragraph('''
The track "%s" has the lowest P-value of %s corresponding to %s %s similarity to the rest of the tracks
as measured by "%s" track similarity measure.
''' % (topTrackTitle,
strWithNatLangFormatting(results[topTrackTitle][1]),
strWithNatLangFormatting(results[topTrackTitle][0]),
summaryFunc, similaryStatClassName))
# core.tableFromDictionary(results, columnNames=['Track title', 'Similarity to rest of tracks in suite (' + summaryFunc+')', 'P-value'], sortable=False)
addTableWithTabularAndGsuiteImportButtons(
core,
choices,
galaxyFn,
cls.Q3_SHORT,
decoratedResultsDict,
columnTitles,
gsuite=gsuite,
results=results,
gsuiteAppendAttrs=['similarity_score', 'p_value'],
sortable=True)
core.divEnd()
core.divEnd()
core.end()
else: # Q4
mcfdrDepth = choices.mcfdrDepth if choices.mcfdrDepth else \
AnalysisDefHandler(REPLACE_TEMPLATES['$MCFDR$']).getOptionsAsText().values()[0][0]
analysisDefString = REPLACE_TEMPLATES[
'$MCFDRv3$'] + ' -> CollectionSimilarityHypothesisWrapperStat'
analysisSpec = AnalysisDefHandler(analysisDefString)
analysisSpec.setChoice('MCFDR sampling depth', mcfdrDepth)
analysisSpec.addParameter('assumptions',
'PermutedSegsAndIntersegsTrack')
analysisSpec.addParameter('rawStatistic',
'MultitrackSummarizedInteractionV2Stat')
analysisSpec.addParameter(
'pairwiseStatistic', GSuiteStatUtils.
PAIRWISE_STAT_LABEL_TO_CLASS_MAPPING[similaryStatClassName])
analysisSpec.addParameter(
'summaryFunc',
GSuiteStatUtils.SUMMARY_FUNCTIONS_MAPPER[summaryFunc])
analysisSpec.addParameter('multitrackSummaryFunc',
'avg') # should it be a choice?
analysisSpec.addParameter('tail', 'right-tail')
results = doAnalysis(analysisSpec, analysisBins,
tracks).getGlobalResult()
pval = results['P-value']
observed = results['TSMC_MultitrackSummarizedInteractionV2Stat']
significanceLevel = 'strong' if pval < 0.01 else (
'weak' if pval < 0.05 else 'no')
core = HtmlCore()
core.begin()
core.divBegin(divId='results-page')
core.divBegin(divClass='results-section')
core.header(analysisQuestion)
core.paragraph('''
The tracks in the suite show %s significance in their collective similarity
(average similarity of a track to the rest) of %s
and corresponding p-value of %s,
as measured by "%s" track similarity measure.
''' % (significanceLevel, strWithNatLangFormatting(observed),
strWithNatLangFormatting(pval), similaryStatClassName))
core.divEnd()
core.divEnd()
core.end()
print str(core)
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(choices, galaxyFn=None, username=''):
#data from choices
gSuite = choices.gSuite
plotType = choices.plotType
columnX = choices.columnX
columnY = choices.columnY
plotSeries = choices.plotSeries
axesScaleX = choices.axesScaleX
axesScaleY = choices.axesScaleY
#'linear', 'log10', 'no uniform scale (sorted values as labels)'
if axesScaleX == 'linear':
#plotRes = choices.plotRes
plotRes = 'combine'
elif axesScaleX == 'log10':
plotRes = 'separate'
elif axesScaleX == 'no uniform scale (sorted values as labels)':
plotRes = 'separate'
#unpack gsuite
gSuite = getGSuiteFromGalaxyTN(gSuite)
#full list of attributes (meta-data)
attributeList = gSuite.attributes
#fill list of attributes plus title
attributeList = [TITLE_COL] + attributeList
#dictNum - include numerical values which can be presented in y-axes
#need to do that because pie can have only one chocie and then it is not returing dict
from quick.gsuite.GSuiteUtils import attributesType
attribute = attributesType(gSuite)
dictNum=OrderedDict()
for key, it in attribute.iteritems():
if it == True:
dictNum[key] = False
#check if it is dict or not
if not isinstance(columnY, dict):
tempDict={}
tempDict[columnY] = 'True'
columnY=tempDict
seriesName=[]
#check if user selected categoriesNumber and it is possible to use combinate
categoriesNumber = False
sortedCat=None
categories=None
if columnX == TITLE_COL:
categories = gSuite.allTrackTitles()
elif columnX == 'line number':
categories = None
else:
if columnX in dictNum.keys():
categoriesBefore = [float(v) for v in gSuite.getAttributeValueList(columnX)]
if axesScaleX == 'log10':
for cbN in range(0, len(categoriesBefore)):
if categoriesBefore[cbN]!=0:
categoriesBefore[cbN]=math.log(categoriesBefore[cbN], 10)
sortedCat = sorted(range(len(categoriesBefore)), key=lambda k: categoriesBefore[k])
categories=[]
for n in sortedCat:
categories.append(categoriesBefore[n])
categoriesNumber=True
else:
categories = gSuite.getAttributeValueList(columnX)
#data are sorted according to numerical values
data=[]
for key, it in columnY.iteritems():
if it == 'True':
dataPart=[]
seriesName.append(key)
dataPart = []
for x in gSuite.getAttributeValueList(key):
try:
if axesScaleY == 'log10':
if x!=0:
dataPart.append(math.log(float(x), 10))
else:
dataPart.append(0)
else:
dataPart.append(float(x))
except:
# need to support None in heatmap
if plotType == 'Heatmap':
dataPart.append(0)
else:
dataPart.append(x)
if sortedCat!=None:
dataPartTemp=[]
for n in sortedCat:
dataPartTemp.append(dataPart[n])
dataPart = dataPartTemp
data.append(dataPart)
label=''
if len(seriesName)!=0:
label = '<b>{series.name}</b>: {point.x} {point.y}'
else:
label = '{point.x} {point.y}'
vg = visualizationGraphs()
# 'Column', 'Scatter', 'Heatmap'
if axesScaleX == 'log10':
xAxisTitle = str(columnX) + ' (' + str(axesScaleX) + ')'
else:
xAxisTitle = str(columnX)
if axesScaleY == 'log10':
yAxisTitle = str('values') + ' (' + str(axesScaleY) + ')'
else:
yAxisTitle = str('values')
minFromList = min(min(d) for d in data)
if minFromList > 0:
minFromList = 0
#combain series with data
if plotRes == 'combine':
if categoriesNumber == True:
newData=[]
for d in data:
newDataPart=[]
for cN in range(0, len(categories)):
newDataPart.append([categories[cN], d[cN]])
newData.append(newDataPart)
data=newData
categories=None
res=''
if plotSeries == 'Single':
if plotType == 'Scatter':
res += vg.drawScatterChart(
data,
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = seriesName,
label = label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Pie':
res += vg.drawPieChart(
data[0],
seriesName = categories,
height = 400,
titleText = seriesName[0],
)
if plotType == 'Column':
res += vg.drawColumnChart(
data,
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = seriesName,
label = label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Line':
res += vg.drawLineChart(
data,
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = seriesName,
label = label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Heatmap':
res += vg.drawHeatmapSmallChart(
data,
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = seriesName,
label = label,
# titleText = 'Plot',
)
elif plotSeries == 'Multi':
if plotType == 'Scatter':
for nrD in range(0, len(data)):
if plotRes == 'combine':
data[nrD]=[data[nrD]]
res += vg.drawScatterChart(
data[nrD],
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = [seriesName[nrD]],
label = label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Column':
res += vg.drawColumnCharts(
data,
categories = [categories for x in range(0, len(data))],
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = [[seriesName[elD]] for elD in range(0, len(data))],
label = label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Line':
for nrD in range(0, len(data)):
if plotRes == 'combine':
data[nrD]=[data[nrD]]
res += vg.drawLineChart(
data[nrD],
categories = categories,
xAxisRotation = 90,
marginTop = 30,
xAxisTitle = xAxisTitle,
yAxisTitle = yAxisTitle,
height = 500,
seriesName = [seriesName[nrD]],
label = label,
minY=minFromList
# titleText = 'Plot',
)
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.divBegin(divId='results-page')
htmlCore.divBegin(divClass='results-section')
htmlCore.line(res)
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.end()
print htmlCore
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=''):
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 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 execute(choices, galaxyFn=None, username=''):
file = choices.file
columnX = choices.columnX
columnY = choices.columnY
plotType = choices.plotType
axesScaleX = choices.axesScaleX
axesScaleY = choices.axesScaleY
plotSeries = choices.plotSeries
if axesScaleX == 'linear':
plotRes = 'combine'
elif axesScaleX == 'log10':
plotRes = 'separate'
elif axesScaleX == 'no uniform scale (sorted values as labels)':
plotRes = 'separate'
inputFile = open(
ExternalTrackManager.extractFnFromGalaxyTN(file.split(':')), 'r')
dataS = OrderedDict()
dataS['xAxis'] = OrderedDict()
dataS['yAxis'] = OrderedDict()
i = 0
with inputFile as f:
for x in f.readlines():
if i == 0:
rowColumn = list(x.strip('\n').split('\t'))
else:
j = 0
for el in list(x.strip('\n').split('\t')):
# if columnX[rowColumn[j]] == 'True':
# if not rowColumn[j] in dataS['xAxis']:
# dataS['xAxis'][rowColumn[j]] = []
# dataS['xAxis'][rowColumn[j]].append(el)
if rowColumn[j] in columnX:
if not rowColumn[j] in dataS['xAxis']:
dataS['xAxis'][rowColumn[j]] = []
dataS['xAxis'][rowColumn[j]].append(el)
if rowColumn[j] in columnY and columnY[
rowColumn[j]] == 'True':
if not rowColumn[j] in dataS['yAxis']:
dataS['yAxis'][rowColumn[j]] = []
dataS['yAxis'][rowColumn[j]].append(float(el))
j += 1
i += 1
inputFile.close()
# this will be used just for x - values
# keysX = dataS['xAxis'].keys()
# if keysX == 1:
# plotSeries = 'Single'
# else:
# plotSeries = 'Multi'
#sorting categories values
categoriesNumber = False
sortedCat = None
categories = None
if columnX == 'line number':
categories = None
else:
#if columnX['xAxis'] in columnY.keys():
if columnX in columnY.keys():
categoriesBefore = [float(v) for v in dataS['xAxis'][columnX]]
if axesScaleX == 'log10':
for cbN in range(0, len(categoriesBefore)):
if categoriesBefore[cbN] != 0:
categoriesBefore[cbN] = math.log(
categoriesBefore[cbN], 10)
sortedCat = sorted(range(len(categoriesBefore)),
key=lambda k: categoriesBefore[k])
categories = []
for n in sortedCat:
categories.append(categoriesBefore[n])
categoriesNumber = True
else:
categories = dataS['xAxis'][columnX]
#dataS are sorted according to numerical values
seriesName = []
data = []
for key, it in columnY.iteritems():
if it == 'True':
dataPart = []
seriesName.append(key)
dataPart = []
for x in dataS['yAxis'][key]:
try:
if axesScaleY == 'log10':
if x != 0:
dataPart.append(math.log(float(x), 10))
else:
dataPart.append(0)
else:
dataPart.append(float(x))
except:
dataPart.append(x)
if sortedCat != None:
dataPartTemp = []
for n in sortedCat:
dataPartTemp.append(dataPart[n])
dataPart = dataPartTemp
data.append(dataPart)
label = ''
if len(seriesName) != 0:
label = '<b>{series.name}</b>: {point.x} {point.y}'
else:
label = '{point.x} {point.y}'
# 'Column', 'Scatter', 'Heatmap'
if axesScaleX == 'log10':
xAxisTitle = str(columnX) + ' (' + str(axesScaleX) + ')'
else:
xAxisTitle = str(columnX)
if axesScaleY == 'log10':
yAxisTitle = str('values') + ' (' + str(axesScaleY) + ')'
else:
yAxisTitle = str('values')
minFromList = min(min(d) for d in data)
if minFromList > 0:
minFromList = 0
#combain series with data
if plotRes == 'combine':
if categoriesNumber == True:
newData = []
for d in data:
newDataPart = []
for cN in range(0, len(categories)):
newDataPart.append([categories[cN], d[cN]])
newData.append(newDataPart)
data = newData
categories = None
vg = visualizationGraphs()
res = ''
if plotSeries == 'Single':
if plotType == 'Scatter':
res += vg.drawScatterChart(data,
categories=categories,
xAxisRotation=90,
marginTop=30,
xAxisTitle=xAxisTitle,
yAxisTitle=yAxisTitle,
height=500,
seriesName=seriesName,
label=label,
minY=minFromList)
if plotType == 'Column':
res += vg.drawColumnChart(data,
categories=categories,
xAxisRotation=90,
marginTop=30,
xAxisTitle=xAxisTitle,
yAxisTitle=yAxisTitle,
height=500,
seriesName=seriesName,
label=label,
minY=minFromList)
elif plotSeries == 'Multi':
if plotType == 'Scatter':
for nrD in range(0, len(data)):
if plotRes == 'combine':
data[nrD] = [data[nrD]]
res += vg.drawScatterChart(
data[nrD],
categories=categories,
xAxisRotation=90,
marginTop=30,
xAxisTitle=xAxisTitle,
yAxisTitle=yAxisTitle,
height=500,
seriesName=[seriesName[nrD]],
label=label,
minY=minFromList
# titleText = 'Plot',
)
if plotType == 'Column':
res += vg.drawColumnCharts(
data,
categories=[categories for x in range(0, len(data))],
xAxisRotation=90,
marginTop=30,
xAxisTitle=xAxisTitle,
yAxisTitle=yAxisTitle,
height=500,
seriesName=[[seriesName[elD]]
for elD in range(0, len(data))],
label=label,
minY=minFromList
# titleText = 'Plot',
)
htmlCore = HtmlCore()
htmlCore.begin()
htmlCore.divBegin(divId='results-page')
htmlCore.divBegin(divClass='results-section')
htmlCore.line(res)
htmlCore.divEnd()
htmlCore.divEnd()
htmlCore.end()
print htmlCore
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=''):
'''
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()