def updateOvlpRegions(inpath,outpath,motifChrom='chr17',window=0, stranded=False):
"""exclude motif sites overlapping ENCODE blacklisted regions"""
for infile in glob.glob(os.path.join(outpath,"*"+motifChrom+".bed")):
(filepath,filename) = os.path.split(infile)
tfname = filename.split(motifChrom)[0]
gcoordsfile = open(infile,'rt')
gcoords = csv.reader(gcoordsfile, delimiter='\t')
for exclinfile in glob.glob(os.path.join(inpath,"*Excludable.bed.gz")):
#wgEncodeDacMapabilityConsensusExcludable.bed.gz
(regpath,regfilename) = os.path.split(exclinfile)
expName = regfilename.split('.')[0]
#print 'updating', expName
with gzip.open(exclinfile,'rt') as bedFile:
bed = csv.reader(bedFile, delimiter='\t')
annoIntvlDict = countBed.getBed6Anno(bed,expName)
intervalStartDict = countBed.sortStart(annoIntvlDict)
intervalEndDict = countBed.sortEnd(annoIntvlDict)
exclfile = open(os.path.join(outpath+"/"+expName,tfname+motifChrom+'.bed'),'wt')
writer = csv.writer(exclfile, delimiter='\t')
if stranded:
for test in gcoords:
exonList = []
motifStart, motifEnd, motifStrand = int(test[1])-window, int(test[2])+window, test[-1]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
if valueList != []:##[(exon_name,strand),]
for i in xrange(len(valueList)):
if valueList[i][1] == motifStrand:
exonList.append(regionList[i])
#if exonList != []:
#row = ["_".join(exonList)]
#writer.writerows([row])
if exonList == []:
#row = ['ok']
writer.writerows([test])
else:
for test in gcoords:
motifStart, motifEnd = int(test[1])-window, int(test[2])+window
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
#if regionList != []:
#row = ["_".join(regionList)]
#writer.writerows([row])
if regionList == []:
#row = ['ok']
writer.writerows([test])
exclfile.close()
return 0
def updateChip(path,db,motifChrom="chr17",window=0):
"""mark motifs if it overlaps ENCODE excluded regions"""
mcollection = db["hg19"+motifChrom]
for infile in glob.glob(os.path.join(path,"*.bed.gz")):
(regpath,regfilename) = os.path.split(infile)
expName = regfilename.split(motifChrom)[0]
with gzip.open(infile,'rt') as bedFile:
bed = csv.reader(bedFile, delimiter = '\t')
annoIntvlDict = countBed.getBed4Anno(bed,expName)
intervalStartDict = countBed.sortStart(annoIntvlDict)
intervalEndDict = countBed.sortEnd(annoIntvlDict)
cursor = mcollection.find({"tf_name":{"$in": ["IRF3","MAFK","NFYA","SIN3A","ZNF384"]}})
#{"tf_name": tfName})
for test in cursor:
motifStart, motifEnd = test["genomic_region"]["start"], \
test["genomic_region"]["end"]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
if valueList != []:
#print regionList, valueList, motifChrom, motifStart, motifEnd
mcollection.update({"_id": test["_id"]},
{"$set": {"chip."+expName: valueList[0][1]}}, upsert = True)
return 0
def updateChip(path,db,motifChrom="chr17",window=0):
"""mark motifs if it overlaps ENCODE excluded regions"""
mcollection = db["hg19"+motifChrom]
for tf in ["CTCF"]:#os.listdir(path):#["CTCF","JUND","MAX","REST","SIN3A","SP1","USF1","YY1"]:
files = path+"/"+tf+"/wgEncodeBroadHistoneGm12878CtcfStdPkForma*"+motifChrom+".bed.gz"
for infile in glob.glob(files):
print "update", infile
(regpath,regfilename) = os.path.split(infile)
expName = regfilename.split(motifChrom)[0]
with gzip.open(infile,'rt') as bedFile:
bed = csv.reader(bedFile, delimiter = '\t')
annoIntvlDict = countBed.getBed4Anno(bed,expName)
intervalStartDict = countBed.sortStart(annoIntvlDict)
intervalEndDict = countBed.sortEnd(annoIntvlDict)
cursor = mcollection.find({"tf_name": tf})
for test in cursor:
motifStart, motifEnd = test["genomic_region"]["start"], \
test["genomic_region"]["end"]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
if valueList != []:
#print regionList, valueList, motifChrom, motifStart, motifEnd
mcollection.update({"_id": test["_id"]},
{"$set": {"chip."+expName: valueList[0][1]}}, upsert = True)
return 0
def updateOvlpRegions(path,db,motifChrom='chr17',window=0, stranded=True):
mcollection = db["hg19"+motifChrom]
for infile in glob.glob(os.path.join(path,"hg19_codingExonCanonical_*.bed.gz")):
expName="exon"#"exon1st"
#(regpath,regfilename) = os.path.split(infile)
#expName = regfilename.split('.')[0]
print 'updating', expName
with gzip.open(infile,'rt') as bedFile:
bed = csv.reader(bedFile, delimiter='\t')
annoIntvlDict = countBed.getBed6Anno(bed,expName)
intervalStartDict = countBed.sortStart(annoIntvlDict)
intervalEndDict = countBed.sortEnd(annoIntvlDict)
cursor = mcollection.find()#{"tf_name":{"$in": ["IRF3","MAFK","NFYA","SIN3A","ZNF384"]}})
if stranded:
for test in cursor:
exonList = []
motifStart, motifEnd, motifStrand = test["genomic_region"]["start"], \
test["genomic_region"]["end"], \
test["genomic_region"]["strand"]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
if valueList != []:##[(exon_name,strand),]
#if len(regionList) < len(valueList):
#print motifStart, motifEnd, motifStrand, valueList, regionList
for i in xrange(len(valueList)):
if valueList[i][1] == motifStrand:
if len(regionList) == len(valueList):
exonList.append(regionList[i])
else:
exonList.append(regionList[i-1])
if exonList != []:
mcollection.update({"_id": test["_id"]},
{"$set":{expName: exonList}}, upsert = True)
else:
for test in cursor:
motifStart, motifEnd = test["genomic_region"]["start"], \
test["genomic_region"]["end"]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalStartDict,intervalEndDict,motifChrom,
motifStart,motifEnd,window)
if regionList != []:
mcollection.update({"_id": test["_id"]},
{"$set":{"map."+expName: regionList}}, upsert = True)
return 0
def updateExcludedRegions(path,tfName,window):
"""mark motifs if it overlaps ENCODE excluded regions"""
for infile in glob.glob(os.path.join(path,"*.bed.gz")):
(regpath,regfilename) = os.path.split(infile)
expName = regfilename.split('.')[0]
with gzip.open(infile,'rt') as bedFile:
bed = csv.reader(bedFile, delimiter = '\t')
annoIntvlDict = countBed.getBed6Anno(bed,expName)
intervalDict = countBed.sortInterval(annoIntvlDict)
cursor = mcollection.find({"tf_name": tfName})
for test in cursor:
motifChrom, motifStart, motifEnd = test["motif_genomic_regions_info"]["chr"], \
test["motif_genomic_regions_info"]["start"], \
test["motif_genomic_regions_info"]["end"]
regionList, valueList = countBed.getMotifAnno(annoIntvlDict,
intervalDict,motifChrom,motifStart,motifEnd,window)
if regionList != []:
print regionList, valueList, motifChrom, motifStart, motifEnd
mcollection.update({"_id": test["_id"]},
{"$set": {"motif_mapability_info":{"exclude": regionList}}}, upsert = True)
return 0