def averagesFromFile(VEPannotatedVCF, column2retain , lSOTerm):
"""
VEPannotatedVCF= vcf annotated using VEP
column2retain= list, index of samples to consider
lSOTerm= list of Sequence Ontolgy terms that define variant consequences
"""
dInfo={}; dSOT={}; dImpact={}
for line in gzip.open(args.f, 'r'):
decodedLine=line.decode() ## line.decode() is necessary to read encoded data using gzip in python3
if re.match('##', decodedLine):
if re.search('ID=CSQ' , decodedLine ): csqHeader=decodedLine.rstrip().split(':')[1].lstrip().rstrip('\'>').split('|')
elif re.match('#', decodedLine):
for ind in sampleToConsider:
column2retain.append(decodedLine.split().index(ind))
else:
linesplit=decodedLine.rstrip().split()
mychr=linesplit[0]; mypos=linesplit[1]; myref=linesplit[3]; myalt=linesplit[4] ## basic info
##~~ split INFO field
tempinfo=linesplit[7]
for i in tempinfo.split(';'):
if re.search('=', i): # check if INFO fields has a value corresponding
temp=i.split('=')
dInfo[temp[0]]=temp[1]
else: pass
##~~ work on dInfo[CSQ]
##~~ split for multiple consequences separated by ','
multipleCsq=dInfo['CSQ'].split(',')
##~~ single consequence
#print ('~~~ this is a consequence in a line ')
for mcsq in multipleCsq:
dCsq=dict(zip(csqHeader, mcsq.split('|') )) ############# ALL VEP INFO
#~~~~~~~~~~~ identify the allele with consequences
mycsqAllele=dCsq['Allele']
GTfields=[]
for col in range(args.n): GTfields+=[linesplit[column2retain[col]]]
#nbAploidSamples=len(GTfields)*2
freqCSQ_REF_ALT=gp.AnnotateFreqCSQ_REF_ALT(mycsqAllele,myref, myalt, GTfields) # calculate allelic frequencies
for cons in dCsq['Consequence'].split('&'):
#~~~~~~~~~~~~ assign severity score at the most severe csq
if freqCSQ_REF_ALT[0]!='NA':
if not dCsq['Consequence'] in dSOT: dSOT[dCsq['Consequence']]=[0,0] #inizialize di dictionary with [counter, allele freq] if the key is not present
dSOT[dCsq['Consequence']][0]+=1 #add +1 to the counter
dSOT[dCsq['Consequence']][1]+=float(freqCSQ_REF_ALT[0]) #add the value of the consequence allele
else: listOfErrors.append((mychr, mypos,myref, myalt, dCsq["Allele"]) ) #to be printed in the error file to compare allele matching
CsqMeans=[mychr,freqCSQ_REF_ALT[4]]
for vcsq in lSOTerm:
if vcsq in dSOT: CsqMeans.append(dSOT[vel][1]/float(dSOT[vel][0]))
else: CsqMeans.append('na')
return CsqMeans
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-f",
help="path to input file ",
type=str,
required=True)
parser.add_argument("-v",
help="path to table of vep consequences ",
type=str,
required=True)
parser.add_argument("-o",
help="path to output file ",
type=str,
required=True)
parser.add_argument("-e",
help="path to error file",
type=str,
required=True)
#parser.add_argument("-m", help="path to metadata file",type=str,required=True)
args = parser.parse_args()
#output = open(args.o,'w')
#print(args)
#############################################################
## READ VEP consequences rank ########
"""read external file with info on VEP consequences """
dRank = {
"HIGH": "HIGH",
"LOW": "LOW",
"MODERATE": "MODERATE",
"MODIFIER": "MODIFIER"
}
dSOTermRank = {}
lSOTerm = [] ### list of SOTerm ordered by severity
countlinesCsq = True
for csqLine in open(args.v, 'r'):
if countlinesCsq:
csqTitle = csqLine.rstrip().split('\t')
countlinesCsq = False
else:
myRowList = csqLine.rstrip().split('\t')
dCsq = dict(zip(csqTitle, myRowList))
dSOTermRank[dCsq['SO term']] = dRank[dCsq['IMPACT']]
lSOTerm.append(myRowList[0])
##########~~~~~~~~~~~~~~ Read metadata and randomize sample to choose
# Region=[]
# Sample=[]
#
# for line in open(args.m, 'r'):
# if re.match('sample', line):
# header= line.rstrip().split()
# else:
# other=line.rstrip().split()
# dMeta= dict(zip(header, other))
#
# Region.append(dMeta['region'])
# Sample.append(dMeta['sample'])
#
# dSampleRegion=dict(zip(Sample, Region))
#
# EUR = [key for (key, value) in dSampleRegion.items() if value == 'EUROPE']
# random.seed(899)
#
# sampleToConsider=random.sample(EUR, 6)
##########~~~~~~~~~~~~~~ Loop of vcf lines
sys.stdout = open(args.o, 'w')
listOfErrors = []
dInfo = {}
#column2retain=[]
print("Chr", "\t", "Pos", "\t", "VariantClass", "\t", "CSQallele", "\t",
"CSQrank", "\t", "Consequence", "\t", "CSQfreq", "\t", "REFfreq",
"\t", "ALTfreq", "\t", "MAF", "\t", "Population")
for line in gzip.open(args.f, 'r'):
decodedLine = line.decode(
) ## line.decode() is necessary to read encoded data using gzip in python3
if re.match('#', decodedLine):
if re.search("ID=CSQ", decodedLine):
csqHeader = decodedLine.rstrip().split(":")[1].lstrip().rstrip(
"\">").split("|")
#filemyres.write(decodedLine)
# elif re.match('#', decodedLine):
# for ind in sampleToConsider:
# column2retain.append(decodedLine.split().index(ind))
else:
linesplit = decodedLine.rstrip().split()
mychr = linesplit[0]
mypos = linesplit[1]
myref = linesplit[3]
myalt = linesplit[4] ## basic info
##~~ split INFO field
tempinfo = linesplit[7]
for i in tempinfo.split(";"):
if re.search('=', i):
temp = i.split("=")
dInfo[temp[0]] = temp[1]
else:
pass
##~~ work on dInfo[CSQ]
##~~ split for multiple consequences separated by ","
multipleCsq = dInfo["CSQ"].split(",")
##~~ single consequence
#print ('~~~ this is a consequence in a line ')
CSQcount = 0
for mcsq in multipleCsq:
CSQcount += 1
dCsq = dict(zip(
csqHeader, mcsq.split("|"))) ############# ALL VEP INFO
#~~~~~~~~~~~ identify the allele with consequences
mycsqAllele = dCsq["Allele"]
GTfields = linesplit[9:]
nbAploidSamples = len(GTfields) * 2
freqCSQ_REF_ALT = gp.AnnotateFreqCSQ_REF_ALT(
mycsqAllele, myref, myalt, nbAploidSamples, GTfields)
for c in dCsq['Consequence'].split("&"):
#~~~~~~~~~~~~ assign severity score at the most severe csq
myindexes = []
myindexes.append(lSOTerm.index(c))
mostSevereCsq = lSOTerm[min(myindexes)]
linesplit[7] = tempinfo # reset info field
print(linesplit[0], "\t", linesplit[1], "\t",
dCsq["VARIANT_CLASS"], "\t", dCsq["Allele"], "\t",
dSOTermRank[mostSevereCsq], "\t", c, "\t",
freqCSQ_REF_ALT[0], "\t", freqCSQ_REF_ALT[1], "\t",
freqCSQ_REF_ALT[2], "\t", freqCSQ_REF_ALT[3], "\t",
"GREP")
#fileToWrite=open(args.e, 'w')
#for i in listOfErrors: fileToWrite.write( i )
sys.stdout.close()
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-f",
help="path to input file ",
type=str,
required=True)
parser.add_argument("-v",
help="path to table of vep consequences ",
type=str,
required=True)
parser.add_argument("-o",
help="path to output file ",
type=str,
required=True)
parser.add_argument("-e",
help="path to error file",
type=str,
required=True)
args = parser.parse_args()
#output = open(args.o,'w')
#print(args)
#############################################################
## READ VEP consequences rank ########
"""read external file with info on VEP consequences """
dRank = {
"HIGH": "HIGH",
"LOW": "LOW",
"MODERATE": "MODERATE",
"MODIFIER": "MODIFIER"
}
dSOTermRank = {}
lSOTerm = [] ### list of SOTerm ordered by severity
countlinesCsq = True
for csqLine in open(args.v, 'r'):
if countlinesCsq:
csqTitle = csqLine.rstrip().split('\t')
countlinesCsq = False
else:
myRowList = csqLine.rstrip().split('\t')
dCsq = dict(zip(csqTitle, myRowList))
dSOTermRank[dCsq['SO term']] = dRank[dCsq['IMPACT']]
lSOTerm.append(myRowList[0])
##########~~~~~~~~~~~~~~ Loop of vcf lines
filemyres = open(args.o, 'w')
listOfErrors = []
dInfo = {}
for line in gzip.open(args.f, 'r'):
decodedLine = line.decode(
) ## line.decode() is necessary to read encoded data using gzip in python3
if re.match('##', decodedLine):
if re.search("ID=CSQ", decodedLine):
csqHeader = decodedLine.rstrip().split(":")[1].lstrip().rstrip(
"\">").split("|")
filemyres.write(decodedLine)
elif re.match('#', decodedLine):
filemyres.write(
'##INFO=<ID=CSQfreq,Number=A,Type=Float,Description="Frequency of CSQ allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=REFfreq,Number=A,Type=Float,Description="Frequency of REF allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=ALTfreq,Number=A,Type=Float,Description="Frequency of ALT allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=MAF,Number=A,Type=Float,Description="Frequency of Minor allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=nbCSQ,Number=A,Type=Integer,Description="1st, 2nd .... CSQ allele">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=CSQallele,Number=A,Type=String,Description="Describe CSQ allele">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=IMPACT,Number=A,Type=String,Description="Impact of CSQ allele">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=Existing_variation,Number=A,Type=String,Description="RS of CSQallele">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=VARIANT_CLASS,Number=A,Type=String,Description="type of variant : MODIFIER ecc..">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=Consequence,Number=A,Type=String,Description="type of variant : SNP, indels, ecc...">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=CSQrank,Number=A,Type=String,Description="Rank of Consequence: HIGH, MODERATE, LOW, MODIFIER">'
)
filemyres.write("\n")
filemyres.write(decodedLine)
else:
#print("this is a new line ") ## line split by tab
linesplit = decodedLine.rstrip().split()
#print(linesplit)
mychr = linesplit[0]
mypos = linesplit[1]
myref = linesplit[3]
myalt = linesplit[4] ## basic info
##~~ split INFO field
tempinfo = linesplit[7]
for i in tempinfo.split(";"):
temp = i.split("=")
dInfo[temp[0]] = temp[1]
##~~ work on dInfo[CSQ]
##~~ split for multiple consequences separated by ","
multipleCsq = dInfo["CSQ"].split(",")
##~~ single consequence
#print ('~~~ this is a consequence in a line ')
CSQcount = 0
for mcsq in multipleCsq:
CSQcount += 1
myres = []
#myres+=[mychr, mypos]
dCsq = dict(zip(
csqHeader, mcsq.split("|"))) ############# ALL VEP INFO
#print (dCsq)
#myres.append(dCsq['Existing_variation'])
#~~~~~~~~~~~ identify the allele with consequences
#linesplit[7]=tempinfo # reset info field
mycsqAllele = dCsq["Allele"]
GTfields = linesplit[9:]
nbAploidSamples = len(GTfields) * 2
freqCSQ_REF_ALT = gp.AnnotateFreqCSQ_REF_ALT(
mycsqAllele, myref, myalt, nbAploidSamples, GTfields)
#print(freqCSQ_REF_ALT)
#print (dCsq['Consequence'].split("&"))
for c in dCsq['Consequence'].split("&"):
#~~~~~~~~~~~~ assign severity score at the most severe csq
myindexes = []
myindexes.append(lSOTerm.index(c))
mostSevereCsq = lSOTerm[min(myindexes)]
linesplit[7] = tempinfo # reset info field
linesplit[7] += ";CSQallele="
linesplit[7] += dCsq["Allele"]
linesplit[7] += ";Consequence="
linesplit[7] += c
linesplit[7] += ";CSQrank="
linesplit[7] += dSOTermRank[mostSevereCsq]
#linesplit[7]+=";IMPACT="
#linesplit[7]+=dCsq["IMPACT"]
linesplit[7] += ";ExistingVariation="
linesplit[7] += dCsq["Existing_variation"]
linesplit[7] += ";VariantClass="
linesplit[7] += dCsq["VARIANT_CLASS"]
linesplit[
7] += ";nbCSQ=" # for specify the number of CSQ allele
linesplit[7] += str(
CSQcount) # for specify the number of CSQ allele
linesplit[7] += ";CSQfreq=%s" % (freqCSQ_REF_ALT[0])
linesplit[7] += ";REFfreq=%s" % (freqCSQ_REF_ALT[1])
linesplit[7] += ";ALTfreq=%s" % (freqCSQ_REF_ALT[2])
linesplit[7] += ";MAF=%s" % (freqCSQ_REF_ALT[3])
filemyres.write("\t".join(linesplit))
filemyres.write("\n")
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-f",
help="path to input file ",
type=str,
required=True)
parser.add_argument("-o",
help="path to output file ",
type=str,
required=True)
parser.add_argument("-e",
help="path to error file",
type=str,
required=True)
args = parser.parse_args()
#output = open(args.o,'w')
#print(args)
##########~~~~~~~~~~~~~~ Loop of vcf lines
filemyres = open(args.o, 'w')
listOfErrors = []
dInfo = {}
for line in gzip.open(args.f, 'r'):
decodedLine = line.decode(
) ## line.decode() is necessary to read encoded data using gzip in python3
if re.match('##', decodedLine):
if re.search("ID=CSQ", decodedLine):
csqHeader = decodedLine.rstrip().split(":")[1].lstrip().rstrip(
"\">").split("|")
filemyres.write(decodedLine)
elif re.match('#', decodedLine):
filemyres.write(
'##INFO=<ID=CSQfreq,Number=A,Type=Float,Description="Frequency of CSQ allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=REFfreq,Number=A,Type=Float,Description="Frequency of REF allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=ALTfreq,Number=A,Type=Float,Description="Frequency of ALT allele in set of samples">'
)
filemyres.write("\n")
filemyres.write(
'##INFO=<ID=nbCSQ,Number=A,Type=Integer,Description="1st, 2nd .... CSQ allele">'
)
filemyres.write("\n")
filemyres.write(decodedLine)
else:
#print("this is a new line ") ## line split by tab
linesplit = decodedLine.rstrip().split()
#print(linesplit)
mychr = linesplit[0]
mypos = linesplit[1]
myref = linesplit[3]
myalt = linesplit[4] ## basic info
#nbOfAltAlleles=len(myalt.split(","))
#if nbOfAltAlleles> 2: #~~ excludes cases with more than two alt allele, want to add the excluded in the error output
#listOfErrors+=( '\t'.join([mychr, mypos, 'more than two alternate alleles', nbOfAltAlleles , '\n']))
#pass
#else:
##~~ split INFO field
tempinfo = linesplit[7]
for i in tempinfo.split(";"):
temp = i.split("=")
dInfo[temp[0]] = temp[1]
##~~ work on dInfo[CSQ]
##~~ split for multiple consequences separated by ","
multipleCsq = dInfo["CSQ"].split(",")
##~~ single consequence
#print ('~~~ this is a consequence in a line ')
CSQcount = 0
for mcsq in multipleCsq:
CSQcount += 1
myres = []
#myres+=[mychr, mypos]
dCsq = dict(zip(
csqHeader, mcsq.split("|"))) ############# ALL VEP INFO
#print (dCsq)
#myres.append(dCsq['Existing_variation'])
#~~~~~~~~~~~ identify the allele with consequences
linesplit[7] = tempinfo # reset info field
mycsqAllele = dCsq["Allele"]
GTfields = linesplit[9:]
nbAploidSamples = len(GTfields) * 2
freqCSQ_REF_ALT = gp.AnnotateFreqCSQ_REF_ALT(
mycsqAllele, myref, myalt, nbAploidSamples, GTfields)
#print(freqCSQ_REF_ALT)
linesplit[7] += ";"
linesplit[
7] += "nbCSQ=" # for specify the number of CSQ allele
linesplit[7] += str(
CSQcount) # for specify the number of CSQ allele
linesplit[7] += ";CSQfreq=%s" % (freqCSQ_REF_ALT[0])
linesplit[7] += ";REFfreq=%s" % (freqCSQ_REF_ALT[1])
linesplit[7] += ";ALTfreq=%s" % (freqCSQ_REF_ALT[2])
#CSQfreq="CSQfreq=%s" %(freqCSQ_REF_ALT[0])
#REFfreq="REFfreq=%s" %(freqCSQ_REF_ALT[1])
#ALTfreq="ALTfreq=%s" %(freqCSQ_REF_ALT[2])
#allfreq=CSQfreq,REFfreq,ALTfreq
#freqField=";".join(map(str,allfreq))
#rowWithFreq=linesplit[7],freqField
#row2print=";".join(map(str,rowWithFreq))
#linesplit[7]=row2print
filemyres.write("\t".join(linesplit))
filemyres.write("\n")
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-f", help="path to input file ",type=str,required=True)
parser.add_argument("-v", help="path to table of vep consequences ",type=str, required= True)
parser.add_argument("-o", help="path to output file ",type=str, required= True)
parser.add_argument("-e", help="path to error file",type=str,required=True)
parser.add_argument("-m", help="path to metadata file",type=str,required=True)
parser.add_argument("-c", help="number of random cycle",type=int,required=False, default=1)
parser.add_argument("-s", help="seed's number",type=int,required=True)
args = parser.parse_args()
#output = open(args.o,'w')
#print(args)
#############################################################
##########~~~~~~~~~~~~~~ READ VEP consequences rank
"""read external file with info on VEP consequences """
dRank={"HIGH":"HIGH", "LOW": "LOW", "MODERATE":"MODERATE", "MODIFIER":"MODIFIER"}
dSOTermRank={}
lSOTerm=[] ### list of SOTerm ordered by severity
countlinesCsq= True
for csqLine in open(args.v, 'r'):
if countlinesCsq:
csqTitle=csqLine.rstrip().split('\t')
countlinesCsq=False
else:
myRowList=csqLine.rstrip().split('\t')
dCsq= dict(zip(csqTitle, myRowList ))
dSOTermRank[dCsq['SO term']]=dRank[dCsq['IMPACT']]
lSOTerm.append(myRowList[0])
##########~~~~~~~~~~~~~~ Read metadata
Region=[]
Sample=[]
for line in open(args.m, 'r'):
if re.match('sample', line):
header= line.rstrip().split()
else:
other=line.rstrip().split()
dMeta= dict(zip(header, other))
Region.append(dMeta['region'])
Sample.append(dMeta['sample'])
dSampleRegion=dict(zip(Sample, Region))
EUR = [key for (key, value) in dSampleRegion.items() if value == 'EUROPE']
EURsorted = sorted(EUR) ## needed for seed
random.seed(args.s) ## need a sorted list of EUR
#sampleToConsider=random.sample(EUR, 6)
##########~~~~~~~~~~~~~~ Loop of vcf lines
sys.stdout=open(args.o, 'w')
listOfErrors=[]
print("Chr","\t", "Pos","\t","VariantClass","\t", "CSQallele","\t","CSQrank","\t","Consequence","\t","CSQfreq","\t","REFfreq","\t","ALTfreq","\t","MAF","\t","Cycle","\t","Population")
cycle=0
while cycle < args.c:
cycle+=1
myinput=gzip.open(args.f, 'r')
dInfo={}
column2retain=[]
sampleToConsider=random.sample(EURsorted, 6)
for line in myinput:
decodedLine=line.decode() ## line.decode() is necessary to read encoded data using gzip in python3
if re.match('##', decodedLine):
if re.search("ID=CSQ" , decodedLine ):
csqHeader=decodedLine.rstrip().split(":")[1].lstrip().rstrip("\">").split("|")
elif re.match('#', decodedLine):
for ind in sampleToConsider:
column2retain.append(decodedLine.split().index(ind))
else:
#print("this is a new line ") ## line split by tab
linesplit=decodedLine.rstrip().split()
mychr=linesplit[0]; mypos=linesplit[1]; myref=linesplit[3]; myalt=linesplit[4] ## basic info
##~~ split INFO field
tempinfo=linesplit[7]
for i in tempinfo.split(";"):
if re.search('=', i): # check if INFO fields has a value corresponding
temp=i.split("=")
dInfo[temp[0]]=temp[1]
else: pass
##~~ work on dInfo[CSQ]
##~~ split for multiple consequences separated by ","
multipleCsq=dInfo["CSQ"].split(",")
##~~ single consequence
#print ('~~~ this is a consequence in a line ')
CSQcount=0
for mcsq in multipleCsq:
CSQcount+=1
dCsq=dict(zip(csqHeader, mcsq.split("|") )) ############# ALL VEP INFO
#~~~~~~~~~~~ identify the allele with consequences
mycsqAllele=dCsq["Allele"]
GTfields=[]
for col in range(6): GTfields+=[linesplit[column2retain[col]]]
nbAploidSamples=len(GTfields)*2
freqCSQ_REF_ALT=gp.AnnotateFreqCSQ_REF_ALT(mycsqAllele,myref, myalt, nbAploidSamples, GTfields)
for cons in dCsq['Consequence'].split("&"):
#~~~~~~~~~~~~ assign severity score at the most severe csq
myindexes=[]
myindexes.append(lSOTerm.index(cons))
mostSevereCsq=lSOTerm[min(myindexes)]
print(linesplit[0],"\t",linesplit[1],"\t",dCsq["VARIANT_CLASS"],"\t",dCsq["Allele"],"\t",dSOTermRank[mostSevereCsq],"\t",cons,"\t",freqCSQ_REF_ALT[0],"\t",freqCSQ_REF_ALT[1],"\t",freqCSQ_REF_ALT[2],"\t",freqCSQ_REF_ALT[3],"\t",cycle,"\t","HGDP")