def main(argv):
if (len(argv) < 1):
sys.stderr.write("Error: Number of arguments incorrect\n")
usage()
sys.exit()
else:
hidden_decoys = False
hidden_pattern = ""
decoy_pattern = "random"
verbose = False
ties = True
label = ""
try:
opts, args = getopt.getopt(
sys.argv[2:], "e:d:vhl:t",
["hidden=", "prefix=", "verbose", "help", "label", "ties"])
except getopt.GetoptError, err:
# print help information and exit:
print str(
err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
for o, a in opts:
if o == "-v":
verbose = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-e", "--hidden"):
hidden_decoys = True
hidden_pattern = a
elif o in ("-d", "--prefix"):
decoy_pattern = a
elif o in ("-l", "--label"):
label = a
elif o in ("-t", "--ties"):
ties = a
else:
assert False, "unhandled option"
if (os.path.isfile(argv[0])):
infile = argv[0]
else:
sys.stderr.write("Error: file " + str(argv[0]) + " not found\n")
sys.exit()
colors = [
"red", "blue", "yellow", "black", "brown", "pink", "cyan",
"darkblue", "darkred"
]
hits = list()
names = list()
##READING ELEMENTS##
for line in open(infile).readlines():
words = line.split()
prob_file = words[0]
names.append(words[1])
if (verbose):
print "reading file " + str(prob_file)
if (os.path.isfile(prob_file)):
hits.append(util.importer(prob_file))
else:
sys.stderr.write("Error: file " + str(prob_file) +
" not found\n")
sys.exit()
pvalues = list(xrange(len(hits)))
pvalues2 = list(xrange(len(hits)))
for x in xrange(len(hits)):
##ESTIMATING PVALUES FROM PEP##
if (hidden_decoys):
pvalues[x] = pep2pvalue([(ele.pep, ele.protein) for ele in [
ele
for ele in hits[x] if ele.protein.find(decoy_pattern) == -1
]], ties)
pvalues[x] = [
ele[-2] for ele in pvalues[x]
if ele[-1].find(hidden_pattern) != -1
]
pvalues[x] = remove_zeroes(pvalues[x])
pvalues[x] = sorted(pvalues[x], reverse=False)
pvalues2[x] = estimate_pvalues([(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(hidden_pattern) != -1],\
[(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) != -1],\
decoy_pattern)
else:
pvalues[x] = pep2pvalue([(ele.pep, ele.protein)
for ele in hits[x]], ties)
pvalues[x] = [
ele[-2] for ele in pvalues[x]
if ele[-1].find(decoy_pattern) != -1
]
pvalues[x] = remove_zeroes(pvalues[x])
pvalues[x] = sorted(pvalues[x], reverse=False)
pvalues2[x] = estimate_pvalues([(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) == -1],\
[(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) != -1],\
decoy_pattern)
if (verbose):
print "generatings plots"
##PLOTTING##
util.plotHist(pvalues2, names, colors, "estimated $p$ value",
"#target " + label, label + "_pvalue_estimated.png")
util.plot_pvalues_calibration(
pvalues, names, colors, label + "_pvalue_calibration_from_pep.png")
util.plot_pvalues_calibration(pvalues2, names, colors,
label + "_pvalue_calibration.png")
if (verbose):
print "plots generated"
def main(argv):
if (len(argv) < 1):
sys.stderr.write("Error: Number of arguments incorrect\n")
usage()
sys.exit()
else:
hidden_decoys = False
hidden_pattern = ""
decoy_pattern = "random"
verbose = False
ties = False
countdecoys = True
label = ""
fdr = 0.01
tdratio = 0.0
pi0 = False
try:
opts, args = getopt.getopt(sys.argv[2:], "e:d:vhl:f:i:r:tc", [
"hidden=", "prefix=", "verbose", "help", "label=", "fdr=",
"pi0=", "tdratio=", "ties", "countdecoys"
])
except getopt.GetoptError, err:
# print help information and exit:
print str(
err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
for o, a in opts:
if o == "-v":
verbose = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-e", "--hidden"):
hidden_decoys = True
hidden_pattern = a
elif o in ("-d", "--prefix"):
decoy_pattern = a
elif o in ("-l", "--label"):
label = a
elif o in ("-f", "--fdr"):
fdr = float(a)
elif o in ("-r", "--tdratio"):
tdratio = float(a)
print "Using Target Decoy Ratio = " + str(tdratio)
elif o in ("-i", "--pi0"):
pi0 = True
print "Using pi0 to adjust empirical q values"
elif o in ("-c", "--countdecoys"):
countdecoys = True
print "Not counting decoys when computing empirical q values"
elif o in ("-t", "--ties"):
ties = True
print "Counting clusters as one"
else:
assert False, "unhandled option"
if (os.path.isfile(argv[0])):
infile = argv[0]
else:
sys.stderr.write("Error: file " + str(argv[0]) + " not found\n")
sys.exit()
colors = [
"red", "blue", "yellow", "black", "brown", "pink", "cyan",
"darkblue", "darkred"
]
hits = list()
names = list()
##READING ELEMENTS##
for line in open(infile).readlines():
words = line.split()
if (line != ""):
prob_file = words[0]
names.append(words[1])
if (os.path.isfile(prob_file)):
hits.append(util.importer(prob_file))
if (verbose):
print "reading file " + str(prob_file)
else:
sys.stderr.write("Error: file " + str(prob_file) +
" not found\n")
sys.exit()
pi0s = list(xrange(len(hits)))
qvaluesEmps = list(xrange(len(hits)))
qvaluesEsts = list(xrange(len(hits)))
if (hidden_decoys):
print "Estimating qvalues for hidden decoys mode..."
for x in xrange(len(hits)):
##ESTIMATING PI0##
if (hidden_decoys):
hits[x] = [
ele for ele in hits[x]
if ele.protein.find(decoy_pattern) == -1
]
if (pi0):
pi0s[x] = estimatePi0(
getPValues(hits[x], True, hidden_pattern))
else:
pi0s[x] = 1.0
else:
if (pi0):
pi0s[x] = estimatePi0(
getPValues(hits[x], True, decoy_pattern))
else:
pi0s[x] = 1.0
if (pi0s[x] > 1.0 or pi0s[x] < 0.0):
pi0s[x] = 1.0
##ESTIMATING QVALUEs##
ndecoys = 0
ntargets = 0
ndecoysEmp = 0
ntargetEmp = 0
if (hidden_decoys):
qvaluesEmps[x], qvaluesEsts[
x], ndecoys, ntargets, ndecoysEmp, ntargetEmp = estimateFDR_hidden_decoys(
hits[x], pi0s[x], hidden_pattern, False, ties,
countdecoys, fdr, tdratio)
else:
qvaluesEmps[x], qvaluesEsts[
x], ndecoys, ntargets, ndecoysEmp, ntargetEmp = estimateQvalues(
hits[x], pi0s[x], decoy_pattern, False, ties,
countdecoys, fdr, tdratio)
if (pi0):
print "pi0 file " + str(x + 1) + " :" + str(pi0s[x])
print "elements with estimated qvalue below " + str(
fdr) + " in file " + str(x + 1) + " :" + str(ntargets)
print "elements with empirical qvalue below " + str(
fdr) + " in file " + str(x + 1) + " :" + str(ntargetEmp)
print "false positive elements with estimated qvalue below " + str(
fdr) + " in file " + str(x + 1) + " :" + str(ndecoys)
print "false positive elements with empirical qvalue below " + str(
fdr) + " in file " + str(x + 1) + " :" + str(ndecoysEmp)
if (verbose):
print "generatings plots"
##PLOTTING##
util.plotHist(qvaluesEsts, names, colors, "estimated $q$ value",
"#target " + label, label + "_qvalue_estimated.png", fdr)
util.plotHist(qvaluesEmps, names, colors, "empirical $q$ value",
"#target " + label, label + "_qvalue_empirical.png", fdr)
util.plotCorrelation(
qvaluesEmps, qvaluesEsts, names, colors, "empirical $q$ value",
"estimated $q$ value",
label + "_qvalue_estimated_VS_qvalue_empirical_low_range.png", 1.0)
util.plotCorrelation(
qvaluesEmps, qvaluesEsts, names, colors, "empirical $q$ value",
"estimated $q$ value",
label + "_qvalue_estimated_VS_qvalue_empirical.png", fdr)
if (verbose):
print "plots generated"
def main(argv):
if( len(argv) < 1):
sys.stderr.write("Error: Number of arguments incorrect\n")
usage()
sys.exit()
else:
hidden_decoys = False
hidden_pattern = ""
decoy_pattern = "random"
verbose = False
ties = True
label = ""
try:
opts, args = getopt.getopt(sys.argv[2:], "e:d:vhl:t", ["hidden=", "prefix=", "verbose", "help", "label", "ties"])
except getopt.GetoptError, err:
# print help information and exit:
print str(err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
for o, a in opts:
if o == "-v":
verbose = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-e", "--hidden"):
hidden_decoys = True
hidden_pattern = a
elif o in ("-d", "--prefix"):
decoy_pattern = a
elif o in ("-l", "--label"):
label = a
elif o in ("-t", "--ties"):
ties = a
else:
assert False, "unhandled option"
if(os.path.isfile(argv[0])):
infile = argv[0]
else:
sys.stderr.write("Error: file " + str(argv[0]) + " not found\n")
sys.exit()
colors = ["red","blue","yellow","black","brown","pink","cyan","darkblue","darkred"]
hits = list()
names = list()
##READING ELEMENTS##
for line in open(infile).readlines():
words = line.split()
prob_file = words[0]
names.append(words[1])
if(verbose):
print "reading file " +str(prob_file)
if(os.path.isfile(prob_file)):
hits.append(util.importer(prob_file))
else:
sys.stderr.write("Error: file " + str(prob_file) + " not found\n")
sys.exit()
pvalues = list(xrange(len(hits)))
pvalues2 = list(xrange(len(hits)))
for x in xrange(len(hits)):
##ESTIMATING PVALUES FROM PEP##
if(hidden_decoys):
pvalues[x] = pep2pvalue([(ele.pep,ele.protein) for ele in [ele for ele in hits[x] if ele.protein.find(decoy_pattern) == -1]],ties)
pvalues[x] = [ele[-2] for ele in pvalues[x] if ele[-1].find(hidden_pattern) != -1]
pvalues[x] = remove_zeroes(pvalues[x])
pvalues[x] = sorted(pvalues[x],reverse=False)
pvalues2[x] = estimate_pvalues([(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(hidden_pattern) != -1],\
[(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) != -1],\
decoy_pattern)
else:
pvalues[x] = pep2pvalue([(ele.pep,ele.protein) for ele in hits[x]],ties)
pvalues[x] = [ele[-2] for ele in pvalues[x] if ele[-1].find(decoy_pattern) != -1]
pvalues[x] = remove_zeroes(pvalues[x])
pvalues[x] = sorted(pvalues[x],reverse=False)
pvalues2[x] = estimate_pvalues([(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) == -1],\
[(ele.score,ele.protein) for ele in hits[x] if ele.protein.find(decoy_pattern) != -1],\
decoy_pattern)
if(verbose):
print "generatings plots"
##PLOTTING##
util.plotHist(pvalues2,names,colors,"estimated $p$ value", "#target " + label, label + "_pvalue_estimated.png")
util.plot_pvalues_calibration(pvalues,names,colors,label + "_pvalue_calibration_from_pep.png")
util.plot_pvalues_calibration(pvalues2,names,colors,label + "_pvalue_calibration.png")
if(verbose):
print "plots generated"
def main(argv):
if( len(argv) < 1):
sys.stderr.write("Error: Number of arguments incorrect\n")
usage()
sys.exit()
else:
hidden_decoys = False
hidden_pattern = ""
decoy_pattern = "random"
verbose = False
ties = False
countdecoys = True
label = ""
fdr = 0.01
tdratio = 0.0
pi0 = False
try:
opts, args = getopt.getopt(sys.argv[2:], "e:d:vhl:f:i:r:tc", ["hidden=", "prefix=", "verbose", "help", "label=", "fdr=", "pi0=", "tdratio=","ties","countdecoys"])
except getopt.GetoptError, err:
# print help information and exit:
print str(err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
for o, a in opts:
if o == "-v":
verbose = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-e", "--hidden"):
hidden_decoys = True
hidden_pattern = a
elif o in ("-d", "--prefix"):
decoy_pattern = a
elif o in ("-l", "--label"):
label = a
elif o in ("-f", "--fdr"):
fdr = float(a)
elif o in ("-r", "--tdratio"):
tdratio = float(a)
print "Using Target Decoy Ratio = " + str(tdratio)
elif o in ("-i", "--pi0"):
pi0 = True
print "Using pi0 to adjust empirical q values"
elif o in ("-c", "--countdecoys"):
countdecoys = True
print "Not counting decoys when computing empirical q values"
elif o in ("-t", "--ties"):
ties = True
print "Counting clusters as one"
else:
assert False, "unhandled option"
if(os.path.isfile(argv[0])):
infile = argv[0]
else:
sys.stderr.write("Error: file " + str(argv[0]) + " not found\n")
sys.exit()
colors = ["red","blue","yellow","black","brown","pink","cyan","darkblue","darkred"]
hits = list()
names = list()
##READING ELEMENTS##
for line in open(infile).readlines():
words = line.split()
if(line != ""):
prob_file = words[0]
names.append(words[1])
if(os.path.isfile(prob_file)):
hits.append(util.importer(prob_file))
if(verbose):
print "reading file " +str(prob_file)
else:
sys.stderr.write("Error: file " + str(prob_file) + " not found\n")
sys.exit()
pi0s = list(xrange(len(hits)))
qvaluesEmps = list(xrange(len(hits)))
qvaluesEsts = list(xrange(len(hits)))
if(hidden_decoys):
print "Estimating qvalues for hidden decoys mode..."
for x in xrange(len(hits)):
##ESTIMATING PI0##
if(hidden_decoys):
hits[x] = [ele for ele in hits[x] if ele.protein.find(decoy_pattern) == -1]
if(pi0):
pi0s[x] = estimatePi0(getPValues(hits[x],True,hidden_pattern))
else:
pi0s[x] = 1.0
else:
if(pi0):
pi0s[x] = estimatePi0(getPValues(hits[x],True,decoy_pattern))
else:
pi0s[x] = 1.0
if(pi0s[x] > 1.0 or pi0s[x] < 0.0):
pi0s[x] = 1.0
##ESTIMATING QVALUEs##
ndecoys = 0
ntargets = 0
ndecoysEmp = 0
ntargetEmp = 0
if(hidden_decoys):
qvaluesEmps[x],qvaluesEsts[x],ndecoys,ntargets,ndecoysEmp,ntargetEmp = estimateFDR_hidden_decoys(hits[x],pi0s[x],hidden_pattern,False,ties,countdecoys,fdr,tdratio)
else:
qvaluesEmps[x],qvaluesEsts[x],ndecoys,ntargets,ndecoysEmp,ntargetEmp = estimateQvalues(hits[x],pi0s[x],decoy_pattern,False,ties,countdecoys,fdr,tdratio)
if(pi0):
print "pi0 file " + str(x+1) + " :" + str(pi0s[x])
print "elements with estimated qvalue below " + str(fdr) + " in file " + str(x+1) + " :" + str(ntargets)
print "elements with empirical qvalue below " + str(fdr) + " in file " + str(x+1) + " :" + str(ntargetEmp)
print "false positive elements with estimated qvalue below " + str(fdr) + " in file " + str(x+1) + " :" + str(ndecoys)
print "false positive elements with empirical qvalue below " + str(fdr) + " in file " + str(x+1) + " :" + str(ndecoysEmp)
if(verbose):
print "generatings plots"
##PLOTTING##
util.plotHist(qvaluesEsts,names,colors,"estimated $q$ value", "#target " + label, label + "_qvalue_estimated.png",fdr)
util.plotHist(qvaluesEmps,names,colors,"empirical $q$ value", "#target " + label, label + "_qvalue_empirical.png",fdr)
util.plotCorrelation(qvaluesEmps,qvaluesEsts,names,colors,"empirical $q$ value","estimated $q$ value",label + "_qvalue_estimated_VS_qvalue_empirical_low_range.png",1.0)
util.plotCorrelation(qvaluesEmps,qvaluesEsts,names,colors,"empirical $q$ value","estimated $q$ value",label + "_qvalue_estimated_VS_qvalue_empirical.png",fdr)
if(verbose):
print "plots generated"
