def runMC(length):
print 'Markov Chain length %d:' % length
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildMarkovChain(counts, coding, noncoding, length)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores, length)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (endTime-startTime, buildTime-startTime, endTime-buildTime)
writeORFs(approved, suspect)
compareResults.compare('trueORFs.txt', 'predicted.txt')
print ''
def runIMM():
print 'IMM max length %d:' % maxLength
#maxLength = maxImmLength
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildIMM(counts, coding, noncoding)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (endTime-startTime, buildTime-startTime, endTime-buildTime)
writeORFs(approved, suspect)
compareResults.compare('trueORFs.txt', 'predicted.txt')
print ''
def runMC(length):
''' Run the fixed-length Markov model classifier and print the accuracy of the results.
'''
print 'Markov Chain length %d:' % length
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildMarkovChain(counts, coding, noncoding, length)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores, length)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (endTime-startTime, buildTime-startTime, endTime-buildTime)
writeORFs(approved, suspect)
compareResults.compare(trueFile, 'predicted.txt')
print ''
def runIMM():
''' Run the IMM classifier and print the accuracy of the results.
'''
print 'IMM max length %d:' % maxLength
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildIMM(counts, coding, noncoding)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (endTime-startTime, buildTime-startTime, endTime-buildTime)
writeORFs(approved, suspect)
compareResults.compare(trueFile, 'predicted.txt')
print ''
def runMC(length):
''' Run the fixed-length Markov model classifier and print the accuracy of the results.
'''
print 'Markov Chain length %d:' % length
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildMarkovChain(counts, coding, noncoding, length)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores, length)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (
endTime - startTime, buildTime - startTime, endTime - buildTime)
writeORFs(approved, suspect)
compareResults.compare(trueFile, 'predicted.txt')
print ''
def runIMM():
''' Run the IMM classifier and print the accuracy of the results.
'''
print 'IMM max length %d:' % maxLength
counts, immScores = init()
startTime = time.time()
coding, noncoding = findLongORFs(genome)
counts = buildIMM(counts, coding, noncoding)
buildTime = time.time()
approved, suspect = glimmer(genome, counts, immScores)
endTime = time.time()
print ' %0.2fs total (%0.2fs to build, %0.2fs to run)' % (
endTime - startTime, buildTime - startTime, endTime - buildTime)
writeORFs(approved, suspect)
compareResults.compare(trueFile, 'predicted.txt')
print ''
def runIterativeMC(length):
''' Repeatedly train on the set of predicted ORFs, and use a fixed-length MM to predict a new set of ORFs.
Continue until the set of predicted ORFs does not change.
'''
print 'Iterative MC length %d:' % length
coding, noncoding = findLongORFs(genome)
oldORFs = []
i = 0
for x in xrange(10):
i += 1
print ' Iteration %d' % i
counts, immScores = init()
startTime = time.time()
print ' Training on %d coding, %d noncoding...' % (len(coding),
len(noncoding))
counts = buildMarkovChain(counts, coding, noncoding, length)
buildTime = time.time()
print ' Done, %0.2fs' % (buildTime - startTime)
print ' Calculating ORFs...'
approved, suspect = glimmer(genome, counts, immScores, length)
endTime = time.time()
print ' Done, %0.2fs' % (endTime - buildTime)
print ' Found %d approved, %d suspected' % (len(approved),
len(suspect))
writeORFs(approved, suspect)
compareResults.compare(trueFile, 'predicted.txt')
print ''
newORFs = approved + suspect
if noChange(oldORFs, newORFs):
print ' Converged'
break
else:
coding, noncoding = updateORFs(approved)
oldORFs = newORFs
print 'Completed after %d iterations' % i
def runIterativeMC(length):
''' Repeatedly train on the set of predicted ORFs, and use a fixed-length MM predict a new set of ORFs.
Continue until the set of predicted ORFs does not change.
'''
print 'Iterative MC length %d:' % length
coding, noncoding = findLongORFs(genome)
oldORFs = []
i = 0
for x in xrange(10):
i += 1
print ' Iteration %d' % i
counts, immScores = init()
startTime = time.time()
print ' Training on %d coding, %d noncoding...' % (len(coding), len(noncoding))
counts = buildMarkovChain(counts, coding, noncoding, length)
buildTime = time.time()
print ' Done, %0.2fs' % (buildTime-startTime)
print ' Calculating ORFs...'
approved, suspect = glimmer(genome, counts, immScores, length)
endTime = time.time()
print ' Done, %0.2fs' % (endTime-buildTime)
print ' Found %d approved, %d suspected' % (len(approved), len(suspect))
writeORFs(approved, suspect)
compareResults.compare('trueORFs.txt', 'predicted.txt')
print ''
newORFs = approved+suspect
if noChange(oldORFs, newORFs):
print ' Converged'
break
else:
coding, noncoding = updateORFs(approved)
oldORFs = newORFs
print 'Completed after %d iterations' % i