def main():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
basFilename = args[0]
logger.debug("bas file: %s" % basFilename)
bf = H5BasFile.BasFile(basFilename)
try:
hole = int(args[1])
except ValueError:
logger.error('ERROR: second parameter must be an integer ZMW number')
sys.exit()
if opt.subreads:
for region in bf.holeRegions(hole):
regionHole, regionType, start, end, score = region
if regionType == 1: # a subread?
printRange(bf, hole, opt, start, end)
else:
printRange(bf, hole, opt, opt.start, opt.end)
logger.debug("complete")
def main ():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
basFilename = args[0]
logger.debug("bas file: %s" % basFilename)
basfile = H5BasFile.BasFile (basFilename)
try:
hole = int(args[1])
except ValueError:
logger.error('ERROR: second parameter must be an integer ZMW number')
sys.exit()
if not opt.reverse:
sequence = basfile.getSequence(hole, opt.start, opt.end) # end==None gets the whole read
else:
sequence = basfile.getRevCompSequence(hole, opt.start, opt.end)
movie = basfile.movieName()
length = len(sequence)
print ">%s/%d/%d_%d" % (movie, hole, opt.start, opt.start+length)
for ix in xrange(0,length,opt.flen):
print sequence[ix:ix+opt.flen]
logger.debug("complete")
def main ():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
basFilename = args[0]
logger.debug("bas file: %s" % basFilename)
bf = H5BasFile.BasFile (basFilename)
try:
hole = int(args[1])
except ValueError:
logger.error('ERROR: second parameter must be an integer ZMW number')
sys.exit()
if opt.subreads:
for region in bf.holeRegions(hole):
regionHole, regionType, start, end, score = region
if regionType == 1: # a subread?
printRange (bf, hole, opt, start, end)
else:
printRange (bf, hole, opt, opt.start, opt.end)
logger.debug("complete")
def submitFinalJobs (opt, chunkList):
chunkFiles = ['%s \\\n' % chk.trimmedChunkName for chk in chunkList]
sh = list()
sh.append('#!/bin/bash\n\n')
sh.append('set -o errexit\n')
sh.append('set -o nounset\n\n')
sh.append('cat \\\n')
sh.extend(chunkFiles)
sh.append(' > %s\n' % opt.output)
if opt.report is not None:
reportFiles = ['%s \\\n' % chk.reportChunkName for chk in chunkList]
sh.append('\ncat \\\n')
sh.extend(reportFiles)
sh.append(' > %s\n' % opt.report)
finalScriptName = '%s/trim_final.sh' % opt.tmpdir
handle = open (finalScriptName, 'w')
handle.writelines (sh)
handle.close()
deps = ':'.join ([chk.jobno for chk in chunkList])
cmd = list()
cmd.append('qsub')
cmd.append('-N trim_final') # job name
cmd.append('-o trim_final.out') # output file
cmd.append('-j oe') # combine stdout and stderr
cmd.append('-l nodes=1:ppn=1,walltime=4:00:00') # resources required
cmd.append('-d . ') # working directory (strangely, ./ is not the default)
cmd.append('-r n') # do NOT attempt to restart on failure
cmd.append('-V') # export all environment variables to job
cmd.append('-W umask=0002') # make logs rw-rw-r--
cmd.append('-m n') # don't send any mail
cmd.append('-W depend=afterok:%s' % deps)
cmd.append(finalScriptName) # script to run
command = ' '.join(cmd)
logger.debug ('running %s' % command)
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error('command failed, rc=%d' % rc)
raise RuntimeError
logger.debug ('jobno is %s' % response)
return response
def submitFinalJobs(opt, chunkList):
chunkFiles = ['%s \\\n' % chk.trimmedChunkName for chk in chunkList]
sh = list()
sh.append('#!/bin/bash\n\n')
sh.append('set -o errexit\n')
sh.append('set -o nounset\n\n')
sh.append('cat \\\n')
sh.extend(chunkFiles)
sh.append(' > %s\n' % opt.output)
if opt.report is not None:
reportFiles = ['%s \\\n' % chk.reportChunkName for chk in chunkList]
sh.append('\ncat \\\n')
sh.extend(reportFiles)
sh.append(' > %s\n' % opt.report)
finalScriptName = '%s/trim_final.sh' % opt.tmpdir
handle = open(finalScriptName, 'w')
handle.writelines(sh)
handle.close()
deps = ':'.join([chk.jobno for chk in chunkList])
cmd = list()
cmd.append('qsub')
cmd.append('-N trim_final') # job name
cmd.append('-o trim_final.out') # output file
cmd.append('-j oe') # combine stdout and stderr
cmd.append('-l nodes=1:ppn=1,walltime=4:00:00') # resources required
cmd.append('-d . ') # working directory (strangely, ./ is not the default)
cmd.append('-r n') # do NOT attempt to restart on failure
cmd.append('-V') # export all environment variables to job
cmd.append('-W umask=0002') # make logs rw-rw-r--
cmd.append('-m n') # don't send any mail
cmd.append('-W depend=afterok:%s' % deps)
cmd.append(finalScriptName) # script to run
command = ' '.join(cmd)
logger.debug('running %s' % command)
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error('command failed, rc=%d' % rc)
raise RuntimeError
logger.debug('jobno is %s' % response)
return response
def countSeqs (filename):
'''Run grep -c ">" on a fasta file to count the sequences it contains.'''
command = 'grep -c ">" %s' % filename
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error("command failed, rc=%d" % rc)
raise RuntimeError
if not response.isdigit():
logger.error("grep -c returned:" % response)
raise RuntimeError
return int(response)
def countSeqs(filename):
'''Run grep -c ">" on a fasta file to count the sequences it contains.'''
command = 'grep -c ">" %s' % filename
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error("command failed, rc=%d" % rc)
raise RuntimeError
if not response.isdigit():
logger.error("grep -c returned:" % response)
raise RuntimeError
return int(response)
def main ():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
if len(args) != 2:
logger.error ("please specify bas.h5 and cmp.h5 files as parameters. See --help")
sys.exit()
# TODO: Actually, all we need from the bas file is the movie name
# (maxHole will default to something clever). We don't need to
# open the bas file to determine the movie name: It's part of the
# filename. The only real reason we specify a bas file as the
# first parameter is to match the command line interface of other
# scripts.
basFilename = args[0]
logger.debug("bas file: %s" % basFilename)
bf = H5BasFile.BasFile (basFilename)
movie = bf.movieName()
cmpFilename = args[1]
logger.debug("cmp file: %s" % cmpFilename)
cf = H5CmpFile.CmpFile (fileName=cmpFilename)
cmp = H5CmpFile.CmpMovie (cmpObject=cf,
movieName=movie,
maxHole=bf.maxZMW())
if opt.ZMW is not None: # did we ask for a specific ZMW?
for align in cmp.getAlignmentsForHole(opt.ZMW):
printAlignment (align, cmp, opt.flen)
else: # else, print all ZMWs
for align in cmp.getAlignmentsByHole():
printAlignment (align, cmp, opt.flen)
logger.debug("complete")
def submitScript(self):
# Dependent job submission will fail if parent has already
# completed. So delay all job startups by a short amount of time.
startAt = datetime.datetime.now() + datetime.timedelta(0, STARTWAIT)
startAtStr = startAt.strftime('%Y%m%d%H%M.%S')
cmd = list()
cmd.append('qsub')
cmd.append('-N %s' % self.jobName) # job name
cmd.append('-o %s' % self.scriptOutput) # output file
cmd.append('-j oe') # combine stdout and stderr
cmd.append('-l nodes=1:ppn=1,walltime=4:00:00') # resources required
cmd.append('-a %s' % startAtStr) # delay start, see above
cmd.append(
'-d . ') # working directory (strangely, ./ is not the default)
cmd.append('-r n') # do NOT attempt to restart on failure
cmd.append('-V') # export all environment variables to job
cmd.append('-W umask=0002') # make logs rw-rw-r--
cmd.append('-m n') # don't send any mail
cmd.append(self.scriptName) # script to run
command = ' '.join(cmd)
logger.debug('running %s' % command)
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error('command failed, rc=%d' % rc)
raise RuntimeError
match = re.match(Chunk.JOBNO_PATTERN, response)
if match is None:
logger.error("invalid job sequence number: %s" % jobSeqStr)
raise RuntimeError
response = match.group(1)
logger.debug('jobno is %s' % response)
self.jobno = response
return response
def submitScript (self):
# Dependent job submission will fail if parent has already
# completed. So delay all job startups by a short amount of time.
startAt = datetime.datetime.now() + datetime.timedelta(0, STARTWAIT)
startAtStr = startAt.strftime('%Y%m%d%H%M.%S')
cmd = list()
cmd.append('qsub')
cmd.append('-N %s' % self.jobName) # job name
cmd.append('-o %s' % self.scriptOutput) # output file
cmd.append('-j oe') # combine stdout and stderr
cmd.append('-l nodes=1:ppn=1,walltime=4:00:00') # resources required
cmd.append('-a %s' % startAtStr) # delay start, see above
cmd.append('-d . ') # working directory (strangely, ./ is not the default)
cmd.append('-r n') # do NOT attempt to restart on failure
cmd.append('-V') # export all environment variables to job
cmd.append('-W umask=0002') # make logs rw-rw-r--
cmd.append('-m n') # don't send any mail
cmd.append(self.scriptName) # script to run
command = ' '.join(cmd)
logger.debug ('running %s' % command)
popen_file = os.popen(command)
response = popen_file.read().strip()
rc = popen_file.close()
if rc is not None:
logger.error('command failed, rc=%d' % rc)
raise RuntimeError
match = re.match (Chunk.JOBNO_PATTERN, response)
if match is None:
logger.error("invalid job sequence number: %s" % jobSeqStr)
raise RuntimeError
response = match.group(1)
logger.debug ('jobno is %s' % response)
self.jobno = response
return response
def main ():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
if len(args) != 2:
logger.error ("please specify bas.h5 and cmp.h5 files as parameters. See --help")
sys.exit()
basFilename = args[0]
logger.debug("bas file: %s" % basFilename)
bf = H5BasFile.BasFile (basFilename)
cmpFilename = args[1]
logger.debug("cmp file: %s" % cmpFilename)
cf = H5CmpFile.CmpFile (fileName=cmpFilename)
cmp = H5CmpFile.CmpMovie (cmpObject=cf,
movieName=bf.movieName(),
maxHole=bf.maxZMW())
cf.printDetails()
print " AlnID RG Hole Set Stb SubRd Seq Ref St Start End RefStrt RefEnd OffStrt OffEnd"
print
if opt.sort == 'hole':
for align in cmp.getAlignmentsByHole():
printAlign(align)
else : # else, must be 'none'
for align in cmp.getAllAlignments(): # generator function, returns a dict
printAlign(align)
logger.debug("complete")
def main ():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
basFilename = args[0]
basfile = H5BasFile.BasFile (basFilename)
try:
hole = int(args[1])
except ValueError:
logger.error('ERROR: second parameter must be an integer ZMW number')
sys.exit()
start = opt.start
end = basfile.readLen(hole) if opt.end is None else opt.end
sequence = basfile.getSequence(hole, start, end)
aln = SWAligner.Aligner()
aln.setRef (sequence)
aln.setRead (H5BasFile.ADAPTER)
aln.fillMatrix()
allScores = aln.allScores()
range = xrange(start,end)
title = "ZMW %d (%d to %d)" % (hole, start, end)
plt.suptitle(title, fontsize=14, fontweight='bold')
plt.plot (range, allScores, COL_NOT_HQ, zorder=1, label='non-HQ')
if not opt.nocol: # finding HQ region takes a long time, so optionally turn it off
# There doesn't seem to be a way to separately specify a
# colour for each point in a plot. So we'll plot in one
# colour, then overlay subregions of that with another
# colour. Plot commands are rendered in increasing zorder.
HQStart, HQEnd = basfile.HQregion(hole)[2:4]
HQRange = xrange(HQStart, HQEnd)
HQScores = allScores[HQStart:HQEnd]
plt.plot (HQRange, HQScores, COL_HQ, zorder=2, label='HQ')
label = 'adapter'; # I will only say this once...
for region in basfile.holeRegions(hole): # loop through the regions looking for adapters
regionType, regionStart, regionEnd = region[1:4]
if regionType == 0: # an adapter?
regionRange = xrange(regionStart, regionEnd)
regionScores = allScores[regionStart:regionEnd]
plt.plot (regionRange, regionScores, COL_ADAPT, zorder=3, label=label)
label = '_nolegend_' # don't generate multiple legend entries
plt.legend(loc='best', prop={'size':10}) # add a legend box to figure
plt.ylim (0, len(H5BasFile.ADAPTER))
if opt.output is not None:
outfile = opt.output
else:
outfile = "ZMW-%05d.png" % hole
plt.savefig (outfile)
logger.debug("complete")
def main():
logger.debug("%s starting" % sys.argv[0])
opt, args = getParms()
basFilename = args[0]
basfile = H5BasFile.BasFile(basFilename)
try:
hole = int(args[1])
except ValueError:
logger.error('ERROR: second parameter must be an integer ZMW number')
sys.exit()
start = opt.start
end = basfile.readLen(hole) if opt.end is None else opt.end
sequence = basfile.getSequence(hole, start, end)
aln = SWAligner.Aligner()
aln.setRef(sequence)
aln.setRead(H5BasFile.ADAPTER)
aln.fillMatrix()
allScores = aln.allScores()
range = xrange(start, end)
title = "ZMW %d (%d to %d)" % (hole, start, end)
plt.suptitle(title, fontsize=14, fontweight='bold')
plt.plot(range, allScores, COL_NOT_HQ, zorder=1, label='non-HQ')
if not opt.nocol: # finding HQ region takes a long time, so optionally turn it off
# There doesn't seem to be a way to separately specify a
# colour for each point in a plot. So we'll plot in one
# colour, then overlay subregions of that with another
# colour. Plot commands are rendered in increasing zorder.
HQStart, HQEnd = basfile.HQregion(hole)[2:4]
HQRange = xrange(HQStart, HQEnd)
HQScores = allScores[HQStart:HQEnd]
plt.plot(HQRange, HQScores, COL_HQ, zorder=2, label='HQ')
label = 'adapter'
# I will only say this once...
for region in basfile.holeRegions(
hole): # loop through the regions looking for adapters
regionType, regionStart, regionEnd = region[1:4]
if regionType == 0: # an adapter?
regionRange = xrange(regionStart, regionEnd)
regionScores = allScores[regionStart:regionEnd]
plt.plot(regionRange,
regionScores,
COL_ADAPT,
zorder=3,
label=label)
label = '_nolegend_' # don't generate multiple legend entries
plt.legend(loc='best', prop={'size': 10}) # add a legend box to figure
plt.ylim(0, len(H5BasFile.ADAPTER))
if opt.output is not None:
outfile = opt.output
else:
outfile = "ZMW-%05d.png" % hole
plt.savefig(outfile)
logger.debug("complete")
