def run(self):
# If the files are in a sub-dir then rip them out.
if os.path.isdir(self.inputSequenceFileOrDirectory):
tempFile = getTempFile(rootDir=self.getGlobalTempDir())
catFiles(
[
os.path.join(self.inputSequenceFileOrDirectory, f)
for f in os.listdir(self.inputSequenceFileOrDirectory)
],
tempFile,
)
inputSequenceFile = tempFile
else:
inputSequenceFile = self.inputSequenceFileOrDirectory
assert inputSequenceFile != self.outputSequenceFile
prepXmlElems = self.configNode.findall("preprocessor")
analysisString = runCactusAnalyseAssembly(inputSequenceFile)
self.logToMaster(
"Before running any preprocessing on the assembly: %s got following stats (assembly may be listed as temp file if input sequences from a directory): %s"
% (self.inputSequenceFileOrDirectory, analysisString)
)
if len(prepXmlElems) == 0: # Just cp the file to the output file
system("cp %s %s" % (inputSequenceFile, self.outputSequenceFile))
else:
logger.info("Adding child batch_preprocessor target")
self.addChildTarget(BatchPreprocessor(prepXmlElems, inputSequenceFile, self.outputSequenceFile, 0))
def run(self, fileStore):
logger.info("Results IDs: %s" % self.resultsFileIDs)
resultsFiles = [readGlobalFileWithoutCache(fileStore, fileID) for fileID in self.resultsFileIDs]
collatedResultsFile = fileStore.getLocalTempFile()
catFiles(resultsFiles, collatedResultsFile)
logger.info("Collated the alignments to the file: %s", collatedResultsFile)
collatedResultsID = fileStore.writeGlobalFile(collatedResultsFile)
for resultsFileID in self.resultsFileIDs:
fileStore.deleteGlobalFile(resultsFileID)
return collatedResultsID
def testCompression(self):
tempSeqFile = os.path.join(self.tempDir, "tempSeq.fa")
tempSeqFile2 = os.path.join(self.tempDir, "tempSeq2.fa")
self.tempFiles.append(tempSeqFile)
self.tempFiles.append(tempSeqFile2)
self.encodePath = os.path.join(self.encodePath, "ENm001")
catFiles([ os.path.join(self.encodePath, fileName) for fileName in os.listdir(self.encodePath) ], tempSeqFile)
startTime = time.time()
compressFastaFile(tempSeqFile)
logger.critical("It took %s seconds to compress the fasta file" % (time.time() - startTime))
startTime = time.time()
system("rm %s" % tempSeqFile + ".bz2")
system("bzip2 --keep --fast %s" % tempSeqFile)
logger.critical("It took %s seconds to compress the fasta file by system functions" % (time.time() - startTime))
startTime = time.time()
decompressFastaFile(tempSeqFile + ".bz2", tempSeqFile2)
logger.critical("It took %s seconds to decompress the fasta file" % (time.time() - startTime))
system("rm %s" % tempSeqFile2)
startTime = time.time()
system("bunzip2 --stdout %s > %s" % (tempSeqFile + ".bz2", tempSeqFile2))
logger.critical("It took %s seconds to decompress the fasta file using system function" % (time.time() - startTime))
logger.critical("File sizes, before: %s, compressed: %s, decompressed: %s" % (os.stat(tempSeqFile).st_size, os.stat(tempSeqFile + ".bz2").st_size, os.stat(tempSeqFile2).st_size))
def testCompression(self):
tempSeqFile = os.path.join(self.tempDir, "tempSeq.fa")
tempSeqFile2 = os.path.join(self.tempDir, "tempSeq2.fa")
self.tempFiles.append(tempSeqFile)
self.tempFiles.append(tempSeqFile2)
self.encodePath = os.path.join(self.encodePath, "ENm001")
catFiles([
os.path.join(self.encodePath, fileName)
for fileName in os.listdir(self.encodePath)
], tempSeqFile)
startTime = time.time()
compressFastaFile(tempSeqFile)
logger.critical("It took %s seconds to compress the fasta file" %
(time.time() - startTime))
startTime = time.time()
system("rm %s" % tempSeqFile + ".bz2")
system("bzip2 --keep --fast %s" % tempSeqFile)
logger.critical(
"It took %s seconds to compress the fasta file by system functions"
% (time.time() - startTime))
startTime = time.time()
decompressFastaFile(tempSeqFile + ".bz2", tempSeqFile2)
logger.critical("It took %s seconds to decompress the fasta file" %
(time.time() - startTime))
system("rm %s" % tempSeqFile2)
startTime = time.time()
system("bunzip2 --stdout %s > %s" %
(tempSeqFile + ".bz2", tempSeqFile2))
logger.critical(
"It took %s seconds to decompress the fasta file using system function"
% (time.time() - startTime))
logger.critical(
"File sizes, before: %s, compressed: %s, decompressed: %s" %
(os.stat(tempSeqFile).st_size,
os.stat(tempSeqFile + ".bz2").st_size,
os.stat(tempSeqFile2).st_size))
def cat(target, genome, output_gtf, unsorted_tmp_file, out_file_tree):
"""
Concatenates all of the results into one big genePred, and sorts it by chromosome/pos
"""
catFiles(out_file_tree.listFiles(), unsorted_tmp_file)
system("sort -k1,1 -k4,4 {} > {}".format(unsorted_tmp_file, output_gtf))
def run(self):
catFiles(self.resultsFiles, self.finalResultsFile)
logger.info("Collated the alignments to the file: %s", self.finalResultsFile)
def cat(target, genome, file_tree, out_db):
tmp_file = os.path.join(target.getGlobalTempDir(), "tmp.txt")
catFiles(file_tree.listFiles(), tmp_file)
target.setFollowOnTargetFn(load_db, args=[genome, tmp_file, out_db])
def main():
##########################################
#Construct the arguments.
##########################################
usage = "usage: %prog [options] <query> <target> <output>\n\n" + \
" <query>: fasta sequence to search for repeats\n" + \
" <target>: fasta sequence to mask\n" + \
" <output>: softmasked version of <target>\n\n" + \
"Example: %prog genome.fa chunk.fa chunk.masked.fa\n\n"
description = "softrepeat mask a fasta file using lastz.\n" + \
"NOTE: NEW VERSION OF LASTZ REQUIRED (MINIMUM 1.02.40)\n" + \
"lastz tools/ dir MUST BE IN SYSTEM PATH"
parser = OptionParser(usage=usage, description=description)
#output stuff
parser.add_option("--fragment", dest="fragment", type="int",
help="The size of chunks passed to lastz (must be at least twice as big as overlap)",
default=200)
parser.add_option("--minPeriod", dest="period", type="int",
help="minimum number of occurrences of a sequence for it to be masked",
default=10)
parser.add_option("--lastzOpts", dest="lastzOptions", type="string",
help="lastz options for repeat identification",
default="")
parser.add_option("--lastzCmd", dest="lastzCmd", type="string",
help="lastz executable",
default="cPecanLastz")
parser.add_option("--unmaskInput", dest="unmaskInput", action="store_true",
help="Makes any previous masking of the input sequence invisible to the repeat masking process",
default=False)
parser.add_option("--unmaskOutput", dest="unmaskOutput", action="store_true",
help="Discards any previous masking from the output sequence, uses just the masking discovered by lastz",
default=False)
parser.add_option("--proportionSampled", dest="proportionSampled", type="float",
help="The amount of the genome that is being sampled for masking, used to adjust the minPeriod parameter according to sampling",
default="1.0")
parser.add_option("--tempDir", dest="tempDir",
help="Location in which to place to temporary files",
default=os.getcwd())
options, args = parser.parse_args()
if len(args) != 2:
parser.print_help()
return 1
queryFile = args[0]
outputFile = args[1]
targetFiles = sys.stdin.readline().split() #Read them from stdin
assert os.path.isfile(queryFile)
assert len(targetFiles) >= 1
for targetFile in targetFiles:
assert os.path.isfile(targetFile)
assert options.fragment > 1
#Adjust the period parameter using the amount of genome sampled
options.period = max(1, round(options.proportionSampled * options.period))
# make sure fragment size is even so they can overlap by exactly one half.
if options.fragment % 2:
options.fragment += 1
# make temporary working directory in same path as output
tempDir = tempfile.mkdtemp(dir=options.tempDir)
maskInfoFile = os.path.join(tempDir, "maskFile.dat")
targetFile = os.path.join(tempDir, "target.fa")
try:
#Make temporary target file, if more than one file
catFiles(targetFiles, targetFile)
# chop up input fasta file into into fragments of specified size. fragments overlap by
# half their length.
fragCmdLine = 'cat ' + queryFile + ' | cactus_fasta_fragments.py ' + '--fragment=' + \
str(options.fragment) + ' --step=' + str(options.fragment / 2) + " --origin=zero "
# lastz each fragment against the entire input sequence. Each time a fragment aligns to a base
# in the sequence, that base's match count is incremented.
# the plus three for the period parameter is a fudge to ensure sufficient alignments are found
lastZSequenceHandling = '[multiple][nameparse=darkspace] /dev/stdin[nameparse=darkspace] '
if options.unmaskInput:
lastZSequenceHandling = '[multiple,unmask][nameparse=darkspace] /dev/stdin[unmask][nameparse=darkspace] '
lastzCmdLine = options.lastzCmd + ' ' + targetFile + \
lastZSequenceHandling + options.lastzOptions + \
(' --querydepth=keep,nowarn:%i --format=general:name1,zstart1,end1,name2,zstart2+,end2+ --markend ' % \
(options.period+3))
#This runs Bob's covered intervals program, which combins the lastz alignment info into intervals of the query.
coveredIntervalsCmdLine = "cactus_covered_intervals --queryoffsets --origin=one M=%s > %s" % (int(options.period*2), maskInfoFile)
system(fragCmdLine + ' | ' + lastzCmdLine + ' | ' + coveredIntervalsCmdLine)
#open(maskInfoFile, "w").close()
# the previous lastz command outputs a file of intervals (denoted with indices) to softmask.
# we finish by applying these intervals to the input file, to produce the final, softmasked output.
unmaskString = ""
if options.unmaskOutput:
unmaskString = "--unmask"
softMaskCmdLine = 'cat ' + queryFile + (' | cactus_fasta_softmask_intervals.py --origin=one %s ' % unmaskString) + \
maskInfoFile + ' > ' + outputFile
system(softMaskCmdLine)
except Exception, e:
# delete the temporary files
shutil.rmtree(tempDir)
raise e
def cat(target, genome, output_gtf, unsorted_tmp_file, out_file_tree):
"""
Concatenates all of the results into one big genePred, and sorts it by chromosome/pos
"""
catFiles(out_file_tree.listFiles(), unsorted_tmp_file)
system("sort -k1,1 -k4,4 {} > {}".format(unsorted_tmp_file, output_gtf))
