def testBlastRandom(self):
"""Make some sequences, put them in a file, call blast with random parameters
and check it runs okay.
"""
tempSeqFile = os.path.join(self.tempDir, "tempSeq.fa")
self.tempFiles.append(tempSeqFile)
for test in xrange(self.testNo):
seqNo = random.choice(xrange(0, 10))
seq = getRandomSequence(8000)[1]
fileHandle = open(tempSeqFile, 'w')
for fastaHeader, seq in [(str(i),
mutateSequence(seq,
0.3 * random.random()))
for i in xrange(seqNo)]:
if random.random() > 0.5:
seq = reverseComplement(seq)
fastaWrite(fileHandle, fastaHeader, seq)
fileHandle.close()
chunkSize = random.choice(xrange(500, 9000))
overlapSize = random.choice(xrange(2, 100))
toilDir = os.path.join(getTempDirectory(self.tempDir), "toil")
runCactusBlast([tempSeqFile], self.tempOutputFile, toilDir,
chunkSize, overlapSize)
#runToilStatusAndFailIfNotComplete(toilDir)
if getLogLevelString() == "DEBUG":
system("cat %s" % self.tempOutputFile)
system("rm -rf %s " % toilDir)
def testBlastRandom(self):
"""Make some sequences, put them in a file, call blast with random parameters
and check it runs okay.
"""
tempSeqFile = os.path.join(self.tempDir, "tempSeq.fa")
self.tempFiles.append(tempSeqFile)
for test in xrange(self.testNo):
seqNo = random.choice(xrange(0, 10))
seq = getRandomSequence(8000)[1]
fileHandle = open(tempSeqFile, 'w')
for fastaHeader, seq in [ (str(i), mutateSequence(seq, 0.3*random.random())) for i in xrange(seqNo) ]:
if random.random() > 0.5:
seq = reverseComplement(seq)
fastaWrite(fileHandle, fastaHeader, seq)
fileHandle.close()
chunkSize = random.choice(xrange(500, 9000))
overlapSize = random.choice(xrange(2, 100))
toilDir = os.path.join(getTempDirectory(self.tempDir), "toil")
runCactusBlast([ tempSeqFile ], self.tempOutputFile, toilDir, chunkSize, overlapSize)
#runToilStatusAndFailIfNotComplete(toilDir)
if getLogLevelString() == "DEBUG":
system("cat %s" % self.tempOutputFile)
system("rm -rf %s " % toilDir)
def getCactusInputs_random(regionNumber=0,
tempDir=None,
sequenceNumber=None,
avgSequenceLength=None,
treeLeafNumber=None):
"""Gets a random set of sequences, each of length given, and a species
tree relating them. Each sequence is a assigned an event in this tree.
"""
if sequenceNumber is None:
sequenceNumber = random.choice(list(range(30)))
if avgSequenceLength is None:
avgSequenceLength = random.choice(list(range(1, 3000)))
if treeLeafNumber is None:
treeLeafNumber = random.choice(list(range(2, 4)))
#Make tree
binaryTree = makeRandomBinaryTree(treeLeafNumber)
newickTreeString = printBinaryTree(binaryTree, includeDistances=True)
newickTreeLeafNames = []
def fn(tree):
if tree.internal:
fn(tree.left)
fn(tree.right)
else:
newickTreeLeafNames.append(tree.iD)
fn(binaryTree)
logger.info("Made random binary tree: %s" % newickTreeString)
sequenceDirs = []
for i in range(len(newickTreeLeafNames)):
seqDir = getTempDirectory(rootDir=tempDir)
sequenceDirs.append(seqDir)
logger.info("Made a set of random directories: %s" %
" ".join(sequenceDirs))
#Random sequences and species labelling
sequenceFile = None
fileHandle = None
parentSequence = getRandomSequence(
length=random.choice(list(range(1, 2 * avgSequenceLength))))[1]
emptySequenceDirs = set(sequenceDirs)
i = 0
while i < sequenceNumber or len(emptySequenceDirs) > 0:
if sequenceFile == None:
if random.random(
) > 0.5: #Randomly choose the files to be attached or not
suffix = ".fa.complete"
else:
suffix = ".fa"
sequenceDir = random.choice(sequenceDirs)
if sequenceDir in emptySequenceDirs:
emptySequenceDirs.remove(sequenceDir)
sequenceFile = getTempFile(rootDir=sequenceDir, suffix=suffix)
fileHandle = open(sequenceFile, 'w')
if random.random() > 0.8: #Get a new root sequence
parentSequence = getRandomSequence(
length=random.choice(list(range(1, 2 * avgSequenceLength))))[1]
sequence = mutateSequence(parentSequence,
distance=random.random() * 0.25)
name = getRandomAlphaNumericString(15)
if random.random() > 0.5:
sequence = reverseComplement(sequence)
fastaWrite(fileHandle, name, sequence)
if random.random() > 0.5:
fileHandle.close()
fileHandle = None
sequenceFile = None
i += 1
if fileHandle != None:
fileHandle.close()
logger.info("Made %s sequences in %s directories" %
(sequenceNumber, len(sequenceDirs)))
return sequenceDirs, newickTreeString
def getCactusInputs_random(regionNumber=0, tempDir=None,
sequenceNumber=None,
avgSequenceLength=None,
treeLeafNumber=None):
"""Gets a random set of sequences, each of length given, and a species
tree relating them. Each sequence is a assigned an event in this tree.
"""
if sequenceNumber is None:
sequenceNumber = random.choice(xrange(30))
if avgSequenceLength is None:
avgSequenceLength = random.choice(xrange(1,3000))
if treeLeafNumber is None:
treeLeafNumber = random.choice(xrange(2, 4))
#Make tree
binaryTree = makeRandomBinaryTree(treeLeafNumber)
newickTreeString = printBinaryTree(binaryTree, includeDistances=True)
newickTreeLeafNames = []
def fn(tree):
if tree.internal:
fn(tree.left)
fn(tree.right)
else:
newickTreeLeafNames.append(tree.iD)
fn(binaryTree)
logger.info("Made random binary tree: %s" % newickTreeString)
sequenceDirs = []
for i in xrange(len(newickTreeLeafNames)):
seqDir = getTempDirectory(rootDir=tempDir)
sequenceDirs.append(seqDir)
logger.info("Made a set of random directories: %s" % " ".join(sequenceDirs))
#Random sequences and species labelling
sequenceFile = None
fileHandle = None
parentSequence = getRandomSequence(length=random.choice(xrange(1, 2*avgSequenceLength)))[1]
emptySequenceDirs = set(sequenceDirs)
i = 0
while i < sequenceNumber or len(emptySequenceDirs) > 0:
#for i in xrange(sequenceNumber):
if sequenceFile == None:
if random.random() > 0.5: #Randomly choose the files to be attached or not
suffix = ".fa.complete"
else:
suffix = ".fa"
sequenceDir = random.choice(sequenceDirs)
if sequenceDir in emptySequenceDirs:
emptySequenceDirs.remove(sequenceDir)
sequenceFile = getTempFile(rootDir=sequenceDir, suffix=suffix)
fileHandle = open(sequenceFile, 'w')
if random.random() > 0.8: #Get a new root sequence
parentSequence = getRandomSequence(length=random.choice(xrange(1, 2*avgSequenceLength)))[1]
sequence = mutateSequence(parentSequence, distance=random.random()*0.5)
name = getRandomAlphaNumericString(15)
if random.random() > 0.5:
sequence = reverseComplement(sequence)
fastaWrite(fileHandle, name, sequence)
if random.random() > 0.5:
fileHandle.close()
fileHandle = None
sequenceFile = None
i += 1
if fileHandle != None:
fileHandle.close()
logger.info("Made %s sequences in %s directories" % (sequenceNumber, len(sequenceDirs)))
return sequenceDirs, newickTreeString