def __init__(self, reads, blastFilenames, databaseFilename=None,
databaseDirectory=None, sqliteDatabaseFilename=None,
scoreClass=HigherIsBetterScore,
sortBlastFilenames=True, randomizeZeroEValues=True):
if type(blastFilenames) == str:
blastFilenames = [blastFilenames]
if sortBlastFilenames:
self.blastFilenames = numericallySortFilenames(blastFilenames)
else:
self.blastFilenames = blastFilenames
self._databaseFilename = databaseFilename
self._sqliteDatabaseFilename = sqliteDatabaseFilename
self._databaseDirectory = databaseDirectory
self._subjectTitleToSubject = None
self.randomizeZeroEValues = randomizeZeroEValues
# Prepare application parameters in order to initialize self.
self._reader = self._getReader(self.blastFilenames[0], scoreClass)
application = self._reader.application
blastParams = copy.deepcopy(self._reader.params)
subjectIsNucleotides = application != 'blastx'
scoreTitle = ('Bit score' if scoreClass is HigherIsBetterScore
else '$- log_{10}(e)$')
applicationParams = ReadsAlignmentsParams(
application, blastParams,
subjectIsNucleotides=subjectIsNucleotides, scoreTitle=scoreTitle)
ReadsAlignments.__init__(self, reads, applicationParams,
scoreClass=scoreClass)
def __init__(self,
reads,
blastFilenames,
scoreClass=HigherIsBetterScore,
sortBlastFilenames=True,
randomizeZeroEValues=True):
if type(blastFilenames) == str:
blastFilenames = [blastFilenames]
if sortBlastFilenames:
self.blastFilenames = numericallySortFilenames(blastFilenames)
else:
self.blastFilenames = blastFilenames
self.randomizeZeroEValues = randomizeZeroEValues
# Prepare application parameters in order to initialize self.
self._reader = self._getReader(self.blastFilenames[0], scoreClass)
application = self._reader.application
blastParams = copy.deepcopy(self._reader.params)
subjectIsNucleotides = application != 'blastx'
scoreTitle = ('Bit score' if scoreClass is HigherIsBetterScore else
'$- log_{10}(e)$')
applicationParams = ReadsAlignmentsParams(
application,
blastParams,
subjectIsNucleotides=subjectIsNucleotides,
scoreTitle=scoreTitle)
ReadsAlignments.__init__(self,
reads,
applicationParams,
scoreClass=scoreClass)
def __init__(self, resultFilenames, database, sortFilenames=True):
if isinstance(resultFilenames, str):
resultFilenames = [resultFilenames]
if sortFilenames:
self.resultFilenames = numericallySortFilenames(resultFilenames)
else:
self.resultFilenames = resultFilenames
self._database = database
# Add a dictionary that will allow us to look up database subjects
# by title.
self._subjects = dict((subject.read.id, subject.read.sequence)
for subject in database.getSubjects())
# Prepare application parameters in order to initialize self.
self._reader = self._getReader(self.resultFilenames[0])
params = copy.deepcopy(self._reader.params)
applicationParams = ReadsAlignmentsParams('light',
applicationParams=params,
subjectIsNucleotides=False)
# We will add to self._reads as we go through the results (which
# contain the read ids and sequences). Note that this means the
# reads will not be available to the ReadsAlignments instance until
# after all results have been read.
self._reads = Reads()
ReadsAlignments.__init__(self, self._reads, applicationParams)
def __init__(self, reads, filenames, databaseFilename=None,
databaseDirectory=None, sqliteDatabaseFilename=None,
scoreClass=HigherIsBetterScore, sortFilenames=False,
randomizeZeroEValues=True):
if type(filenames) == str:
filenames = [filenames]
if sortFilenames:
self.filenames = numericallySortFilenames(filenames)
else:
self.filenames = filenames
self._databaseFilename = databaseFilename
self._sqliteDatabaseFilename = sqliteDatabaseFilename
self._databaseDirectory = databaseDirectory
self._subjectTitleToSubject = None
self.randomizeZeroEValues = randomizeZeroEValues
# Prepare diamondTask parameters in order to initialize self.
self._reader = self._getReader(self.filenames[0], scoreClass)
diamondTask = self._reader.diamondTask
diamondParams = copy.deepcopy(self._reader.params)
scoreTitle = ('Bit score' if scoreClass is HigherIsBetterScore
else '$- log_{10}(e)$')
diamondTaskParams = ReadsAlignmentsParams(
diamondTask, diamondParams,
subjectIsNucleotides=False, # DIAMOND dbs are always protein.
scoreTitle=scoreTitle)
ReadsAlignments.__init__(self, reads, diamondTaskParams,
scoreClass=scoreClass)
def testSeveralNames(self):
"""
A list with several numeric names should result in a correctly
sorted list of names being returned.
"""
self.assertEqual(['1.json', '2.json', '3.json'],
numericallySortFilenames(
['3.json', '1.json', '2.json']))
def testSeveralNames(self):
"""
A list with several numeric names should result in a correctly
sorted list of names being returned.
"""
self.assertEqual(
['1.json', '2.json', '3.json'],
numericallySortFilenames(['3.json', '1.json', '2.json']))
def testSeveralNamesWithUnequalPrefixLengths(self):
"""
A list with several numeric names whose numeric prefixes differ
in length should result in a correctly sorted list of names being
returned.
"""
self.assertEqual(
['2.json', '3.json', '21.json', '35.json', '250.json'],
numericallySortFilenames(
['3.json', '21.json', '35.json', '250.json', '2.json']))
def testBasename(self):
"""
Sorting must be according to file basename.
"""
self.assertEqual(
['../output/2.json', '../output/3.json', '../output/21.json',
'../output/35.json', '../output/250.json'],
numericallySortFilenames(
['../output/3.json', '../output/21.json', '../output/35.json',
'../output/250.json', '../output/2.json']))
def testSeveralNamesWithUnequalPrefixLengths(self):
"""
A list with several numeric names whose numeric prefixes differ
in length should result in a correctly sorted list of names being
returned.
"""
self.assertEqual(
['2.json', '3.json', '21.json', '35.json', '250.json'],
numericallySortFilenames(
['3.json', '21.json', '35.json', '250.json', '2.json']))
def testBasename(self):
"""
Sorting must be according to file basename.
"""
self.assertEqual(
['../output/2.json', '../output/3.json', '../output/21.json',
'../output/35.json', '../output/250.json'],
numericallySortFilenames(
['../output/3.json', '../output/21.json', '../output/35.json',
'../output/250.json', '../output/2.json']))
def testOneNumericName(self):
"""
A list with a single numeric name should result in that same
name being returned.
"""
self.assertEqual(['3.json'], numericallySortFilenames(['3.json']))
def testNoNames(self):
"""
An empty list must be returned when an empty list is given.
"""
self.assertEqual([], numericallySortFilenames([]))
def testOneNumericName(self):
"""
A list with a single numeric name should result in that same
name being returned.
"""
self.assertEqual(['3.json'], numericallySortFilenames(['3.json']))
def testNoNames(self):
"""
An empty list must be returned when an empty list is given.
"""
self.assertEqual([], numericallySortFilenames([]))
# Flatten lists of lists that we get from using both nargs='+' and
# action='append'. We use both because it allows people to use (e.g.)
# --json on the command line either via "--json file1 --json file2" or
# "--json file1 file2", or a combination of these. That way it's not
# necessary to remember which way you're supposed to use it and you also
# can't be hit by the subtle problem encountered in
# https://github.com/acorg/dark-matter/issues/453
jsonFiles = list(chain.from_iterable(args.json))
whitelist = (
set(chain.from_iterable(args.whitelist)) if args.whitelist else None)
blacklist = (
set(chain.from_iterable(args.blacklist)) if args.blacklist else None)
if args.fasta:
if args.sortFilenames:
files = numericallySortFilenames(chain.from_iterable(args.fasta))
else:
files = list(chain.from_iterable(args.fasta))
reads = FastaReads(files)
else:
if args.sortFilenames:
files = numericallySortFilenames(chain.from_iterable(args.fastq))
else:
files = list(chain.from_iterable(args.fastq))
reads = FastqReads(files)
if args.matcher == 'blast':
from dark.blast.alignments import BlastReadsAlignments
readsAlignments = BlastReadsAlignments(
reads, jsonFiles, databaseFilename=args.databaseFastaFilename,
databaseDirectory=args.databaseFastaDirectory,
# --json on the command line either via "--json file1 --json file2" or
# "--json file1 file2", or a combination of these. That way it's not
# necessary to remember which way you're supposed to use it and you also
# can't be hit by the subtle problem encountered in
# https://github.com/acorg/dark-matter/issues/453
jsonFiles = list(chain.from_iterable(args.json))
whitelist = (
set(chain.from_iterable(args.whitelist)) if args.whitelist else None)
blacklist = (
set(chain.from_iterable(args.blacklist)) if args.blacklist else None)
# TODO: Add a --readClass command-line option in case we want to
# process FASTA containing AA sequences.
if args.fasta:
if args.sortFilenames:
files = numericallySortFilenames(chain.from_iterable(args.fasta))
else:
files = list(chain.from_iterable(args.fasta))
reads = FastaReads(files)
else:
if args.sortFilenames:
files = numericallySortFilenames(chain.from_iterable(args.fastq))
else:
files = list(chain.from_iterable(args.fastq))
reads = FastqReads(files)
if args.matcher == 'blast':
from dark.blast.alignments import BlastReadsAlignments
readsAlignments = BlastReadsAlignments(
reads, jsonFiles, databaseFilename=args.databaseFastaFilename,
databaseDirectory=args.databaseFastaDirectory,
