def saveLocal(self, name):
'''
Save processed SAM file info to a local compressed pickle file.
'name' can exclude extension: .baga will be added
'''
fileout = 'baga.AlignReads.SAMs-{}.baga'.format(name)
with _tarfile.open(fileout, "w:gz") as tar:
print('Writing to {} . . . '.format(fileout))
for att_name, att in self.__dict__.items():
if isinstance(att, _array):
io = _StringIO(att.tostring())
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name = att_name)
thisone.size = length
tar.addfile(tarinfo = thisone, fileobj = io)
else:
# try saving everything else here by jsoning
try:
io = _StringIO()
_json.dump(att, io)
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name = att_name)
thisone.size = length
tar.addfile(tarinfo = thisone, fileobj = io)
except TypeError:
# ignore non-jsonable things like functions
# include unicodes, strings, lists etc etc
#print('omitting {}'.format(att_name))
pass
def saveLocal(self, name):
'''
Save processed SAM file info to a local compressed pickle file.
'name' can exclude extension: .baga will be added
'''
fileout = 'baga.AlignReads.SAMs-{}.baga'.format(name)
with _tarfile.open(fileout, "w:gz") as tar:
print('Writing to {} . . . '.format(fileout))
for att_name, att in self.__dict__.items():
if isinstance(att, _array):
io = _StringIO(att.tostring())
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name=att_name)
thisone.size = length
tar.addfile(tarinfo=thisone, fileobj=io)
else:
# try saving everything else here by jsoning
try:
io = _StringIO()
_json.dump(att, io)
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name=att_name)
thisone.size = length
tar.addfile(tarinfo=thisone, fileobj=io)
except TypeError:
# ignore non-jsonable things like functions
# include unicodes, strings, lists etc etc
#print('omitting {}'.format(att_name))
pass
def __init__(self, reads=False, genome=False, baga=False):
'''
Initialise with:
a baga.PrepareReads.Reads object and,
a baga.CollectData.Genome object.
OR
a path to baga.AlignReads.SAMs (like this one) object that
was previously saved.
'''
if (reads and genome) and not baga:
try:
self.read_files = reads.trimmed_read_files
except AttributeError:
text = 'WARNING: baga was not used to quality-score trim these reads. Read trimming is recommended for most types of analysis. This can be achieved with the "trim()" method of the Reads class in the PrepareReads module.'
print(text)
try:
self.read_files = reads.adaptorcut_read_files
except AttributeError:
text = 'WARNING: baga was not used to remove library preparation adaptor sequences from these reads. Adaptor removal is highly recommended so hopefully you already removed adaptor sequences! This can be achieved with the "cutAdaptors()" method of the Reads class in the PrepareReads module.'
self.read_files = reads.read_files
print(text)
print('continuing with these reads . . .')
# currently baga CollectData includes path to reads in pairname keys to read file pair values
# check and remove here
for pairname, files in self.read_files.items():
if _os.path.sep in pairname:
self.read_files[pairname.split(_os.path.sep)[-1]] = files
del self.read_files[pairname]
self.genome_sequence = genome.sequence
self.genome_id = genome.id
elif baga and not (reads and genome):
# for reloading a previous instantiation
with _tarfile.open(baga, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
#print('json failed: {}'.format(member.name))
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
else:
raise NameError(
'instantiate baga.AlignReads.SAMs with either loaded baga.PrepareReads.Reads-*.baga and baga.CollectData.Genome-*.baga objects or previous saved alignments (baga.AlignReads.SAMs-*.baga)'
)
def loadFrombaga(local_path):
with _tarfile.open(local_path, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
def loadFrombaga(local_path):
with _tarfile.open(local_path, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
def __init__(self, reads = False, genome = False, baga = False):
'''
Initialise with:
a baga.PrepareReads.Reads object and,
a baga.CollectData.Genome object.
OR
a path to baga.AlignReads.SAMs (like this one) object that
was previously saved.
'''
if (reads and genome) and not baga:
try:
self.read_files = reads.trimmed_read_files
except AttributeError:
text = 'WARNING: baga was not used to quality-score trim these reads. Read trimming is recommended for most types of analysis. This can be achieved with the "trim()" method of the Reads class in the PrepareReads module.'
print(text)
try:
self.read_files = reads.adaptorcut_read_files
except AttributeError:
text = 'WARNING: baga was not used to remove library preparation adaptor sequences from these reads. Adaptor removal is highly recommended so hopefully you already removed adaptor sequences! This can be achieved with the "cutAdaptors()" method of the Reads class in the PrepareReads module.'
self.read_files = reads.read_files
print(text)
print('continuing with these reads . . .')
# currently baga CollectData includes path to reads in pairname keys to read file pair values
# check and remove here
for pairname, files in self.read_files.items():
if _os.path.sep in pairname:
self.read_files[pairname.split(_os.path.sep)[-1]] = files
del self.read_files[pairname]
self.genome_sequence = genome.sequence
self.genome_id = genome.id
elif baga and not (reads and genome):
# for reloading a previous instantiation
with _tarfile.open(baga, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
#print('json failed: {}'.format(member.name))
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
else:
raise NameError('instantiate baga.AlignReads.SAMs with either loaded baga.PrepareReads.Reads-*.baga and baga.CollectData.Genome-*.baga objects or previous saved alignments (baga.AlignReads.SAMs-*.baga)')
def saveLocal(self, name = False):
'''
Save a reference genome to a local compressed baga file. This saves
Internet bandwidth if downloading from NCBI and time if loading a
genbank file.
'filename' can exclude extension: .baga will be added
A .baga file is mostly Python dictionaries in JSON strings and
array.array objects in a tar.gz format.
'''
if name:
fileout = 'baga.CollectData.Genome-{}.baga'.format(name)
else:
fileout = 'baga.CollectData.Genome-{}.baga'.format(self.id)
with _tarfile.open(fileout, "w:gz") as tar:
print('Writing to {} . . . '.format(fileout))
for att_name, att in self.__dict__.items():
if isinstance(att, _array):
io = _StringIO(att.tostring())
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name = att_name)
thisone.size = length
tar.addfile(tarinfo = thisone, fileobj = io)
elif isinstance(att, dict) or isinstance(att, str):
# ensure only dicts or strings for genome objects but shouldn't be anything else anyway
io = _StringIO()
_json.dump(att, io)
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name = att_name)
thisone.size = length
tar.addfile(tarinfo = thisone, fileobj = io)
def saveLocal(self, name=False):
'''
Save a reference genome to a local compressed baga file. This saves
Internet bandwidth if downloading from NCBI and time if loading a
genbank file.
'filename' can exclude extension: .baga will be added
A .baga file is mostly Python dictionaries in JSON strings and
array.array objects in a tar.gz format.
'''
if name:
fileout = 'baga.CollectData.Genome-{}.baga'.format(name)
else:
fileout = 'baga.CollectData.Genome-{}.baga'.format(self.id)
with _tarfile.open(fileout, "w:gz") as tar:
print('Writing to {} . . . '.format(fileout))
for att_name, att in self.__dict__.items():
if isinstance(att, _array):
io = _StringIO(att.tostring())
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name=att_name)
thisone.size = length
tar.addfile(tarinfo=thisone, fileobj=io)
elif isinstance(att, dict) or isinstance(att, str):
# ensure only dicts or strings for genome objects but shouldn't be anything else anyway
io = _StringIO()
_json.dump(att, io)
io.seek(0, _os.SEEK_END)
length = io.tell()
io.seek(0)
thisone = _tarfile.TarInfo(name=att_name)
thisone.size = length
tar.addfile(tarinfo=thisone, fileobj=io)
def __init__(self, genome = False,
baga = False,
num_individuals = 1,
large_deletions = {},
large_del_padding = 1000,
random_seed = False):
'''
Initialise with:
a baga.CollectData.Genome object.
OR
a path to baga.SimulateReads.Reads (like this one) object that
was previously saved.
Large deletions can be included to simulate e.g. missing genomic islands or prophage.
Currently, a set of genomes are generated with and a set without the large deletions if
specified.
large_deletions should be a dict with arbitrary names of deletions as keys and
tuples of (start,end) for python slices delineating each deletion. If supplied a
set of genomes with and without deletions are generated, each set consisting of
num_individuals members.
No variants will be generated within large_del_padding: a 'safe' distance around large
deletions outside of which variant calling should be reliable. Small deletions might run
over into this zone.
'''
if random_seed:
if not isinstance(random_seed, int):
raise ValueError('random_seed must be integer')
_random.seed(random_seed) # 684651
if genome and not baga:
self.genome = genome
elif baga and not genome:
# for reloading a previous instantiation
with _tarfile.open(baga, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
#print('json failed: {}'.format(member.name))
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
else:
raise NameError('instantiate baga.SimulateReads.Reads with a loaded baga.CollectData.Genome-*.baga object or previously saved baga.SimulateReads.Reads object')
omit = set()
for name,(start,end) in large_deletions.items():
omit.update(range(start-large_del_padding, end+large_del_padding))
samplefrom = sorted(set(range(len(self.genome.sequence))) - omit)
self.samplefrom = samplefrom
self.large_deletions = large_deletions
self.num_individuals = num_individuals
# to be optionally populated with methods
self.SNPs_per_genome = []
self.indel_dict_by_pos_pergenome = []
def __init__(self,
genome=False,
baga=False,
num_individuals=1,
large_deletions={},
large_del_padding=1000,
random_seed=False):
'''
Initialise with:
a baga.CollectData.Genome object.
OR
a path to baga.SimulateReads.Reads (like this one) object that
was previously saved.
Large deletions can be included to simulate e.g. missing genomic islands or prophage.
Currently, a set of genomes are generated with and a set without the large deletions if
specified.
large_deletions should be a dict with arbitrary names of deletions as keys and
tuples of (start,end) for python slices delineating each deletion. If supplied a
set of genomes with and without deletions are generated, each set consisting of
num_individuals members.
No variants will be generated within large_del_padding: a 'safe' distance around large
deletions outside of which variant calling should be reliable. Small deletions might run
over into this zone.
'''
if random_seed:
if not isinstance(random_seed, int):
raise ValueError('random_seed must be integer')
_random.seed(random_seed) # 684651
if genome and not baga:
self.genome = genome
elif baga and not genome:
# for reloading a previous instantiation
with _tarfile.open(baga, "r:gz") as tar:
for member in tar:
contents = _StringIO(tar.extractfile(member).read())
try:
# either json serialised conventional objects
contents = _json.loads(contents.getvalue())
except ValueError:
#print('json failed: {}'.format(member.name))
# or longer python array.array objects
contents = _array('c', contents.getvalue())
setattr(self, member.name, contents)
else:
raise NameError(
'instantiate baga.SimulateReads.Reads with a loaded baga.CollectData.Genome-*.baga object or previously saved baga.SimulateReads.Reads object'
)
omit = set()
for name, (start, end) in large_deletions.items():
omit.update(
range(start - large_del_padding, end + large_del_padding))
samplefrom = sorted(set(range(len(self.genome.sequence))) - omit)
self.samplefrom = samplefrom
self.large_deletions = large_deletions
self.num_individuals = num_individuals
# to be optionally populated with methods
self.SNPs_per_genome = []
self.indel_dict_by_pos_pergenome = []
