def setup_example(config):
root = os.path.join(config.destination, 'zonkey_pipeline')
with tarfile.TarFile(config.tablefile) as tar_handle:
example_files = []
existing_files = []
for member in tar_handle.getmembers():
if os.path.dirname(member.name) == 'examples' and member.isfile():
example_files.append(member)
destination = fileutils.reroot_path(root, member.name)
if os.path.exists(destination):
existing_files.append(destination)
if existing_files:
print_err("Output files already exist at destination:\n - %s"
% ("\n - ".join(map(repr, existing_files))))
return 1
elif not example_files:
print_err("Sample database %r does not contain example data; "
"cannot proceed." % (config.tablefile,))
return 1
if not os.path.exists(root):
fileutils.make_dirs(root)
for member in example_files:
destination = fileutils.reroot_path(root, member.name)
src_handle = tar_handle.extractfile(member)
with open(destination, 'w') as out_handle:
shutil.copyfileobj(src_handle, out_handle)
print_info("Sucessfully saved example data in %r" % (root,))
return 0
def _teardown(self, _config, temp):
move_file(reroot_path(temp, self._out_prefix + ".phy"),
self._out_prefix + ".phy")
move_file(
reroot_path(temp, self._out_prefix + ".partitions"),
self._out_prefix + ".partitions",
)
def _run(self, _config, temp):
table = {}
for filename in self.input_files:
coverage.read_table(table, filename)
coverage.write_table(table, reroot_path(temp, self._output_file))
move_file(reroot_path(temp, self._output_file), self._output_file)
def setup_example(config):
root = os.path.join(config.destination, 'zonkey_pipeline')
with tarfile.TarFile(config.tablefile) as tar_handle:
example_files = []
existing_files = []
for member in tar_handle.getmembers():
if os.path.dirname(member.name) == 'examples' and member.isfile():
example_files.append(member)
destination = fileutils.reroot_path(root, member.name)
if os.path.exists(destination):
existing_files.append(destination)
if existing_files:
print_err("Output files already exist at destination:\n - %s"
% ("\n - ".join(map(repr, existing_files))))
return 1
elif not example_files:
print_err("Sample database %r does not contain example data; "
"cannot proceed." % (config.tablefile,))
return 1
if not os.path.exists(root):
fileutils.make_dirs(root)
for member in example_files:
destination = fileutils.reroot_path(root, member.name)
src_handle = tar_handle.extractfile(member)
with open(destination, 'w') as out_handle:
shutil.copyfileobj(src_handle, out_handle)
print_info("Sucessfully saved example data in %r" % (root,))
return 0
def _teardown(self, config, temp):
with open(fileutils.reroot_path(temp, self._params_file), "w") as out:
out.write("k: %i\n" % (self._param_k, ))
out.write("m: %i\n" % (self._param_m, ))
out.write("outgroup: %r\n" % (list(self._param_outgroup), ))
open(fileutils.reroot_path(temp, self._parameters_hash), "w").close()
CommandNode._teardown(self, config, temp)
def _teardown(self, config, temp):
with open(fileutils.reroot_path(temp, self._params_file), "w") as out:
out.write("k: %i\n" % (self._param_k,))
out.write("m: %i\n" % (self._param_m,))
out.write("outgroup: %r\n" % (list(self._param_outgroup),))
open(fileutils.reroot_path(temp, self._parameters_hash), "w").close()
CommandNode._teardown(self, config, temp)
def _teardown(self, config, temp):
for postfix in ("ALIGNMENT", "PARTITION"):
filenames = [self._kwargs["TEMP_IN_" + postfix],
self._kwargs["TEMP_IN_" + postfix] + ".reduced",
self._kwargs["OUT_" + postfix]]
for (source, destination) in zip(filenames, filenames[1:]):
source = fileutils.reroot_path(temp, source)
destination = fileutils.reroot_path(temp, destination)
if not os.path.exists(destination):
fileutils.copy_file(source, destination)
os.remove(source)
CommandNode._teardown(self, config, temp)
def _check_output_files(cls, output_files):
"""Checks dict of output files to nodes for cases where
multiple nodes create the same output file.
The directory component of paths are realized in order to
detect cases where nodes create the same file, but via
different paths (e.g. due to relative/absolute paths, or
due to use of symbolic links). Since output files are
replaced, not modified in place, it is not nessesary to
compare files themselves."""
dirpath_cache, real_output_files = {}, {}
for (filename, nodes) in output_files.items():
dirpath = os.path.dirname(filename)
if dirpath not in dirpath_cache:
dirpath_cache[dirpath] = os.path.realpath(dirpath)
real_output_file = reroot_path(dirpath_cache[dirpath], filename)
real_output_files.setdefault(real_output_file, []).extend(nodes)
for (filename, nodes) in real_output_files.items():
if len(nodes) > 1:
nodes = _summarize_nodes(nodes)
yield (
"Multiple nodes create the same (clobber) output-file:\n"
"\tFilename: %s\n\tNodes: %s" %
(filename, "\n\t ".join(nodes)))
def _setup(self, config, temp):
CommandNode._setup(self, config, temp)
input_files = [
self._input_file,
fileutils.swap_ext(self._input_file, ".bim"),
fileutils.swap_ext(self._input_file, ".fam"),
]
for filename in input_files:
basename = os.path.basename(filename)
os.symlink(os.path.abspath(filename), os.path.join(temp, basename))
if self._supervised:
fam_filename = fileutils.swap_ext(self._input_file, ".fam")
pop_filename = fileutils.swap_ext(fam_filename, ".pop")
pop_filename = fileutils.reroot_path(temp, pop_filename)
key = "Group(%i)" % (self._k_groups,)
with open(fam_filename) as fam_handle:
with open(pop_filename, "w") as pop_handle:
for line in fam_handle:
sample, _ = line.split(None, 1)
group = self._samples.get(sample, {}).get(key, "-")
pop_handle.write("%s\n" % (group,))
def _setup(self, config, temp):
"""See CommandNode._setup."""
infile = os.path.abspath(self._infile)
outfile = reroot_path(temp, self._infile)
os.symlink(infile, outfile)
CommandNode._setup(self, config, temp)
def _setup(self, config, temp):
CommandNode._setup(self, config, temp)
input_files = [
self._input_file,
fileutils.swap_ext(self._input_file, ".bim"),
fileutils.swap_ext(self._input_file, ".fam"),
]
for filename in input_files:
basename = os.path.basename(filename)
os.symlink(os.path.abspath(filename), os.path.join(temp, basename))
if self._supervised:
fam_filename = fileutils.swap_ext(self._input_file, ".fam")
pop_filename = fileutils.swap_ext(fam_filename, ".pop")
pop_filename = fileutils.reroot_path(temp, pop_filename)
key = "Group(%i)" % (self._k_groups, )
with open(fam_filename) as fam_handle:
with open(pop_filename, "w") as pop_handle:
for line in fam_handle:
sample, _ = line.split(None, 1)
group = self._samples.get(sample, {}).get(key, "-")
pop_handle.write("%s\n" % (group, ))
def _setup(self, config, temp):
"""See CommandNode._setup."""
infile = os.path.abspath(self._infile)
outfile = reroot_path(temp, self._infile)
os.symlink(infile, outfile)
CommandNode._setup(self, config, temp)
def _teardown(self, config, temp):
# Validate output from MAFFT
output_file = reroot_path(temp, self._output_file)
try:
MSA.from_file(output_file)
except MSAError, error:
raise NodeError("Invalid MSA produced by MAFFT:\n%s" % (error,))
def _check_output_files(cls, output_files):
"""Checks dict of output files to nodes for cases where
multiple nodes create the same output file.
The directory component of paths are realized in order to
detect cases where nodes create the same file, but via
different paths (e.g. due to relative/absolute paths, or
due to use of symbolic links). Since output files are
replaced, not modified in place, it is not nessesary to
compare files themselves."""
dirpath_cache, real_output_files = {}, {}
for (filename, nodes) in output_files.iteritems():
dirpath = os.path.dirname(filename)
if dirpath not in dirpath_cache:
dirpath_cache[dirpath] = os.path.realpath(dirpath)
real_output_file = reroot_path(dirpath_cache[dirpath], filename)
real_output_files.setdefault(real_output_file, []).extend(nodes)
for (filename, nodes) in real_output_files.iteritems():
if (len(nodes) > 1):
nodes = _summarize_nodes(nodes)
yield "Multiple nodes create the same (clobber) output-file:" \
"\n\tFilename: %s\n\tNodes: %s" \
% (filename, "\n\t ".join(nodes))
def _teardown(self, config, temp):
# Validate output from MAFFT
output_file = reroot_path(temp, self._output_file)
try:
MSA.from_file(output_file)
except MSAError, error:
raise NodeError("Invalid MSA produced by MAFFT:\n%s" % (error,))
def _teardown(self, config, temp):
os.remove(os.path.join(temp, "RAxML_info.output"))
source = os.path.join(temp, "RAxML_parsimonyTree.output.0")
destination = fileutils.reroot_path(temp, self._output_tree)
fileutils.move_file(source, destination)
CommandNode._teardown(self, config, temp)
def _teardown(self, config, temp):
os.remove(os.path.join(temp, "RAxML_info.output"))
source = os.path.join(temp, "RAxML_parsimonyTree.output.0")
destination = fileutils.reroot_path(temp, self._output_tree)
fileutils.move_file(source, destination)
CommandNode._teardown(self, config, temp)
def _run(self, _config, temp):
alignment = MSA.from_file(self._input_file)
for (to_filter, groups) in self._filter_by.items():
alignment = alignment.filter_singletons(to_filter, groups)
temp_filename = fileutils.reroot_path(temp, self._output_file)
with open(temp_filename, "w") as handle:
alignment.to_file(handle)
fileutils.move_file(temp_filename, self._output_file)
def _run(self, _config, temp):
alignment = MSA.from_file(self._input_file)
for (to_filter, groups) in self._filter_by.iteritems():
alignment = alignment.filter_singletons(to_filter, groups)
temp_filename = fileutils.reroot_path(temp, self._output_file)
with open(temp_filename, "w") as handle:
alignment.to_file(handle)
fileutils.move_file(temp_filename, self._output_file)
def _teardown(self, config, temp):
template = self._output_template
bootstraps = self._bootstrap_num
start = self._bootstrap_start
for (src_file, dst_file) in self._bootstraps(template, bootstraps, start):
src_file = os.path.join(temp, src_file)
dst_file = fileutils.reroot_path(temp, dst_file)
fileutils.move_file(src_file, dst_file)
CommandNode._teardown(self, config, temp)
def _run(self, config, temp):
try:
CommandNode._run(self, config, temp)
except NodeError, error:
if self._command.join() == [1, None]:
with open(fileutils.reroot_path(temp, "template.stdout")) as handle:
lines = handle.readlines()
if lines and ("Giving up." in lines[-1]):
error = NodeError("%s\n\n%s" % (error, lines[-1]))
raise error
def _run(self, config, temp):
try:
CommandNode._run(self, config, temp)
except NodeError, error:
if self._command.join() == [1, None]:
with open(fileutils.reroot_path(temp,
"template.stdout")) as handle:
lines = handle.readlines()
if lines and ("Giving up." in lines[-1]):
error = NodeError("%s\n\n%s" % (error, lines[-1]))
raise error
def _run(self, _config, temp):
# Read and check that MSAs share groups
msas = [MSA.from_file(filename) for filename in sorted(self.input_files)]
MSA.validate(*msas)
blocks = []
for msa in msas:
blocks.append(sequential_phy(msa, add_flag = self._add_flag))
with open(reroot_path(temp, self._out_phy), "w") as output:
output.write("\n\n".join(blocks))
def setup_example(config):
root = os.path.join(config.destination, "zonkey_pipeline")
log = logging.getLogger(__name__)
log.info("Copying example project to %r", root)
with tarfile.TarFile(config.database.filename) as tar_handle:
example_files = []
existing_files = []
for member in tar_handle.getmembers():
if os.path.dirname(member.name) == "examples" and member.isfile():
example_files.append(member)
destination = fileutils.reroot_path(root, member.name)
if os.path.exists(destination):
existing_files.append(destination)
if existing_files:
log.error("Output files already exist at destination:")
for filename in sorted(existing_files):
log.error(" - %r", filename)
return 1
elif not example_files:
log.error(
"Sample database %r does not contain example data; cannot proceed.",
config.database.filename,
)
return 1
if not os.path.exists(root):
fileutils.make_dirs(root)
for member in example_files:
destination = fileutils.reroot_path(root, member.name)
src_handle = tar_handle.extractfile(member)
with open(destination, "wb") as out_handle:
shutil.copyfileobj(src_handle, out_handle)
log.info("Sucessfully saved example data in %r", root)
return 0
def build_msa_nodes(options, settings, regions, filtering, dependencies):
if settings["Program"].lower() != "mafft":
raise RuntimeError("Only MAFFT support has been implemented!")
sequencedir = os.path.join(options.destination, "alignments",
regions["Name"])
# Run on full set of sequences
sequences = regions["Sequences"][None]
node = CollectSequencesNode(
fasta_files=regions["Genotypes"],
destination=sequencedir,
sequences=sequences,
dependencies=dependencies,
)
if settings["Enabled"]:
fasta_files = {}
algorithm = settings["MAFFT"]["Algorithm"]
for sequence in sequences:
input_file = os.path.join(sequencedir, sequence + ".fasta")
output_file = os.path.join(sequencedir, sequence + ".afa")
fasta_files[output_file] = MAFFTNode(
input_file=input_file,
output_file=output_file,
algorithm=algorithm,
options=settings["MAFFT"],
dependencies=node,
)
else:
fasta_files = dict((filename, node) for filename in node.output_files)
if not any(filtering.values()):
return list(fasta_files.values())
destination = sequencedir + ".filtered"
filtering = dict(filtering)
filtered_nodes = []
for (filename, node) in fasta_files.items():
output_filename = fileutils.reroot_path(destination, filename)
filtered_node = FilterSingletonsNode(
input_file=filename,
output_file=output_filename,
filter_by=filtering,
dependencies=node,
)
filtered_nodes.append(filtered_node)
return filtered_nodes
def _run(self, _config, temp):
# Read and check that MSAs share groups
msas = [
MSA.from_file(filename) for filename in sorted(self.input_files)
]
MSA.validate(*msas)
blocks = []
for msa in msas:
blocks.append(sequential_phy(msa, add_flag=self._add_flag))
with open(reroot_path(temp, self._out_phy), "w") as output:
output.write("\n\n".join(blocks))
def _run(self, _config, temp):
merged_msas = []
for (name, files_dd) in sorted(self._infiles.iteritems()):
partitions = files_dd["partitions"]
msas = dict((key, []) for key in partitions)
for filename in files_dd["filenames"]:
msa = MSA.from_file(filename)
if self._excluded:
msa = msa.exclude(self._excluded)
for (key, msa_part) in msa.split(partitions).iteritems():
msas[key].append(msa_part)
msas.pop("X", None)
for (key, msa_parts) in sorted(msas.iteritems()):
merged_msa = MSA.join(*msa_parts)
if self._reduce:
merged_msa = merged_msa.reduce()
if merged_msa is not None:
merged_msas.append(("%s_%s" % (name, key),
merged_msa))
out_fname_phy = reroot_path(temp, self._out_prefix + ".phy")
with open(out_fname_phy, "w") as output_phy:
final_msa = MSA.join(*(msa for (_, msa) in merged_msas))
output_phy.write(interleaved_phy(final_msa))
partition_end = 0
out_fname_parts = reroot_path(temp, self._out_prefix + ".partitions")
with open(out_fname_parts, "w") as output_part:
for (name, msa) in merged_msas:
length = msa.seqlen()
output_part.write("DNA, %s = %i-%i\n"
% (name,
partition_end + 1,
partition_end + length))
partition_end += length
def _run(self, _config, temp):
merged_msas = []
for (name, files_dd) in sorted(self._infiles.items()):
partitions = files_dd["partitions"]
msas = dict((key, []) for key in partitions)
for filename in files_dd["filenames"]:
msa = MSA.from_file(filename)
if self._excluded:
msa = msa.exclude(self._excluded)
for (key, msa_part) in msa.split(partitions).items():
msas[key].append(msa_part)
msas.pop("X", None)
for (key, msa_parts) in sorted(msas.items()):
merged_msa = MSA.join(*msa_parts)
if self._reduce:
merged_msa = merged_msa.reduce()
if merged_msa is not None:
merged_msas.append(("%s_%s" % (name, key), merged_msa))
out_fname_phy = reroot_path(temp, self._out_prefix + ".phy")
with open(out_fname_phy, "w") as output_phy:
final_msa = MSA.join(*(msa for (_, msa) in merged_msas))
output_phy.write(interleaved_phy(final_msa))
partition_end = 0
out_fname_parts = reroot_path(temp, self._out_prefix + ".partitions")
with open(out_fname_parts, "w") as output_part:
for (name, msa) in merged_msas:
length = msa.seqlen()
output_part.write(
"DNA, %s = %i-%i\n" %
(name, partition_end + 1, partition_end + length))
partition_end += length
def __init__(self, description, destination, source_nodes):
source_nodes = safe_coerce_to_tuple(source_nodes)
input_files = []
for source_node in source_nodes:
input_files.extend(source_node.output_files)
output_files = [reroot_path(destination, fpath) for fpath in input_files]
self._files = zip(input_files, output_files)
Node.__init__(self,
description = "<Copy %s output to %r>" % (description, destination),
input_files = input_files,
output_files = output_files,
dependencies = source_nodes)
def _run(self, config, temp):
output_file, = self.output_files
samples = self._samples
groups = set(sample["Group(3)"] for sample in samples.itervalues())
colors = dict(zip(groups, _DEFAULT_COLORS))
with open(fileutils.reroot_path(temp, output_file), "w") as handle:
handle.write("Name\tGroup\tColor\n")
for name, sample in sorted(samples.iteritems()):
group = sample["Group(3)"]
color = colors[group]
handle.write("%s\t%s\t%s\n" % (name, group, color))
handle.write("Sample\t-\t#000000\n")
def _run(self, config, temp):
output_file, = self.output_files
samples = self._samples
groups = set(sample["Group(3)"] for sample in samples.itervalues())
colors = dict(zip(groups, _DEFAULT_COLORS))
with open(fileutils.reroot_path(temp, output_file), "w") as handle:
handle.write("Name\tGroup\tColor\n")
for name, sample in sorted(samples.iteritems()):
group = sample["Group(3)"]
color = colors[group]
handle.write("%s\t%s\t%s\n" % (name, group, color))
handle.write("Sample\t-\t#000000\n")
def build_msa_nodes(options, settings, regions, filtering, dependencies):
if settings["Program"].lower() != "mafft":
raise RuntimeError("Only MAFFT support has been implemented!")
sequencedir = os.path.join(options.destination, "alignments", regions["Name"])
# Run on full set of sequences
sequences = regions["Sequences"][None]
node = CollectSequencesNode(fasta_files=regions["Genotypes"],
destination=sequencedir,
sequences=sequences,
dependencies=dependencies)
if settings["Enabled"]:
fasta_files = {}
algorithm = settings["MAFFT"]["Algorithm"]
for sequence in sequences:
input_file = os.path.join(sequencedir, sequence + ".fasta")
output_file = os.path.join(sequencedir, sequence + ".afa")
mafft = MAFFTNode.customize(input_file=input_file,
output_file=output_file,
algorithm=algorithm,
dependencies=node)
apply_options(mafft.command, settings["MAFFT"])
fasta_files[output_file] = mafft.build_node()
else:
fasta_files = dict((filename, node) for filename in node.output_files)
if not any(filtering.itervalues()):
return fasta_files.values()
destination = sequencedir + ".filtered"
filtering = dict(filtering)
filtered_nodes = []
for (filename, node) in fasta_files.iteritems():
output_filename = fileutils.reroot_path(destination, filename)
filtered_node = FilterSingletonsNode(input_file=filename,
output_file=output_filename,
filter_by=filtering,
dependencies=node)
filtered_nodes.append(filtered_node)
return filtered_nodes
def _run(self, config, temp):
likelihoods = []
for fileset in self._files:
for filename in fileset:
if filename.endswith(".log"):
likelihoods.append(
(self._read_admixture_log(filename), fileset))
break
else:
raise NodeError("No log-file found in list of admixture "
"output-files: %r" % (fileset, ))
_, fileset = max(likelihoods)
for src_filename in fileset:
dst_filename = fileutils.reroot_path(self._output_root,
src_filename)
fileutils.copy_file(src_filename, dst_filename)
def _run(self, config, temp):
likelihoods = []
for fileset in self._files:
for filename in fileset:
if filename.endswith(".log"):
likelihoods.append((self._read_admixture_log(filename),
fileset))
break
else:
raise NodeError("No log-file found in list of admixture "
"output-files: %r" % (fileset,))
_, fileset = max(likelihoods)
for src_filename in fileset:
dst_filename = fileutils.reroot_path(self._output_root,
src_filename)
fileutils.copy_file(src_filename, dst_filename)
def _run(self, config, temp):
rois = self._stat_areas_of_interest(self._prefixes)
genomes = self._stat_prefixes(self._prefixes)
with open(reroot_path(temp, self._output_file), "w") as table:
table.write("# Command:\n")
table.write("# %s\n" % (" ".join(sys.argv)))
table.write("#\n")
table.write("# Directory:\n")
table.write("# %s\n" % (os.getcwd()))
table.write("#\n")
self._write_genomes(table, genomes)
table.write("#\n")
self._write_areas_of_interest(table, rois)
table.write("#\n#\n")
for roi in rois.values():
genomes[roi["Label"]] = {"Size": roi["Size"]}
self._write_tables(table, genomes)
def __init__(self, description, destination, source_nodes):
source_nodes = safe_coerce_to_tuple(source_nodes)
input_files = []
for source_node in source_nodes:
input_files.extend(source_node.output_files)
output_files = [
reroot_path(destination, fpath) for fpath in input_files
]
self._files = zip(input_files, output_files)
Node.__init__(self,
description="<Copy %s output to %r>" %
(description, destination),
input_files=input_files,
output_files=output_files,
dependencies=source_nodes)
def _run(self, config, temp):
(output_file, ) = self.output_files
samples = self._samples
group = self._groups[max(self._groups)]
group_colors = dict(zip(sorted(set(group.values())), _DEFAULT_COLORS))
with open(fileutils.reroot_path(temp, output_file), "w") as handle:
handle.write("Name\tGroup\tColor\n")
for sample_name in sorted(samples):
group_name = group[sample_name]
group_color = group_colors[group_name]
handle.write("%s\t%s\t%s\n" %
(sample_name, group_name, group_color))
handle.write("Sample\t-\t#000000\n")
def _run(self, _config, temp):
end = 0
partitions = collections.defaultdict(list)
for (filename, msa) in _read_sequences(self._infiles):
length = msa.seqlen()
start, end = end + 1, end + length
for (group, offsets) in self._get_partition_by(filename):
if len(offsets) != 3:
parts = [("%i-%i\\3" % (start + offset, end)) for offset in offsets]
else:
parts = ["%i-%i" % (start, end)]
name = "%s_%s" % (self._infiles[filename]["name"], group)
partitions[name].extend(parts)
with open(reroot_path(temp, self._out_part), "w") as part_file:
for (name, parts) in sorted(partitions.items()):
part_file.writelines("DNA, %s = %s\n" % (name, ", ".join(parts)))
def _run(self, _config, temp):
if self._seed is not None:
rng = random.Random(self._seed)
partitions = _read_partitions(self._input_part)
header, names, sequences = _read_sequences(self._input_phy)
bootstraps = self._bootstrap_sequences(sequences, partitions, rng)
temp_fpath = reroot_path(temp, self._output_phy)
with open(temp_fpath, "w") as output_phy:
output_phy.write(header)
for (name, fragments) in zip(names, bootstraps):
output_phy.write(name)
output_phy.write(" ")
for sequence in fragments:
output_phy.write(sequence)
output_phy.write("\n")
move_file(temp_fpath, self._output_phy)
def _run(self, config, temp):
contigs = {}
with open(self._fai_file) as handle:
for line in handle:
name, length, _ = line.split('\t', 2)
if name in contigs:
raise NodeError('Reference genome contains multiple '
'identically named contigs (%r)!' %
(name, ))
contigs[name] = int(length)
with open(reroot_path(temp, self._outfile), 'w') as handle:
for record in read_bed_file(self._infile, contigs=contigs):
max_length = contigs[record.contig]
record.start = max(0, record.start - self._amount)
record.end = min(record.end + self._amount, max_length)
handle.write('%s\n' % (record, ))
def _run(self, _config, temp):
if self._seed is not None:
rng = random.Random(self._seed)
partitions = _read_partitions(self._input_part)
header, names, sequences = _read_sequences(self._input_phy)
bootstraps = self._bootstrap_sequences(sequences, partitions, rng)
temp_fpath = reroot_path(temp, self._output_phy)
with open(temp_fpath, "w") as output_phy:
output_phy.write(header)
for (name, fragments) in zip(names, bootstraps):
output_phy.write(name)
output_phy.write(" ")
for sequence in fragments:
output_phy.write(sequence)
output_phy.write("\n")
move_file(temp_fpath, self._output_phy)
def _run(self, config, temp):
contigs = {}
with open(self._fai_file) as handle:
for line in handle:
name, length, _ = line.split('\t', 2)
if name in contigs:
raise NodeError('Reference genome contains multiple '
'identically named contigs (%r)!'
% (name,))
contigs[name] = int(length)
with open(reroot_path(temp, self._outfile), 'w') as handle:
for record in read_bed_file(self._infile, contigs=contigs):
max_length = contigs[record.contig]
record.start = max(0, record.start - self._amount)
record.end = min(record.end + self._amount, max_length)
handle.write('%s\n' % (record,))
def _run(self, config, temp):
contigs = {}
with open(self._fai_file) as handle:
for line in handle:
name, length, _ = line.split("\t", 2)
if name in contigs:
raise NodeError("Reference genome contains multiple "
"identically named contigs (%r)!" %
(name, ))
contigs[name] = int(length)
with open(reroot_path(temp, self._outfile), "w") as handle:
records = list(read_bed_file(self._infile, contigs=contigs))
pad_bed_records(records=records,
padding=self._amount,
max_sizes=contigs)
for record in merge_bed_records(records):
handle.write("%s\n" % (record, ))
def _run(self, _config, temp):
end = 0
partitions = collections.defaultdict(list)
for (filename, msa) in _read_sequences(self._infiles):
length = msa.seqlen()
start, end = end + 1, end + length
for (group, offsets) in self._get_partition_by(filename):
if len(offsets) != 3:
parts = [("%i-%i\\3" % (start + offset, end))
for offset in offsets]
else:
parts = ["%i-%i" % (start, end)]
name = "%s_%s" % (self._infiles[filename]["name"], group)
partitions[name].extend(parts)
with open(reroot_path(temp, self._out_part), "w") as part_file:
for (name, parts) in sorted(partitions.items()):
part_file.writelines("DNA, %s = %s\n" %
(name, ", ".join(parts)))
def _run(self, config, temp):
rois = self._stat_areas_of_interest(self._prefixes)
genomes = self._stat_prefixes(self._prefixes)
with open(reroot_path(temp, self._output_file), "w") as table:
table.write("# Command:\n")
table.write("# %s\n" % (" ".join(sys.argv)),)
table.write("#\n")
table.write("# Directory:\n")
table.write("# %s\n" % (os.getcwd()),)
table.write("#\n")
table.write("# Makefile:\n")
table.write("# Filename: %s\n" % (self._makefile["Filename"],))
table.write("# SHA1Sum: %s\n" % (self._makefile["Hash"],))
table.write("# MTime: %s\n" % (self._makefile["MTime"],))
table.write("#\n")
self._write_genomes(table, genomes)
table.write("#\n")
self._write_areas_of_interest(table, rois)
table.write("#\n#\n")
for roi in rois.itervalues():
genomes[roi["Label"]] = {"Size" : roi["Size"]}
self._write_tables(table, genomes)
def _setup(self, _config, temp):
os.symlink(self._in_reference, reroot_path(temp, self._in_reference))
def test_reroot_path__rel_rel__w_final_dash():
assert_equal(reroot_path("etc/apt/", "tmp/sources.list"),
"etc/apt/sources.list")
def main(argv):
config, args = parse_options(argv)
if config is None:
return 1
# Get default options for bam_pipeline
bam_config, _ = bam_cfg.parse_config(args, "bam")
makefiles = bam_pipeline.read_makefiles(bam_config, args)
# Build .fai files for reference .fasta files
bam_pipeline.index_references(bam_config, makefiles)
for makefile in makefiles:
mkfile_fname = makefile["Statistics"]["Filename"]
bam_config.destination = os.path.dirname(mkfile_fname)
tasks = bam_pipeline.build_pipeline_full(bam_config, makefile,
return_nodes=False)
make_dirs(config.destination)
makefile_name = add_postfix(makefile["Statistics"]["Filename"],
config.postfix)
makefile_path = reroot_path(config.destination, makefile_name)
if samefile(makefile["Statistics"]["Filename"], makefile_path):
sys.stderr.write("ERROR: Would overwrite source makefile at %r\n" % (makefile_path,))
sys.stderr.write(" Please set --destination and/or --output-name-postfix\n")
sys.stderr.write(" before continuing.\n")
return 1
print("Writing makefile", makefile_path)
found_prefix = False
for prefix in makefile["Prefixes"]:
if prefix != config.prefix:
print("%sSkipping %s" % (_INDENTATION, prefix))
else:
found_prefix = True
if not found_prefix:
sys.stderr.write("\nERROR:\n")
sys.stderr.write("Could not find prefix %r in %r! Aborting ...\n"
% (config.prefix, mkfile_fname))
return 1
with open(makefile_path, "w") as makefile_handle:
template = bam_mkfile.build_makefile(add_sample_tmpl=False)
makefile_handle.write(template)
makefile_handle.write("\n" * 3)
for target in tasks:
target_name = add_postfix(target.name, config.postfix)
print("%sTarget: %s -> %s" % (_INDENTATION,
target.name,
target_name))
makefile_handle.write('%s"%s":\n' % (_INDENTATION * 0,
target_name))
for prefix in target.prefixes:
if prefix.name != config.prefix:
continue
for sample in prefix.samples:
print("%sSample: %s" % (_INDENTATION * 2, sample.name))
makefile_handle.write('%s"%s":\n' % (_INDENTATION * 1,
sample.name))
for library in sample.libraries:
print("%sLibrary: %s" % (_INDENTATION * 3,
library.name))
makefile_handle.write('%s"%s":\n'
% (_INDENTATION * 2,
library.name))
sink_cache = {}
destination = os.path.join(target_name,
"reads",
sample.name,
library.name)
for lane in library.lanes:
convert_reads(config, destination, lane, sink_cache)
ReadSink.close_all_sinks()
for lane_name in sorted(sink_cache):
makefile_handle.write('%s"%s":\n' % (_INDENTATION * 3, lane_name))
for (reads_type, sink) in sorted(sink_cache[lane_name].items()):
makefile_handle.write('%s%s "%s"\n'
% (_INDENTATION * 4,
("%s:" % (reads_type,)).ljust(20),
sink.filename))
makefile_handle.write("\n")
print("\tDone ...")
print()
return 0
def _setup(self, _config, temp):
# Ensure that Picard CreateSequenceDict cannot reuse any existing
# sequence dictionaries, if the underlying files have changed.
os.symlink(self._in_reference, reroot_path(temp, self._in_reference))
def _teardown(self, _config, temp):
move_file(reroot_path(temp, self._output_file), self._output_file)
def _teardown(self, config, temp):
"""See CommandNode._teardown."""
os.remove(reroot_path(temp, self._infile))
CommandNode._teardown(self, config, temp)
def _teardown(self, config, temp):
"""See CommandNode._teardown."""
os.remove(reroot_path(temp, self._infile))
CommandNode._teardown(self, config, temp)
def _teardown(self, _config, temp):
move_file(reroot_path(temp, self._out_phy), self._out_phy)
def _setup(self, _config, temp):
# Ensure that Picard CreateSequenceDict cannot reuse any existing
# sequence dictionaries, if the underlying files have changed.
os.symlink(self._in_reference, reroot_path(temp, self._in_reference))
def _run(self, _config, temp):
msa = MSA.join(*(MSA.from_file(filename) for filename in sorted(self.input_files)))
with open(reroot_path(temp, self._out_phy), "w") as output:
output.write(interleaved_phy(msa, add_flag = self._add_flag))
def _teardown(self, _config, temp):
for destination in sorted(self._outfiles):
source = fileutils.reroot_path(temp, destination)
fileutils.move_file(source, destination)
def _teardown(self, config, temp):
(destination, ) = self.output_files
source = fileutils.reroot_path(temp, destination)
fileutils.move_file(source, destination)
def _teardown(self, _config, temp):
move_file(reroot_path(temp, self._out_prefix + ".phy"),
self._out_prefix + ".phy")
move_file(reroot_path(temp, self._out_prefix + ".partitions"),
self._out_prefix + ".partitions")
