def __call__(self, identifier, partner_identifier, row_target, col_target,
scale, reduction_axis):
scale_dir = ensure_dir(os.path.join(self.work_dir, str(scale)))
next_scale_dir = ensure_dir(os.path.join(self.work_dir,
str(scale + 1)))
temp_dir = ensure_dir(os.path.join(scale_dir, "work"))
input_img = [identifier, partner_identifier]
axis_name = ['rows', 'cols']
print("Partners over {}: {} and {}".format(
axis_name[reduction_axis], os.path.basename(identifier),
os.path.basename(partner_identifier)))
target_name = "{}_{}".format(row_target, col_target)
target_file = "{}.{}".format(target_name, img_data_fmt)
dir_stitch = setup_directories(os.path.join(temp_dir, target_name))
stitched_img = os.path.join(next_scale_dir, target_file)
if reduction_axis == 0:
success = assemble_row(dir_stitch, input_img, stitched_img)
else:
success = assemble_column(dir_stitch, input_img, stitched_img)
return success, stitched_img
def mux_batch_blast(dataset, bin_type):
"""Send batch jobs to Blast. Muxes to multiple reference DBs."""
# Identify the genome
nickname = dataset['nickname']
# Determine the input file root
root_dir = dirs['parcels_dir']+nickname+"/"
file_root = root_dir+nickname+bin_type
# Identify the references to blast against
ref_nicks = dataset['ref_nicks']
for ref_nick in ref_nicks:
# Identify Blast DB
db_path = dirs['blast_db_dir']+ref_nick
# Prep output directory
out_dir = dirs['blast_out_dir']+nickname+"/"+ref_nick+"/"
ensure_dir(out_dir)
# Signal process start
print "--- Blasting", nickname+bin_type, "against", ref_nick, "---"
print datetime.now()
index = 1
while os.path.isfile(file_root+"_"+str(index)+".fas"):
query_file = file_root+"_"+str(index)+".fas"
outfile = out_dir+nickname+bin_type+"_"+str(index)+"_blast.out"
print "\tblasting", query_file
local_blastn_2file(query_file, db_path, outfile, blast_prefs)
index +=1
print "--- Finished BLAST run ---"
print datetime.now()
print index, "parcel files blasted"
return "OK"
def build_a_bear():
"""
the output of bear differs between versions, so we build the
latest bear rather than trying to support multiple versions.
FIXME: might be better to handle multiple versions instead.
"""
if os.path.isdir(c.BEAR_PREFIX):
logging.debug("skipping Bear installation")
return
# download
if not os.path.isfile(c.BEAR_ARCHIVE):
curl = get_cmd_or_die("curl")
curl['-s', c.BEAR_URL, '-o', c.BEAR_ARCHIVE] & pb.TEE
# remove any existing build dir since we don't know if
# bear was built for the current host environment.
if os.path.isdir(c.BEAR_SRC):
shutil.rmtree(c.BEAR_SRC, ignore_errors=True)
# unpack
tar = get_cmd_or_die("tar")
with pb.local.cwd(c.DEPS_DIR):
tar['xf', c.BEAR_ARCHIVE] & pb.TEE
# cmake
bear_build_dir = os.path.join(c.BEAR_SRC, "build")
bear_install_prefix = "-DCMAKE_INSTALL_PREFIX=" + c.BEAR_PREFIX
ensure_dir(bear_build_dir)
with pb.local.cwd(bear_build_dir):
cmake = get_cmd_or_die("cmake")
cmake["..", bear_install_prefix] & pb.TEE
make = get_cmd_or_die("make")
make["install"] & pb.TEE
def apply_blankfield(files, blank_field_file, dest_dir, percentage=100,
blur_alpha=0.7, threads=1, is_dark=False):
files = sorted(files)
ensure_dir(dest_dir)
blank_field = image_load_resize(blank_field_file, percentage)
if blur_alpha > 0:
blank_field = omomorphic_shading_extraction(blank_field, blur_alpha)
blankfield_wgt = get_blankfield_weights(blank_field)
ims = (image_load_resize(i, percentage) for i in files)
job_args = [ (blankfield_wgt, im_i, f, dest_dir) for f, im_i
in itertools.izip(files, ims) ]
if threads == 1:
results = [ apply_blankfield_weights(*args) for args in job_args ]
else:
pool = mp.Pool(processes=threads)
jobs = [ pool.apply_async(apply_blankfield_weights, args) for args
in job_args ]
pool.close()
pool.join()
results = [ job.get() for job in jobs ]
success = all(results)
return success
def generate_write_blankfield(files_in, output_dir, percentage=100, threads=1):
bfield = generate_blankfield(files_in, int(percentage), threads)
ensure_dir(output_dir)
success = cv2.imwrite(os.path.join(output_dir, "blankfield-stat.png"),
bfield)
return success
def simple_q2a(dataset, trim_file):
"""."""
# Identify the genome
nickname = dataset['nickname']
# Identify the trim file type
ttype = trim_file['type']
# Identify the source file
source_file = trim_file['name']
# Prep output files
dir_root = dirs['mft_dir']+nickname+"/"
ensure_dir(dir_root)
out_file = dir_root+nickname+ttype+'.fas'
track_file = dir_root+nickname+ttype+'_track.txt'
# Save filenames for later reference
dataset['mft_files'].append({'type': ttype,
'name': out_file,
'track': track_file})
# Signal the process start
print "-- Converting ", nickname+ttype, "to multifasta --"
print datetime.now()
# Set up iterator
multifasta = open(out_file, 'w')
tracker = open(track_file, 'w')
read_count = 0
for title, seq, qual in FastqGeneralIterator(open(source_file)) :
read_count +=1
id_string = nickname+"_"+str(read_count)
multifasta.write(">"+id_string+"\n"+seq+"\n")
tracker.write(title+"\t"+id_string+"\n")
if read_count%100000==0:
print "\t"+str(read_count), "reads processed"
multifasta.close()
tracker.close()
return read_count
def chop_multifasta(dataset, mft_file):
"""Split a master file into smaller multifasta files.
This is useful for making reasonably-sized batch BLAST jobs.
Iterator function adapted from http://biopython.org/wiki/Split_large_file
"""
# Identify the genome
nickname = dataset['nickname']
# Identify the trim file type
ttype = mft_file['type']
# Prep output file
dir_root = dirs['parcels_dir']+nickname+"/"
out_file_root = dir_root+nickname+ttype
ensure_dir(dir_root)
# Save filenames for later reference
dataset['parcel_files'] = {'root': out_file_root, 'suffix': '.fas'}
# Unpack chopping parameters
parceln = chop_param['parceln']
# Signal the process start
print "-- Splitting ", nickname+ttype, "into batches of", parceln, "reads --"
print datetime.now()
# Set up iterator function
record_iter = SeqIO.parse(open(mft_file['name']),"fasta")
parcel_files = []
for i, batch in enumerate(batch_iterator(record_iter, parceln)) :
filename = out_file_root+"_%i.fas" % (i+1)
handle = open(filename, "w")
count = SeqIO.write(batch, handle, "fasta")
handle.close()
parcel_files.append(filename)
print "\twrote %i records to %s" % (count, filename)
print "-- Finished --"
print datetime.now()
return len(parcel_files)
def _main():
if on_mac():
die("Cross-checking is only supported on Linux hosts.")
setup_logging()
logging.debug("args: %s", " ".join(sys.argv))
# earlier plumbum versions are missing features such as TEE
if pb.__version__ < c.MIN_PLUMBUM_VERSION:
err = "locally installed version {} of plumbum is too old.\n" \
.format(pb.__version__)
err += "please upgrade plumbum to version {} or later." \
.format(c.MIN_PLUMBUM_VERSION)
die(err)
args = _parse_args()
if args.clean_all:
logging.info("cleaning all dependencies and previous built files")
shutil.rmtree(c.CLANG_XCHECK_PLUGIN_BLD, ignore_errors=True)
make = get_cmd_or_die('make')
with pb.local.cwd(c.LIBFAKECHECKS_DIR):
make('clean')
# clang 3.6.0 is known to work; 3.4.0 known to not work.
ensure_clang_version([3, 6, 0])
# NOTE: it seems safe to disable this check since we now
# that we use a rust-toolchain file for rustc versioning.
# ensure_rustc_version(c.CUSTOM_RUST_RUSTC_VERSION)
ensure_dir(c.CLANG_XCHECK_PLUGIN_BLD)
ensure_dir(c.BUILD_DIR)
git_ignore_dir(c.BUILD_DIR)
build_clang_plugin(args)
def test_js_engine_path(self):
# Test that running JS commands works for node, d8, and jsc and is not path dependent
restore_and_set_up()
sample_script = test_file('print_args.js')
# Fake some JS engines
# Note that the path contains 'd8'.
test_path = self.in_dir('fake', 'abcd8765')
ensure_dir(test_path)
jsengines = [('d8', config.V8_ENGINE),
('d8_g', config.V8_ENGINE),
('js', config.SPIDERMONKEY_ENGINE),
('node', config.NODE_JS),
('nodejs', config.NODE_JS)]
for filename, engine in jsengines:
try_delete(SANITY_FILE)
if type(engine) is list:
engine = engine[0]
if not engine:
print('WARNING: Not testing engine %s, not configured.' % (filename))
continue
print(filename, engine)
test_engine_path = os.path.join(test_path, filename)
with open(test_engine_path, 'w') as f:
f.write('#!/bin/sh\n')
f.write('exec %s $@\n' % (engine))
make_executable(test_engine_path)
out = self.run_js(sample_script, engine=test_engine_path, args=['--foo'])
self.assertEqual('0: --foo', out.strip())
def __init__(self,
conf,
uuid,
namespace=None,
service=None,
pids_path=None,
default_cmd_callback=None,
cmd_addl_env=None,
pid_file=None,
run_as_root=False):
self.conf = conf
self.uuid = uuid
self.namespace = namespace
self.default_cmd_callback = default_cmd_callback
self.cmd_addl_env = cmd_addl_env
self.pids_path = pids_path or self.conf.external_pids
self.pid_file = pid_file
self.run_as_root = run_as_root
if service:
self.service_pid_fname = 'pid.' + service
self.service = service
else:
self.service_pid_fname = 'pid'
self.service = 'default-service'
common_utils.ensure_dir(os.path.dirname(self.get_pid_file_name()))
def _main():
setup_logging()
logging.debug("args: %s", " ".join(sys.argv))
# FIXME: allow env/cli override of LLVM_SRC and LLVM_BLD
# FIXME: check that cmake and ninja are installed
# FIXME: option to build LLVM/Clang from master?
args = _parse_args()
if args.clean_all:
logging.info("cleaning all dependencies and previous built files")
shutil.rmtree(c.LLVM_SRC, ignore_errors=True)
shutil.rmtree(c.LLVM_BLD, ignore_errors=True)
shutil.rmtree(c.BUILD_DIR, ignore_errors=True)
shutil.rmtree(c.AST_EXPO_PRJ_DIR, ignore_errors=True)
cargo = get_cmd_or_die("cargo")
with pb.local.cwd(c.ROOT_DIR):
invoke(cargo, "clean")
ensure_dir(c.LLVM_BLD)
ensure_dir(c.BUILD_DIR)
git_ignore_dir(c.BUILD_DIR)
download_llvm_sources()
configure_and_build_llvm(args)
build_transpiler(args)
print_success_msg(args)
def _main():
setup_logging()
logging.debug("args: %s", " ".join(sys.argv))
# earlier plumbum versions are missing features such as TEE
if pb.__version__ < c.MIN_PLUMBUM_VERSION:
err = "locally installed version {} of plumbum is too old.\n" \
.format(pb.__version__)
err += "please upgrade plumbum to version {} or later." \
.format(c.MIN_PLUMBUM_VERSION)
die(err)
args = _parse_args()
if args.clean_all:
logging.info("cleaning all dependencies and previous built files")
shutil.rmtree(c.CLANG_XCHECK_PLUGIN_BLD, ignore_errors=True)
# prerequisites
if not have_rust_toolchain(c.CUSTOM_RUST_NAME):
die("missing rust toolchain: " + c.CUSTOM_RUST_NAME, errno.ENOENT)
# clang 3.6.0 is known to work; 3.4.0 known to not work.
ensure_clang_version([3, 6, 0])
ensure_rustc_version(c.CUSTOM_RUST_RUSTC_VERSION)
ensure_dir(c.CLANG_XCHECK_PLUGIN_BLD)
ensure_dir(c.DEPS_DIR)
git_ignore_dir(c.DEPS_DIR)
build_clang_plugin(args)
def importNavteq(options):
ensure_dir(options.output_dir)
netconvert = sumolib.checkBinary('netconvert')
polyconvert = sumolib.checkBinary('polyconvert')
for idx, config in enumerate(options.config.split(",")):
netconvert_call = [netconvert, '--output-file', options.netfile, '-c', config]
if idx > 0:
tmp_net = os.path.join(
options.output_dir, options.net_prefix + "_tmp.net.xml")
os.rename(options.netfile, tmp_net)
netconvert_call += ['--sumo-net-file', tmp_net]
else:
netconvert_call += ['--dlr-navteq', options.prefix]
if options.verbose:
print(' '.join(netconvert_call))
sys.stdout.flush()
subprocess.call(netconvert_call)
polyconvertCmd = [
polyconvert,
'--verbose',
'--dlr-navteq-poly-files', options.prefix + '_polygons.txt',
#'--dlr-navteq-poi-files', options.prefix + '_points_of_interest.txt',
'--output', os.path.join(options.output_dir, "shapes.xml"),
'-n', options.netfile
]
if options.verbose:
print(polyconvertCmd)
sys.stdout.flush()
subprocess.call(polyconvertCmd)
def annot_ref(ref_name, ctg_fas, prot_db_name, fixed_dirs, project_id,
blast_prefs):
"""Annotate reference contig (predict ORFs and assign function)."""
# locate the COG database
prot_db = fixed_dirs['ref_dbs_dir']+prot_db_name
# set inputs and outputs
g_gbk_ctgs_root = fixed_dirs['gbk_contigs_dir']+ref_name+"/"
ctg_cds_root = fixed_dirs['ctg_cds_dir']+ref_name+"/"
ctg_prot_root = fixed_dirs['ctg_prot_dir']+ref_name+"/"
ctg_blast_root = fixed_dirs['ctg_blast_dir']+ref_name+"/"
annot_trn_root = fixed_dirs['annot_trn_dir']
ensure_dir([g_gbk_ctgs_root, ctg_cds_root, ctg_prot_root,
ctg_blast_root, annot_trn_root])
trn_file = annot_trn_root+ref_name+"_annot.trn"
g_ctg_gbk = g_gbk_ctgs_root+ref_name+"_1.gbk"
annot_gbk = ctg_cds_root+ref_name+"_1_cds.gbk"
annot_aa = ctg_prot_root+ref_name+"_1_aa.fas"
blast_out = ctg_blast_root+ref_name+"_1.xml"
if path.exists(blast_out) and os.stat(blast_out)[6]==0:
os.remove(blast_out)
if not path.exists(g_ctg_gbk):
l_tag_base = ref_name+"_1"
record = annot_ctg(ctg_fas, ctg_fas, annot_gbk,
annot_aa, trn_file, prot_db,
blast_out, l_tag_base, blast_prefs)
record.description = ref_name+"_re-annotated"
record.name = ref_name+"_1"
record.dbxrefs = ["Project: "+project_id+"/"+ref_name
+"-like backbones"]
record.seq.alphabet = generic_dna
write_genbank(g_ctg_gbk, record)
else:
record = load_genbank(g_ctg_gbk)
return record
def _get_conf_base(cfg_root, uuid, ensure_conf_dir):
#TODO(mangelajo): separate responsibilities here, ensure_conf_dir
# should be a separate function
conf_dir = os.path.abspath(os.path.normpath(cfg_root))
conf_base = os.path.join(conf_dir, uuid)
if ensure_conf_dir:
common.ensure_dir(conf_dir)
return conf_base
def _get_conf_base(cfg_root, uuid, ensure_conf_dir):
#TODO(mangelajo): separate responsibilities here, ensure_conf_dir
# should be a separate function
conf_dir = os.path.abspath(os.path.normpath(cfg_root))
conf_base = os.path.join(conf_dir, uuid)
if ensure_conf_dir:
common.ensure_dir(conf_dir)
return conf_base
def run_fires(site, region):
url = site + '/jobs/archive'
try:
p = getPage(url)
except Exception as e:
logging.error("Can't load {}".format(url))
logging.error(e)
return None
a = p.findAll('a')
zips = [x.get('href') for x in a if x.get('href').endswith('.zip')]
fires = sorted(set([x[x.rindex('/') + 1:x.index('_')] for x in zips]))
times = {}
recent = {}
simtimes = {}
dates = []
totaltime = 0
dir_download = common.ensure_dir(os.path.join(DIR, region))
dir_ext = common.ensure_dir(os.path.join(EXT_DIR, region))
logging.debug("Checking {} fires".format(len(fires)))
for f in fires:
times[f] = [
datetime.datetime.strptime(x[x.rindex('_') + 1:x.rindex('.')],
'%Y%m%d%H%M%S%f') for x in zips
if x[x.rindex('/') + 1:x.index('_')] == f
]
recent[f] = {
'time':
max(times[f]),
'url': [
x for x in zips
if x[x.rindex('/') +
1:x.index('_')] == f and datetime.datetime.strptime(
x[x.rindex('_') +
1:x.rindex('.')], '%Y%m%d%H%M%S%f') == max(times[f])
][0],
}
logging.debug('{}: {}'.format(f, recent[f]['time']))
z = common.save_http(dir_download,
site + recent[f]['url'],
ignore_existing=True)
cur_dir = os.path.join(dir_ext, os.path.basename(z)[:-4])
common.unzip(z, cur_dir)
fgmj = os.path.join(cur_dir, 'job.fgmj')
if os.path.exists(fgmj):
try:
t0 = timeit.default_timer()
log_name = firestarr.do_run(fgmj)
t1 = timeit.default_timer()
if log_name is not None:
simtimes[f] = t1 - t0
totaltime = totaltime + simtimes[f]
logging.info("Took {}s to run {}".format(simtimes[f], f))
d = os.path.basename(os.path.dirname(log_name))[:8]
if d not in dates:
dates.append(d)
except Exception as e:
logging.error(e)
return simtimes, totaltime, dates
def make_genome_DB(genome, fixed_dirs):
"""Make a Blast DB from a genome FastA file."""
# load inputs
fas_dir = fixed_dirs['mfas_contigs_dir']
db_dir = fixed_dirs['blast_db_dir']
ensure_dir([fas_dir, db_dir])
g_name = genome['name']
# make DB
make_blastDB(db_dir+g_name, fas_dir+g_name+'_contigs.fas', 'nucl')
def _main():
setup_logging()
logging.debug("args: %s", " ".join(sys.argv))
# FIXME: allow env/cli override of LLVM_SRC, LLVM_VER, and LLVM_BLD
# FIXME: check that cmake and ninja are installed
# FIXME: option to build LLVM/Clang from master?
# earlier plumbum versions are missing features such as TEE
if pb.__version__ < c.MIN_PLUMBUM_VERSION:
err = "locally installed version {} of plumbum is too old.\n" \
.format(pb.__version__)
err += "please upgrade plumbum to version {} or later." \
.format(c.MIN_PLUMBUM_VERSION)
die(err)
args = _parse_args()
if args.clean_all:
logging.info("cleaning all dependencies and previous built files")
shutil.rmtree(c.LLVM_SRC, ignore_errors=True)
shutil.rmtree(c.LLVM_BLD, ignore_errors=True)
shutil.rmtree(c.DEPS_DIR, ignore_errors=True)
# prerequisites
if not have_rust_toolchain(c.CUSTOM_RUST_NAME):
die("missing rust toolchain: " + c.CUSTOM_RUST_NAME, errno.ENOENT)
# clang 3.6.0 is known to work; 3.4.0 known to not work.
ensure_clang_version([3, 6, 0])
ensure_rustc_version(c.CUSTOM_RUST_RUSTC_VERSION)
ensure_dir(c.LLVM_BLD)
ensure_dir(c.DEPS_DIR)
git_ignore_dir(c.DEPS_DIR)
if on_linux():
build_a_bear()
if not os.path.isfile(c.BEAR_BIN):
die("bear not found", errno.ENOENT)
download_llvm_sources()
integrate_ast_exporter()
cc_db = install_tinycbor()
configure_and_build_llvm(args)
# NOTE: we're not doing this anymore since it is
# faster and takes less space to simply pull the
# prebuilt nightly binaries with rustup
# download_and_build_custom_rustc(args)
build_ast_importer(args.debug)
if not on_mac() and args.sanity_test:
test_ast_exporter(cc_db)
def make_ref_DB(reference, run_id, fixed_dirs, r_root_dir, run_dirs):
"""Make a Blast DB from a reference FastA file."""
# load inputs
fas_dir = r_root_dir+run_id+"/"+run_dirs['ref_fas_dir']
db_dir = fixed_dirs['blast_db_dir']
ensure_dir([fas_dir, db_dir])
g_name = reference['name']
# make DB
make_blastDB(db_dir+g_name, fas_dir+g_name+'.fas', 'nucl')
def _main():
setup_logging()
logging.debug("args: %s", " ".join(sys.argv))
# FIXME: allow env/cli override of LLVM_SRC, LLVM_VER, and LLVM_BLD
# FIXME: check that cmake and ninja are installed
# FIXME: option to build LLVM/Clang from master?
# earlier plumbum versions are missing features such as TEE
if pb.__version__ < c.MIN_PLUMBUM_VERSION:
err = "locally installed version {} of plumbum is too old.\n" \
.format(pb.__version__)
err += "please upgrade plumbum to version {} or later." \
.format(c.MIN_PLUMBUM_VERSION)
die(err)
args = _parse_args()
# prerequisites
if not have_rust_toolchain(c.CUSTOM_RUST_NAME):
die("missing rust toolchain: " + c.CUSTOM_RUST_NAME, errno.ENOENT)
# clang 3.6.0 is known to work; 3.4.0 known to not work.
ensure_clang_version([3, 6, 0])
if args.clean_all:
logging.info("cleaning all dependencies and previous built files")
shutil.rmtree(c.LLVM_SRC, ignore_errors=True)
shutil.rmtree(c.LLVM_BLD, ignore_errors=True)
shutil.rmtree(c.DEPS_DIR, ignore_errors=True)
shutil.rmtree(c.AST_EXPO_PRJ_DIR, ignore_errors=True)
cargo = get_cmd_or_die("cargo")
with pb.local.cwd(c.ROOT_DIR):
invoke(cargo, "clean")
ensure_dir(c.LLVM_BLD)
ensure_dir(c.DEPS_DIR)
git_ignore_dir(c.DEPS_DIR)
download_llvm_sources()
update_cmakelists()
configure_and_build_llvm(args)
build_transpiler(args)
# print a helpful message on how to run c2rust bin directly
c2rust_bin_path = 'target/debug/c2rust' if args.debug \
else 'target/release/c2rust'
c2rust_bin_path = os.path.join(c.ROOT_DIR, c2rust_bin_path)
# if os.path.curdir
abs_curdir = os.path.abspath(os.path.curdir)
common_path = os.path.commonpath([abs_curdir, c2rust_bin_path])
if common_path != "/":
c2rust_bin_path = "." + c2rust_bin_path[len(common_path):]
print("success! you may now run", c2rust_bin_path)
def make_fake_llc(filename, targets):
"""Create a fake llc that only handles --version and writes target
list to stdout.
"""
print('make_fake_llc: %s' % filename)
ensure_dir(os.path.dirname(filename))
with open(filename, 'w') as f:
f.write('#!/bin/sh\n')
f.write('echo "llc fake output\nRegistered Targets:\n%s"' % targets)
make_executable(filename)
def extract_natives(self):
if not self.metadata:
self.get_meta()
natives_tmpdir = os.path.join(self.version_directory,
'natives-' + str(int(time.time())))
ensure_dir(natives_tmpdir)
for lib in self.metadata['libraries']:
skiplib = False
# Check Rules
if 'rules' in lib:
for rule in lib['rules']:
if 'action' in rule:
if rule['action'] == 'allow' and 'os' in rule:
if not rule['os']['name'] == platform():
skiplib = True
if rule['action'] == 'noallow' and 'os' in rule:
if rule['os']['name'] == platform():
skiplib = True
if skiplib:
continue
# Skip non-download-included for now
if not 'downloads' in lib:
continue
dl = lib['downloads']
if 'natives' in lib and 'extract' in lib:
if platform() in lib['natives']:
platform_native = lib['natives'][platform()]
if platform_native in dl['classifiers']:
try:
zip_ref = zipfile.ZipFile(
os.path.join(
self.client_root, 'libraries',
dl['classifiers'][platform_native]
['path']), 'r')
zip_ref.extractall(natives_tmpdir)
zip_ref.close()
except Exception as e:
print(
'Failed to extract native library %s due to errors.'
% (lib['name']))
raise e
metainf = os.path.join(natives_tmpdir, 'META-INF')
if os.path.exists(metainf):
shutil.rmtree(metainf)
self.natives = natives_tmpdir
def start_new_game():
# ask for player names
try:
# ask for name of game
game_name = input("Enter name of game (names with the same name will be overwritten): ")
# ask for number of worlds
num_worlds = int(
input("Enter number of worlds (the less, the earlier the complete wavefunction collapse will happen): "))
#
i = 1
name = "temp"
print(
"Entering names of player. Press enter without a name if you're finished. If you need to make changes later,"
"you can opt to edit the configuration directly")
player_names = []
while name != "":
print()
name = input("Please enter name of player %02d: " % i)
if name != "":
player_names.append(name)
i += 1
# ask for number of villagers, seer, players
while True:
num_villagers = int(input("Enter number of villagers: "))
num_wolves = int(input("Enter number of wolves: "))
num_seers = int(input("Enter number of seers: "))
if (num_seers + num_wolves + num_villagers) != len(player_names):
print("Number of players / number of roles mismatch.")
else:
break
except ValueError:
print("Invalid input. Please repeat the setup process.")
start_new_game()
return
# save to configuration
game_config = {
"name": game_name,
"num_worlds": num_worlds,
"players": player_names,
"num_villagers": num_villagers,
"num_wolves": num_wolves,
"num_seers": num_seers
}
game_dir = os.path.join("games", game_name)
ensure_dir(game_dir)
game_config["game_dir"] = game_dir
with open(os.path.join(game_dir, "config"), "w+") as f:
json.dump(game_config, f, indent=2)
print("Setup completed successfully.")
# load game from config
game = Game(dict_to_game_config(game_config))
play_game(game, game_config, True, 0)
def make_fake_tool(filename, version, report_name=None):
if not report_name:
report_name = os.path.basename(filename)
print('make_fake_tool: %s' % filename)
ensure_dir(os.path.dirname(filename))
with open(filename, 'w') as f:
f.write('#!/bin/sh\n')
f.write('echo "%s version %s"\n' % (report_name, version))
f.write('echo "..."\n')
f.write('exit 0\n')
make_executable(filename)
def ensure_directory_exists_without_file(path):
dirname = os.path.dirname(path)
if os.path.isdir(dirname):
try:
os.unlink(path)
except OSError:
with excutils.save_and_reraise_exception() as ctxt:
if not os.path.exists(path):
ctxt.reraise = False
else:
common.ensure_dir(dirname)
def ensure_directory_exists_without_file(path):
dirname = os.path.dirname(path)
if os.path.isdir(dirname):
try:
os.unlink(path)
except OSError:
with excutils.save_and_reraise_exception() as ctxt:
if not os.path.exists(path):
ctxt.reraise = False
else:
common.ensure_dir(dirname)
def make_qiime_reports(dataset, run_id):
"""Generate HTML output for reporting with Qiime.
Makes an interactive heatmap, a Cytoscape network and a summary of
community composition.
"""
# identify inputs and outputs
set_id = dataset['set_id']
print " ", set_id
run_root = root_dir+set_id+"/"+run_id+"/"
otus_dir = run_root+dirs['otus']
table_file = otus_dir+run_id+"_otu_table.txt"
map_file = otus_dir+run_id+"_map.txt"
heatmap_dir = run_root+dirs['reports']+"otu_heatmap"
network_dir = run_root+dirs['reports']
wf_taxa_sum = run_root+dirs['reports']+"communities"
ensure_dir(heatmap_dir)
ensure_dir(network_dir)
# generate a dummy Qiime map file
dummy_comps = ["#SampleID Barcode Primer Treat DOB Descript",
"#Dummy map file to make Qiime happy"]
for sample_id in dataset['samples']:
dummy_comps.append("\t".join([sample_id,"NA","NA","NA","NA","NA"]))
open(map_file, 'w').write("\n".join(dummy_comps))
# make OTU heatmap
comps = ["macqiime", "make_otu_heatmap_html.py", "-i", table_file,
"-o", heatmap_dir]
cline = " ".join(comps)
try:
child = subprocess.Popen(str(cline), stdout=subprocess.PIPE,
shell=True)
output, error = child.communicate()
except: raise
else: print "\t", "OTU heatmap generated"
# make OTU network
comps = ["macqiime", "make_otu_network.py", "-i", table_file,
"-m", map_file, "-o", network_dir]
cline = " ".join(comps)
try:
child = subprocess.Popen(str(cline), stdout=subprocess.PIPE,
shell=True)
output, error = child.communicate()
except: raise
else: print "\t", "OTU network generated"
# summarize communities by taxonomic composition
comps = ["macqiime", "summarize_taxa_through_plots.py", "-i", table_file,
"-o", wf_taxa_sum, "-m", map_file]
cline = " ".join(comps)
try:
child = subprocess.Popen(str(cline), stdout=subprocess.PIPE,
shell=True)
output, error = child.communicate()
except: raise
else: print "\t", "Taxonomic composition of communities summarized"
def __init__(self, archive_path, base_mount_dir):
self.archive_path = archive_path
self.mount_dir = pth.join(base_mount_dir,
"mnt-" + pth.basename(archive_path))
self.mount_available = False
if pth.isdir(self.mount_dir):
raise OSError("Mount directory {} already exists".format(
self.mount_dir))
ensure_dir(self.mount_dir)
self.mount_available = True
def write_object(self, obj: gitobj.GitObject):
content = obj.bcontent()
sha = compute_sha1(content)
dirname, filename = parse_sha(sha)
dirpath = self.path_in_gitdir('objects', dirname)
ensure_dir(dirpath)
path = os.path.join(dirpath, filename)
with open(path, 'wb') as f:
f.write(zlib.compress(content))
return sha
def set_stack_structure(stack_name, work_dir, make_subdirs=True):
st_name = pth.abspath(stack_name)
source_dirs = {kind: pth.join(st_name, kind) for kind in kinds}
base_dir = ensure_dir(pth.join(work_dir, pth.basename(st_name)))
target_dirs = {kind: pth.join(base_dir, kind) for kind in kinds}
if make_subdirs and all(
pth.exists(src_dir) for src_dir in source_dirs.values()):
[ensure_dir(tdir) for tdir in target_dirs.values()]
return source_dirs, target_dirs, base_dir
def main():
opt = process_command_line()
print opt
ensure_dir(opt.work_dir)
data = gather_images_data(opt.files, opt.crop_size, opt.threads,
opt.use_borders)
success = align_images(data, opt.work_dir, opt.first_image_is_absolute)
result = "done" if success else "failed"
debug_log("Registration job", result)
def main():
opt = process_command_line()
print opt
ensure_dir(opt.work_dir)
success = register_images(opt.files, opt.crop_size, opt.threads,
opt.work_dir, opt.use_borders,
opt.first_image_is_absolute,
make_jpeg=opt.write_also_jpeg)
result = "done" if success else "failed"
debug_log("Registration job", result)
def rasterize_perim(run_output, perim, year, name, raster=None):
"""!
Convert a perimeter to a raster
@param run_output Folder to save perimeter to
@param perim Perimeter to convert to raster
@param year Year to find reference raster for projection
@param name Name of fire to use for file name
@param raster Specific name of file name to output to
@return Perimeter that was rasterized
@return Path to raster output
"""
prj = os.path.join(run_output,
os.path.basename(perim).replace('.shp', '_NAD1983.shp'))
ensure_dir(os.path.dirname(prj))
ref_NAD83 = osr.SpatialReference()
ref_NAD83.SetWellKnownGeogCS('NAD83')
#~ try:
Project(perim, prj, ref_NAD83)
del ref_NAD83
r = find_best_raster(Extent(prj).XCenter, year)
prj_utm = os.path.join(
run_output,
os.path.basename(perim).replace('.shp',
os.path.basename(r)[9:14] + '.shp'))
Delete(prj_utm)
zone = GetSpatialReference(r)
Project(perim, prj_utm, zone)
del zone
cellsize = GetCellSize(r)
size = 0.0
dataSource = ogr.GetDriverByName('ESRI Shapefile').Open(
prj_utm, gdal.GA_ReadOnly)
layer = dataSource.GetLayer()
for feature in layer:
geom = feature.GetGeometryRef()
area = geom.GetArea()
size += area / (cellsize * cellsize)
del geom
del feature
del layer
del dataSource
if size < 1:
# this is less than one cell in area so don't use perimeter
perim = None
raster = None
else:
if not raster:
raster = os.path.join(run_output, name + '.tif')
Rasterize(prj_utm, raster, r)
return perim, raster
def main():
opt = process_command_line()
print opt
ensure_dir(opt.work_dir)
data = gather_images_data(opt.files, opt.crop_size, opt.threads)
delta_xp = match_ppl_xpl_opaques(data, opt.crop_size, opt.work_dir)
debug_log("PPL->XPL mismatch is", delta_xp)
success = align_images(data, delta_xp, opt.work_dir)
result = "done" if success else "failed"
debug_log("Registration job", result)
def __init__(self, name, no_download=False):
"""!
Constructor
@param self Pointer to this
@param name Name for weather being loaded
@param no_download Whether or not to not download files
"""
## Name for weather being loaded
self.name = name
common.ensure_dir(self.DIR_DATA)
## Folder to save downloaded weather to
self.DIR_DATA = os.path.join(self.DIR_DATA, self.name)
common.ensure_dir(self.DIR_DATA)
## Whether or not to download files
self.no_download = no_download
def get_assets(self):
if self.metadata:
self.get_meta()
print('Verifying assets..')
assets_dir = os.path.join(self.client_root, 'assets')
assets_versions = os.path.join(assets_dir, 'indexes')
ensure_dir(assets_versions)
asset_index = self.metadata['assetIndex']
assets_file = os.path.join(assets_versions,
'%s.json' % (asset_index['id']))
if not os.path.exists(assets_file):
r = requests.get(asset_index['url'], stream=True)
try:
save_to_file_sha1(assets_file, r, asset_index['sha1'])
except Exception:
print('Failed to download assets!')
raise
with open(assets_file) as json_data:
assets = json.load(json_data)
for key, data in assets['objects'].items():
first = data['hash'][0:2]
asset_url = 'http://resources.download.minecraft.net/%s/%s' % (
first, data['hash'])
asset_dir = os.path.join(assets_dir, 'objects', first)
ensure_dir(asset_dir)
asset_file = os.path.join(asset_dir, data['hash'])
if os.path.exists(asset_file):
continue
r = requests.get(asset_url, stream=True)
try:
save_to_file(asset_file, r)
except Exception as e:
print('Failed to download asset %s!' % (key))
raise e
print('All assets verified.')
def pick_otus(dataset, run_id):
"""Pick OTUs using Uclust with Qiime."""
# identify inputs and outputs
set_id = dataset['set_id']
print " ", set_id
run_root = root_dir+set_id+"/"+run_id+"/"
otus_dir = run_root+dirs['otus']
ensure_dir(otus_dir)
master_file = run_root+dirs['merged']+run_id+".fas"
# run the command
comps = ["macqiime", "pick_otus.py", "-i", master_file, "-o", otus_dir]
cline = " ".join(comps)
try:
child = subprocess.Popen(str(cline), stdout=subprocess.PIPE,
shell=True)
output, error = child.communicate()
except: raise
else: print "\t", "OTUs picked"
def download_image(img, window=None) :
if os.name == "nt" :
path = LM_CACHE_PATH + os.sep + 'media' + os.sep + os.sep.join(img.split('/'))
else :
path = LM_CACHE_PATH + os.sep + 'media' + os.sep + img
img_path = 'http://www.pirates-caraibes.com/' + img
try :
ensure_dir(path)
try :
src = urllib2.urlopen(url_fix(img_path))
except urllib2.HTTPError :
return False
else :
dst = open(path, 'wb')
shutil.copyfileobj(src, dst)
return True
except :
return False
def __init__(self, conf, uuid, namespace=None, service=None,
pids_path=None, default_cmd_callback=None,
cmd_addl_env=None, pid_file=None, run_as_root=False):
self.conf = conf
self.uuid = uuid
self.namespace = namespace
self.default_cmd_callback = default_cmd_callback
self.cmd_addl_env = cmd_addl_env
self.pids_path = pids_path or self.conf.external_pids
self.pid_file = pid_file
self.run_as_root = run_as_root
if service:
self.service_pid_fname = 'pid.' + service
self.service = service
else:
self.service_pid_fname = 'pid'
self.service = 'default-service'
common_utils.ensure_dir(os.path.dirname(self.get_pid_file_name()))
def download_images(img_list) :
for img in img_list :
if os.name == "nt" :
path = LM_CACHE_PATH + os.sep + 'media' + os.sep + os.sep.join(img.split('/'))
else :
path = LM_CACHE_PATH + os.sep + 'media' + os.sep + img
img_path = 'http://www.pirates-caraibes.com/' + img
print "Récupération de l'image '%s' vers %s" % (img_path, path)
try :
ensure_dir(path)
try :
src = urllib2.urlopen(img_path)
except urllib2.HTTPError :
pass
else :
dst = open(path, 'wb')
shutil.copyfileobj(src, dst)
print "Succès"
except :
(exctype, value, traceback) = sys.exc_info()
print "Erreur - %s : %s" % (exctype, value)
def iter_align(coord_array, ref_rec, query_rec, aln_dir, segs_file):
"""Iterate through array of coordinates to make pairwise alignments."""
# set up the root subdirectories
seqs = aln_dir+"input_seqs/"
alns = aln_dir+"output_alns/"
ensure_dir([seqs, alns])
aln_id = 0
aln_len = 0
# cycle through segments
for segment_pair in coord_array:
xa, xb, xc, xd = segment_pair
# extract the corresponding sequence slices
ref_seq = ref_rec[abs(xa):abs(xb)]
query_seq = query_rec[abs(xc):abs(xd)]
# reverse-complement sequences with negative sign
if xa < 0 :
ref_seq = ref_seq.reverse_complement()
if xc < 0 :
query_seq = query_seq.reverse_complement()
# write sequences to file
mscl_in = seqs+str(xa)+"_"+str(xb)+"_"+str(xc)+"_"+str(xd)+".fas"
write_fasta(mscl_in, [ref_seq, query_seq])
# skip segments that are too small to align
if abs(abs(xa)-abs(xb)) < 10:
idp = 0
else:
# set up outfiles
mscl_out = alns+str(xa)+"_"+str(xb)+"_"+str(xc)+"_"+str(xd)+".aln"
logfile = aln_dir+"muscle_log.txt"
# perform alignment
align_muscle(mscl_in, mscl_out, logfile)
idntot = parse_clustal_idstars(mscl_out)
idp = int((float(idntot)/len(query_seq))*100)
aln_id += idntot
aln_len += len(query_seq)
# write details out to segments file
line = "\t".join([str(xa), str(xb), str(xc), str(xd), str(idp)+"\n"])
open(segs_file, 'a').write(line)
overall_id = int((float(aln_id)/aln_len)*100)
return overall_id
def save_parameters(dataset, max_pairs, run_id, timestamp):
"""Save a copy of the dataset-specific parameters to file."""
set_id = dataset['set_id']
print " ", set_id
run_root = root_dir+set_id+"/"+run_id+"/"
report_root = run_root+dirs['reports']
param_file = report_root+run_id+"_parameters.txt"
ensure_dir(report_root)
# primer data
primers = dataset['primers']
primers_list = []
for primer_ID in primers:
primers_list.append("\t".join([primer_ID, primers[primer_ID]]))
primers_str = "\n".join(primers_list)
# samples + barcode data
samples = dataset['samples']
samples_list = []
for sample_ID in samples:
samples_list.append("\t".join([sample_ID,
samples[sample_ID][0],
samples[sample_ID][1]]))
samples_str = "\n".join(samples_list)
# text block
txt = ["# Run ID", run_id,
"# Date generated", dataset['date'],
"# Date processing initiated", timestamp,
"# Special processing parameters",
"# read pair length min threshold (ensures overlap)",
str(rp_min_len),
"# max number of read pairs to process",
str(max_pairs),
"# Dataset specifications",
"# Illumina FastQ master files",
dataset['source_fwd'], dataset['source_rev'],
"# Amplification primers", primers_str,
"# Sample ID\tLeft tag\tRight tag", samples_str]
# write to file
open(param_file, 'w').write("\n".join(txt))
print "\t", "Run parameters saved to file"
def map_ref_segs(run_ref, run_id, r_root_dir, run_dirs, min_size,
fct_flags, fct_colors, idpt):
"""Generate map of reference contig with segment details.
This provides a comparison of the original reference and the
re-annotated version.
"""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ori_file = run_ref.file
ref_maps_root = run_root+run_dirs['ref_map_dir']
ensure_dir([ref_maps_root])
gbk_file = run_root+run_dirs['ref_gbk_dir']+ref_n+"_re-annot.gbk"
map_file = ref_maps_root+ref_n+"_ref.pdf"
# start mapping
try:
# make mock segment, full-length with 100% id
record = load_genbank(gbk_file)
length = len(record.seq)
segdata = [[1, length, 1, length, 100]]
# deactivate offsetting
g_offset = (0,0)
q_invert = False
# generate graphical map
pairwise_draw(ref_n+"_ra", ref_n+"_ori", gbk_file, ori_file,
segdata, map_file, q_invert, g_offset, 'dual', 'dual',
'm', 'fct', 'product', min_size, fct_flags,
fct_colors, idpt)
except IOError:
msg = "\nERROR: could not load segments data"
run_ref.log(msg)
print msg
except StopIteration:
msg = "\nERROR: could not make map"
run_ref.log(msg)
print msg
def basic_batch_blast(genomes, run_ref, blast_mode, r_root_dir, run_dirs,
fixed_dirs, blast_prefs, run_id, timestamp):
"""Send batch jobs to Blast. Muxes to multiple reference DBs."""
# load inputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
in_root = run_root+run_dirs['ref_seg_dir']+ref_n+"/"
print " ", ref_n
# log
logstring = "".join(["\n\n# Blast segs to genomes @", timestamp, "\n\n"])
run_ref.log(logstring)
# do blast
for seg in run_ref.segs:
input_file = in_root+ref_n+"_"+seg['name']+".fas"
# translate if required
if blast_mode == 'tn':
record = load_fasta(input_file)
record.seq = record.seq.translate()
input_file = in_root+ref_n+"_"+seg['name']+"_aa.fas" # substitute
write_fasta(input_file, record)
out_dir = run_root+run_dirs['blast_out_dir']+ref_n+"/"+seg['name']+"/"
ensure_dir([out_dir])
print "\t", seg['name'],
for genome in genomes:
g_name = genome['name']
db_path = fixed_dirs['blast_db_dir']+g_name
outfile = out_dir+g_name+"_out.txt"
print ".",
if blast_mode == 'n':
local_blastn_2file(input_file, db_path, outfile, blast_prefs)
elif blast_mode == 'tx':
local_tblastx_2file(input_file, db_path, outfile, blast_prefs)
elif blast_mode == 'tn':
local_tblastn_2file(input_file, db_path, outfile, blast_prefs)
print ""
run_ref.log("All OK")
return "OK"
def annot_genome_contigs(run_ref, prot_db_name, fixed_dirs, r_root_dir,
run_id, run_dirs, genomes, project_id, timestamp,
blast_prefs):
"""Annotate genome contigs (predict ORFs and assign function)."""
# locate the COG database
prot_db = fixed_dirs['ref_dbs_dir']+prot_db_name
# TODO: add other DB / pfams?
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
fas_ctgs_root = run_root+run_dirs['match_out_dir']+ref_n+"/"
ctg_cds_root = fixed_dirs['ctg_cds_dir']
ctg_prot_root = fixed_dirs['ctg_prot_dir']
ctg_blast_root = fixed_dirs['ctg_blast_dir']
g_gbk_ctgs_root = fixed_dirs['gbk_contigs_dir']
r_gbk_ctgs_root = run_root+run_dirs['run_gbk_ctgs_dir']+ref_n+"/"
annot_trn_root = fixed_dirs['annot_trn_dir']
print " ", ref_n
# log
logstring = "".join(["\n\n# Annotate genome contigs @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs
g_name = genome['name']
fas_ctgs_dir = fas_ctgs_root+g_name+"/"
g_file = fixed_dirs['ori_g_dir']+genome['file']
print '\t', g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# set output files
training_file = annot_trn_root+g_name+"_annot.trn"
# set output dirs
ctg_cds_dir = ctg_cds_root+g_name+"/"
ctg_prot_dir = ctg_prot_root+g_name+"/"
ctg_blast_dir = ctg_blast_root+g_name+"/"
g_gbk_ctgs_dir = g_gbk_ctgs_root+g_name+"/"
r_gbk_ctgs_dir = r_gbk_ctgs_root+g_name+"/"
ensure_dir([ctg_cds_dir, ctg_prot_dir, ctg_blast_dir,
g_gbk_ctgs_dir, r_gbk_ctgs_dir])
# list fasta files in matches directory
dir_contents = listdir(fas_ctgs_dir)
for item in dir_contents:
pattern = re.compile(r'.*_(\d*)\.fas$')
match = pattern.match(item)
if match:
ctg_num = match.group(1)
print ctg_num,
logstring = "".join(["\t", ctg_num])
run_ref.log(logstring)
# set inputs and outputs
ctg_fas = fas_ctgs_dir+item
g_ctg_gbk = g_gbk_ctgs_dir+g_name+"_"+ctg_num+".gbk"
r_ctg_gbk = r_gbk_ctgs_dir+g_name+"_"+ctg_num+".gbk"
annot_gbk = ctg_cds_dir+g_name+"_"+ctg_num+"_cds.gbk"
annot_aa = ctg_prot_dir+g_name+"_"+ctg_num+"_aa.fas"
blast_out = ctg_blast_dir+g_name+"_"+ctg_num+".xml"
if path.exists(blast_out) and os.stat(blast_out)[6]==0:
os.remove(blast_out)
if not path.exists(r_ctg_gbk):
if not path.exists(g_ctg_gbk):
l_tag_base = g_name+"_"+ctg_num
record = annot_ctg(g_file, ctg_fas, annot_gbk,
annot_aa, training_file, prot_db,
blast_out, l_tag_base, blast_prefs)
record.description = g_name+"_"+ctg_num
record.name = g_name+"_"+ctg_num
record.dbxrefs = ["Project: "+project_id+"/"+ref_n
+"-like backbones"]
record.seq.alphabet = generic_dna
write_genbank(g_ctg_gbk, record)
copyfile(g_ctg_gbk, r_ctg_gbk)
print ""
def batch_contig_annot(dataset):
"""Extract and annotate contigs."""
# identify dataset contig file
contigs_file = dirs['assembly_dir']+dataset['f_nick']+'/'+'contigs.fa'
# locate the COG database
cog_db = dirs['blast_db_dir']+'Cog_LE/Cog'
# make the training file
training_file = dirs['annot_dir']+dataset['f_nick']+'/'+'contigs.trn'
#train_prodigal(contigs_file, training_file)
# set output dirs
fas_out_dir = dirs['annot_dir']+dataset['f_nick']+'/fasta/'
gbk_out_dir = dirs['annot_dir']+dataset['f_nick']+'/predict/'
aa_out_dir = dirs['annot_dir']+dataset['f_nick']+'/aa/'
blast_out_dir = dirs['annot_dir']+dataset['f_nick']+'/rpsblast/'
solid_out_dir = dirs['annot_dir']+dataset['f_nick']+'/genbank/'
maps_out_dir = dirs['annot_dir']+dataset['f_nick']+'/maps/'
ensure_dir(fas_out_dir)
ensure_dir(gbk_out_dir)
ensure_dir(aa_out_dir)
ensure_dir(blast_out_dir)
ensure_dir(solid_out_dir)
# set phage hit collector
contig_hits = {}
sp_hit_list = dirs['annot_dir']+dataset['f_nick']+'/'\
+dataset['f_nick']+'_kw_hits.html'
all_hit_list = dirs['annot_dir']+dataset['f_nick']+'/'\
+dataset['f_nick']+'_all_hits.html'
sp_hit_list_handle = open(sp_hit_list, 'w')
all_hit_list_handle = open(all_hit_list, 'w')
sp_hit_list_handle.write("<ul>")
all_hit_list_handle.write("<ul>")
# load all contigs
contigs_list = load_multifasta(contigs_file)
# cycle through contigs
ctg_count = 0
gene_count = 0
for contig in contigs_list:
ctg_count +=1
# use regex to acquire relevant record ID info
pattern = re.compile(r'NODE_(\d*)_length_(\d*)_cov_(\d*)')
match = pattern.match(contig.id)
nick = match.group(1)+'_'+match.group(2)+'_'+match.group(3)
contig.id = nick
fasta_out = fas_out_dir+nick+'.fas'
# write record to file
write_fasta(fasta_out, contig)
# create contig entry in dict
contig_hits[nick] = []
# run the annotation
annot_gbk = gbk_out_dir+nick+'.gbk'
annot_aa = aa_out_dir+nick+'.fas'
#run_prodigal(fasta_out, annot_gbk, annot_aa, training_file)
# blast the amino acids against COG
print '\tblasting', dataset['f_nick'], nick
blast_out = blast_out_dir+nick+'.xml'
if path.isfile(blast_out):
print "\t\talready blasted"
else:
local_rpsblast_2file(annot_aa, cog_db, blast_out, blast_prefs)
# collect best hits
rec_cogs = collect_cogs(blast_out)
map_file = maps_out_dir+nick+'.pdf'
# consolidate annotated genbank file
record = load_fasta(fasta_out)
aa_defs = load_multifasta(annot_aa)
features = []
counter = 1
ctg_flag_1 = 0
ctg_flag_2 = 0
for protein in aa_defs:
gene_count +=1
# get feature details from description line
# necessary because the prodigal output is not parser-friendly
pattern = re.compile(r'\d+_\d+_\d+_\d+_\d+\s+\S+\s+(\d+)\s+\S+\s+(\d+)\s+\S+\s+(\S*\d)')
match = pattern.match(protein.description)
start_pos = int(match.group(1))
end_pos = int(match.group(2))
strand_pos = int(match.group(3))
feat_loc = FeatureLocation(start_pos, end_pos)
annotation = rec_cogs['Query_'+str(counter)]
if ctg_flag_1 is 0:
all_hit_list_handle.write("</ul><br><a href='"
+"../../../../"
+map_file
+"'>Contig "
+nick+"</a><ul>")
ctg_flag_1 = 1
all_hit_list_handle.write("<li>"+str(counter)
+'. '+annotation+"</li>")
# detect phage content in annotation
phi_pattern = re.compile(r".+(COG\d+).+"
"(phage|capsid|muramidase|tail|"
"replication|helicase|polymerase|"
"integrase|recombinase"
"suppressor|hydrolase|transposase).+",
re.IGNORECASE)
phi_match = phi_pattern.match(annotation)
if phi_match:
hit_flag = 'on'
hit_dict = {'CDS': counter,
'annot': annotation,
'COGs': phi_match.group}
contig_hits[nick].append(hit_dict)
# write out to summary file
if ctg_flag_2 is 0:
sp_hit_list_handle.write("</ul><br><a href='"
+"../../../../"
+map_file
+"'>Contig "
+nick+"</a><ul>")
ctg_flag_2 = 1
sp_hit_list_handle.write("<li>"+str(counter)
+'. '+annotation+"</li>")
else:
hit_flag = 'off'
# consolidation feature annotations
quals = {'note': protein.description,
'fct': annotation,
'flag': hit_flag}
feature = SeqFeature(location=feat_loc,
strand=strand_pos,
id=protein.id,
type='CDS',
qualifiers=quals)
features.append(feature)
counter +=1
record.features = features
record.description = dataset['f_nick']+'_contig_'+nick
record.name = nick
record.dbxrefs = ['Project:np1']
record.seq.alphabet = generic_dna
gbk_out = solid_out_dir+nick+'.gbk'
write_genbank(gbk_out, record)
# generate graphical map
ContigDraw(nick, gbk_out, map_file)
sp_hit_list_handle.write("</ul>")
all_hit_list_handle.write("</ul>")
sp_hit_list_handle.close()
all_hit_list_handle.close()
print "\t", gene_count, "predicted genes in", ctg_count, "contigs"
def demux_illumina(dataset, max_pairs, run_id):
"""Demultiplex Illumina dataset.
From separate forward/reverse read sets, combine read pairs and output
to separate files for each sample based on barcode tags. As part of the
process, reject read pairs that have mismatching tags or primers and trim
the rest, removing primer+tag and low-quality sequences.
"""
# identify inputs and outputs
set_id = dataset['set_id']
print " ", set_id
run_root = root_dir+set_id+"/"+run_id+"/"
ori_root = root_dir+set_id+"/"+dirs['master']
fwd_file = ori_root+dataset['source_fwd']
rev_file = ori_root+dataset['source_rev']
demux_root = run_root+dirs['demux']
report_root = run_root+dirs['reports']
qc_dir = "qc_details/"
qc_main_file = report_root+"quality_control.html"
cntsplt = report_root+"sample_counts"
ensure_dir(ori_root)
ensure_dir(demux_root)
ensure_dir(report_root)
ensure_dir(report_root+qc_dir)
# set up files for reporting
html_comps = ["<p><b>Quality control for run "+run_id+"</b></p>",
"<p><img src='sample_counts.png' alt='sample_counts'/></p>",
"<p><table border='1'><tr>",
"<th>Sample</th>",
"<th>Accepted</th>",
"<th>Rejected</th>",
"<th>Total</th>",
"<th>% OK</th></tr>"]
html_block = "".join(html_comps)
open(qc_main_file, 'w').write(html_block)
# prepare primers and barcodes info
primers = dataset['primers']
samples = dataset['samples']
tag_pairs = samples.values()
assert len(primers) >= 2
assert len(samples) >= 1
assert len(tag_pairs) >= 1
# prepare container and files for output batching and reporting
hits_dict = {}
for sample_id in samples:
hits_dict[sample_id] = {'buffer': [], 'countY': 0, 'countN': 0}
# add containers for rejected read pairs
hits_dict['bad_tags'] = {'buffer': [], 'countY': 0, 'countN': 0}
hits_dict['bad_qual'] = {'buffer': [], 'countY': 0, 'countN': 0}
# initialize files
for sample_id in samples:
dmx_out = demux_root+sample_id+"_readpairs.txt"
open(dmx_out, 'w').write('')
open(demux_root+"bad_tags"+"_readpairs.txt", 'w').write('')
open(demux_root+"bad_qual"+"_readpairs.txt", 'w').write('')
# iterate through reads
pair_count = 0
for titles, seqs, quals in FastqJointIterator(open(fwd_file),
open(rev_file)) :
F_title = titles[0][0]
R_title = titles[0][1]
F_seq = seqs[0][0].upper()
R_seq = seqs[0][1].upper()
F_qual = quals[0][0]
R_qual = quals[0][1]
flip = False
sample_id = False
# iterate through barcode tags
# TODO: implement more robust solution to ambiguous base problem
for tag_pair in tag_pairs:
L_tag1 = (tag_pair[0]+primers['fwdRA']).upper()
L_tag2 = (tag_pair[0]+primers['fwdRG']).upper()
R_tag = (tag_pair[1]+primers['rev']).upper()
tag_hit = False
while True:
# start by checking For R_tag since there's only one
if not R_seq.find(R_tag, 0, len(R_tag)) is 0:
if not F_seq.find(R_tag, 0, len(R_tag)) is 0:
# no R_tag match -> reject
break
else: # is there an L_tag in R_seq?
while True:
if not R_seq.find(L_tag1, 0, len(L_tag1)) is 0:
if not R_seq.find(L_tag2, 0, len(L_tag2)) is 0:
# no L_tag match -> reject
break
else:
R_clip = len(L_tag2)
else:
R_clip = len(L_tag1)
tag_hit = True
flip = True
F_clip = len(R_tag)
break
else: # is there an L_tag in F_seq?
while True:
if not F_seq.find(L_tag1, 0, len(L_tag1)) is 0:
if not F_seq.find(L_tag2, 0, len(L_tag2)) is 0:
# no L_tag match -> reject
break
else:
F_clip = len(L_tag2)
else:
F_clip = len(L_tag1)
tag_hit = True
R_clip = len(R_tag)
break
break
if not tag_hit: # continue iterating
sample_id = False
else: # got it, stop iterating
sample_id = key_by_value(samples, tag_pair)[0]
break
# in case no matches were found with any of the tags
if not sample_id:
sample_id = 'bad_tags'
# for matched read pairs, clip off tag+primer and strip low qual runs
else:
F_trim = F_qual[F_clip:].find('##')
if F_trim > -1:
F_seq = F_seq[F_clip:F_clip+F_trim]
F_qual = F_qual[F_clip:F_clip+F_trim]
else:
F_seq = F_seq[F_clip:]
F_qual = F_qual[F_clip:]
R_trim = R_qual[R_clip:].find('##')
if R_trim > -1:
R_seq = R_seq[R_clip:R_clip+R_trim]
R_qual = R_qual[R_clip:R_clip+R_trim]
else:
R_seq = R_seq[R_clip:]
R_qual = R_qual[R_clip:]
if len(F_seq)+len(R_seq) < rp_min_len:
# increment sample hit 'No' counter
hits_dict[sample_id]['countN'] +=1
sample_id = 'bad_qual'
# bundle read data in ordered string
readF = str("@%s\n%s\n+\n%s\n" % (F_title, F_seq, F_qual))
readR = str("@%s\n%s\n+\n%s\n" % (R_title, R_seq, R_qual))
if flip:
read_pair = readR+readF
else:
read_pair = readF+readR
# output to the appropriate buffer
hits_dict[sample_id]['buffer'].append(read_pair)
# increment sample 'Yes' hit counter
hits_dict[sample_id]['countY'] +=1
# when buffer capacity is reached, output to file and reset buffer
if hits_dict[sample_id]['countY']% 100000==0:
dmx_out = demux_root+sample_id+"_readpairs.txt"
dump_buffer(dmx_out, hits_dict[sample_id]['buffer'])
hits_dict[sample_id]['buffer'] = []
# increment counter
pair_count +=1
# report on the progress
if pair_count%1000000==0:
print "\t", pair_count, "reads processed", datetime.now()
if pair_count == max_pairs: # for testing purposes
break
print "\t", "Total", pair_count, "read pairs processed"
print "\t", "Counts per sample:"
# prepare graphing data containers
pcntY = []
pcntN = []
sample_ids = []
# write out whatever remains in each of the samples buffers
for sample_id in samples:
dmx_out = demux_root+sample_id+"_readpairs.txt"
dump_buffer(dmx_out, hits_dict[sample_id]['buffer'])
hits_dict[sample_id]['buffer'] = []
acc = hits_dict[sample_id]['countY']
rej = hits_dict[sample_id]['countN']
print "\t\t", sample_id, acc, "pairs", datetime.now()
pcntY.append(acc)
pcntN.append(rej)
sample_ids.append(sample_id)
# generate FastQC report (use --noextract to not open zipped reports)
run_FastQC(dmx_out, report_root+qc_dir, '--quiet', ' ')
#print "see QC report"
# add line in QC file
link = qc_dir+sample_id+"_readpairs_fastqc/fastqc_report.html"
html_comps = ["<tr>",
"<th><a href='"+link+"'>"+sample_id+"</a></th>",
"<td>", str(acc), "</td>",
"<td>", str(rej), "</td>",
"<td>", str(acc+rej), "</td>",
"<td>", str(int((float(acc)/(acc+rej))*100)),
"</td></tr>"]
html_block = "".join(html_comps)
open(qc_main_file, 'a').write(html_block)
# write out whatever remains in the bad_qual buffer
dmx_out = demux_root+"bad_qual_readpairs.txt"
dump_buffer(dmx_out, hits_dict['bad_qual']['buffer'])
hits_dict['bad_qual']['buffer'] = []
print "\t\t", "rejected (low quality)", hits_dict['bad_qual']['countY'],\
datetime.now()
# generate FastQC report (use --noextract to not open zipped reports)
run_FastQC(dmx_out, report_root+qc_dir, '--quiet', ' ')
#print "see QC report"
# add line in QC file
link = qc_dir+"bad_qual_readpairs_fastqc/fastqc_report.html"
html_comps = ["<tr>",
"<th><a href='"+link+"'>"+"bad_qual"+"</a></th>",
"<td>", '0', "</td>",
"<td>", str(hits_dict['bad_qual']['countY']), "</td>",
"<td>", str(hits_dict['bad_qual']['countY']), "</td>",
"<td>", '0',"</td></tr>"]
html_block = "".join(html_comps)
open(qc_main_file, 'a').write(html_block)
# write out whatever remains in the bad_tags buffer
dmx_out = demux_root+"bad_tags_readpairs.txt"
dump_buffer(dmx_out, hits_dict['bad_tags']['buffer'])
hits_dict['bad_tags']['buffer'] = []
print "\t\t", "rejected (bad tags)", hits_dict['bad_tags']['countY'],\
datetime.now()
# generate FastQC report (use --noextract to not open zipped reports)
run_FastQC(dmx_out, report_root+qc_dir, '--quiet', ' ')
#print "see QC report"
# add line in QC file
link = qc_dir+"bad_tags_readpairs_fastqc/fastqc_report.html"
html_comps = ["<tr>",
"<th><a href='"+link+"'>"+"bad_tags"+"</a></th>",
"<td>", '0', "</td>",
"<td>", str(hits_dict['bad_tags']['countY']), "</td>",
"<td>", str(hits_dict['bad_tags']['countY']), "</td>",
"<td>", '0',"</td></tr>"]
html_block = "".join(html_comps)
open(qc_main_file, 'a').write(html_block)
# close table and add notes
line_bq = "rejected after demultiplexing due to low sequence quality \
(top stacks in bar chart)"
line_bt = "could not be assigned to a sample due to mismatches in tag \
and/or primer"
html_comps = ["</table></p>",
"<p><b>", "bad_qual", ": </b>", line_bq,
"<br><b>", "bad_tags", ": </b>", line_bt, "</p>",]
html_block = "".join(html_comps)
open(qc_main_file, 'a').write(html_block)
# add bad tags category for counts graphing (switch is on purpose)
pcntY.append(hits_dict['bad_tags']['countN'])
pcntN.append(hits_dict['bad_tags']['countY'])
sample_ids.append('bad_tags')# check that the totals add up
assert pair_count == sum(pcntY)+sum(pcntN)
# plot the read counts per sample
series = pcntY, pcntN
legend = 'Accepted', 'Rejected'
colors = 'g', 'r'
titles = 'Number of read pairs', 'Read pairs per sample'
two_storey_bar_chart(series, sample_ids, legend, colors, cntsplt, titles)
def merge_pair_libs(dataset, run_id):
"""Merge read pairs from Illumina sample libs and output FastA."""
# identify inputs and outputs
set_id = dataset['set_id']
print " ", set_id
run_root = root_dir+set_id+"/"+run_id+"/"
dmx_root = run_root+dirs['demux']
merged_root = run_root+dirs['merged']
report_root = run_root+dirs['reports']
master_file = run_root+dirs['merged']+run_id+".fas"
ensure_dir(merged_root)
ensure_dir(report_root)
merger_file = report_root+"merged_pairs.html"
cntsplt = report_root+"merge_counts"
samples = dataset['samples']
# set up files for reporting
html_comps = ["<p><b>Read pairs merged for run ", run_id, "</b></p>",
"<p><img src='merge_counts.png' alt='merge_counts'/></p>",
"<p><table border='1'><tr>",
"<th>Sample</th>",
"<th>Accepted</th>",
"<th>Rejected</th>",
"<th>Total</th>",
"<th>% OK</th></tr>"]
html_block = "".join(html_comps)
open(merger_file, 'w').write(html_block)
# initialize master file
open(master_file, 'w').write('')
# merge per sample (demuxed)
merge_countA = []
merge_countR = []
sample_ids = samples.keys()
for sample_id in sample_ids:
print "\t", sample_id,
lib_file = dmx_root+sample_id+"_readpairs.txt"
merge_out = merged_root+sample_id+"_merged.fas"
open(merge_out, 'w').write('')
# prepare container and files for output batching and reporting
buffer = []
countY = 0
countF = 0
countN = 0
# iterate through the read pairs
count = 0
for titles, seqs, quals in FastqGGIterator(open(lib_file)):
count +=1
seq1 = seqs[0]
seq2 = seqs[1]
qual1 = quals[0]
qual2 = quals[1]
# merge reads TODO: better safeguard against merge failure
try: merged = merge_overlaps(seq1, qual1, seq2, qual2)
except: countF +=1
else:
if merged.find('N') > -1:
countN +=1 # if there are still N quality must be too low
else:
countY +=1
# compose string for output
mcomps = [">",sample_id,"_",str(count),"\n",merged,"\n"]
mstring = "".join(mcomps)
# output to buffer
buffer.append(mstring)
# when buffer capacity is reached, output to file and reset buffer
if countY % 10000==0:
dump_buffer(merge_out, buffer)
dump_buffer(master_file, buffer)
buffer = []
# write out whatever remains in the buffer
dump_buffer(merge_out, buffer)
dump_buffer(master_file, buffer)
# sum up
assert countY+countF+countN == count
print count, "pairs", datetime.now()
print "\t\t", str(countY), "merged and accepted"
print "\t\t", str(countN), "merged but rejected due to residual Ns"
print "\t\t", str(countF), "failed to merge"
# add line in QC file
html_comps = ["<tr>",
"<th>", sample_id, "</b></th>",
"<td>", str(countY), "</td>",
"<td>", str(countN + countF), "</td>",
"<td>", str(count), "</td>",
"<td>", str(int((float(countY)/count)*100)),
"</td></tr>"]
html_block = "".join(html_comps)
open(merger_file, 'a').write(html_block)
# pass values
merge_countA.append(countY)
merge_countR.append(countN+countF)
# close table and add notes
line_N = "either failed to merge or still contained Ns after merging"
html_comps = ["</table></p>",
"<p><b>", "Rejected", ":</b> ", line_N, "</p>"]
html_block = "".join(html_comps)
open(merger_file, 'a').write(html_block)
# plot the read counts per sample
series = merge_countA, merge_countR
lgnd = 'Accepted', 'Rejected'
colors = 'g', 'r'
titles = 'Number of read pairs', 'Read pairs merged per sample'
two_storey_bar_chart(series, sample_ids, lgnd, colors, cntsplt, titles)
def process_ref(ref, ref_annot_flag, r_root_dir, fixed_dirs, run_dirs,
run_id, timestamp, prot_db_name, project_id):
"""Re-annotate contig and extract reference segments using coordinates."""
# set inputs and outputs
run_root = r_root_dir+run_id+"/"
ref_name = ref['name']
in_file = fixed_dirs['ori_g_dir']+ref['file']
seg_out_root = run_root+run_dirs['ref_seg_dir']+ref_name+"/"
gen_fas_root = fixed_dirs['fas_contigs_dir']+ref_name+"/"
if ref_annot_flag:
ref_gbk = run_root+run_dirs['ref_gbk_dir']+ref_name+"_re-annot.gbk"
else: ## bypass re-annotated ONLY IF ORIGINAL INPUT IS GBK #todo: fix
ref_gbk = in_file
ref_fas = run_root+run_dirs['ref_fas_dir']+ref_name+".fas"
genome_fas = gen_fas_root+ref_name+"_1.fas"
report_root = run_root+run_dirs['reports']+ref_name+"/"
ref_log = report_root+run_id+"_"+ref_name+"_log.txt"
ensure_dir([seg_out_root, report_root, gen_fas_root])
print " ", ref_name, "...",
# initialize run_ref object
run_ref = Reference(ref_name, in_file, ref['input'], ref['seg_mode'],
ref['capture'], ref_fas, ref_gbk, seg_out_root,
ref_log)
# initialize reference log
cl_header = ["# Console log:", run_id, "/", ref_name, timestamp, "\n\n"]
open(ref_log, 'w').write(" ".join(cl_header))
# open record and ensure we have a fasta in the right place
if not path.exists(ref_fas):
if run_ref.input == 'fas':
copyfile(in_file, ref_fas)
elif run_ref.input == 'gbk':
record = load_genbank(in_file)
record.id = ref_name
write_fasta(ref_fas, record)
else:
msg = "ERROR: Input not recognized for "+ref_name
run_ref.log(msg)
raise Exception(msg)
# make a BLAST DB
make_ref_DB(ref, run_id, fixed_dirs, r_root_dir, run_dirs)
copyfile(ref_fas, genome_fas)
# re-annotate ref contig
if ref_annot_flag:
record = annot_ref(ref_name, ref_fas, prot_db_name, fixed_dirs,
project_id)
else: ## bypass re-annotation ONLY IF ORIGINAL INPUT IS GBK #todo: fix
record = load_genbank(in_file)
# load or generate segment definitions
if run_ref.seg_mode == 'chop':
run_ref.get_segs_from_chop(len(record.seq), ref['chop_size'])
elif run_ref.seg_mode == 'list':
run_ref.get_segs_from_list(ref['segs'])
elif run_ref.seg_mode == 'feats':
run_ref.get_segs_from_feats(ref['feat_type'])
# extract segment sequences
rec_annot = run_ref.extract_segs_seqs(record, seg_out_root)
# write re-annotated reference sequence to file
write_genbank(ref_gbk, rec_annot)
# report results
logstring = " ".join([str(len(run_ref.segs)), "segments"])
print logstring
run_ref.log(logstring)
return run_ref
def unpack_genomes(genome, separator, fixed_dirs, ctg_thresholds):
"""Unpack genome files.
Here, unpacking means extracting data and producing specific files to
standardize how the information is made available to downstream analysis.
Depending on the input file format, different unpacking methods are
invoked. In all cases, this ensures that for each genome, there is a
multifasta file of the contigs all together as well as a separate Genbank
file for each contig.
Supported input file formats are the following:
- mfas: Basic whole genome sequence in multifasta file of contigs. This
can be used to process a finished genome in a single Fasta file as well.
- cgbk: All contigs concatenated in a single GenBank file (Genoscope,
French WGS). This can be used to process a finished genome in a single
GanBank file as well.
# TODO: provide support for other possible input formats
Unpacking 'cgbk' genomes involves an initial step to detect occurrences
of the sequence separator and collect the start and stop coordinates of
each contig. Each pair of coordinates can then be used to extract the
contig sequence and create a SeqRecord for that contig, which SeqIO
normally does when it unpacks multifasta files.
"""
# set up inputs
infile = genome['file'] #TODO: make GUI input loader (upstream)
inpath = fixed_dirs['ori_g_dir']+infile
g_name = genome['name']
print " ", g_name, "...",
# prep output destinations
mfas_dir = fixed_dirs['mfas_contigs_dir']
fas_dir = fixed_dirs['fas_contigs_dir']+g_name+"/"
ensure_dir([mfas_dir, fas_dir])
mfas_file = mfas_dir+g_name+"_contigs.fas"
records = []
# select unpacking method
if genome['input'] is 'fas':
try: path.exists(inpath) is True
except ValueError: raise Exception("Bad input file path")
genome_recs = load_multifasta(inpath)
# generate GenBank files
counter = 0
for rec in genome_recs:
counter +=1
ctg_num = str(counter)
new_id = g_name+"_"+ctg_num # workaround for long ids
new_seq = rec.seq
new_seq.alphabet = generic_dna
new_rec = SeqRecord(seq=new_seq, id=new_id)
records.append(new_rec) # for multifasta output
fas_file = fas_dir+new_id+".fas"
write_fasta(fas_file, new_rec)
elif genome['input'] is 'gbk':
# load in genome data
genome_rec = load_genbank(inpath)
g_string = genome_rec.seq
# find split coordinates
coord_pairs = multisplit_finder(g_string, separator)
# split record
counter = 0
for (start, stop) in coord_pairs:
counter +=1
ctg_num = str(counter)
new_record = genome_rec[start:stop]
new_record.id = g_name+"_"+ctg_num
records.append(new_record) # for multifasta output
fas_file = fas_dir+g_name+"_"+ctg_num+".fas"
write_fasta(fas_file, new_record)
else:
xmsg = "Input file format "+genome['input']+" unspecified/unsupported"
raise Exception(xmsg)
print counter, "contigs"
# write master file
write_fasta(mfas_file, records)
# pass records to stats logger
ctg_stats(g_name, fixed_dirs, ctg_thresholds, records)
def build_scaffolds(run_ref, r_root_dir, run_dirs, prox_D, separator,
genomes, run_id, timestamp, mtype, mode):
"""Build a scaffold of contigs based on the reference.
This takes contigs that gave positive hits when blasted with reference
segments. The contigs were aligned against the complete reference in a
previous step for mapping purposes. Now the output of that step is re-used
determine their position. A caveat is that if there are natural local
rearrangements in the sequence relative to the reference, they may not be
resolved appropriately. The problem is somewhat moderated by the fact that
this function takes the best (usually the largest) hit region as "anchor"
to position the contig within the scaffold. But if the rearranged region
takes up a significant portion of the contig length, the anchoring will
probably not be called correctly. Visual inspection of the finalized
maps should help diagnose any such problems. The order can be fixed
manually using the Mauve Contig Mover, which is part of Mauve 2.
Note that not all hit contigs are "real" hits, so filtering should be
applied before scaffolding to generate constructs.
Model-based filtering produces a list of contigs that will be passed to
the scaffolder. If filtering manually by looking at the maps,
there are two options available: either select exclusively OR exclude a
subset of contigs for the scaffolding process. This is done by listing
their ID number in the genome dictionaries in the config file then
resuming the pipeline from this step.
"""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ctgs_root = run_root+run_dirs['run_gbk_ctgs_dir']+ref_n+"/"
mauve_root = run_root+run_dirs['mauve_out_dir']+ref_n+"/contigs/"
scaffolds_dir = run_root+run_dirs['scaffolds_dir']+ref_n+"/"
print " ", ref_n
# log
logstring = "".join(["\n\n# Build scaffold constructs @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs
g_name = genome['name']
ctgs_dir = ctgs_root+g_name+"/"
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# set outputs
mauve_dir = mauve_root+g_name+"/"
ensure_dir([mauve_dir, scaffolds_dir])
scaff_fas = scaffolds_dir+g_name+"_"+ref_n+"_scaffold.fas"
scaff_gbk = scaffolds_dir+g_name+"_"+ref_n+"_scaffold.gbk"
# list genbank files in matches directory
dir_contents = listdir(ctgs_dir)
anchors_array = np.zeros(1, dtype=[('ctg', 'i4'),
('start', 'i4'),
('end', 'i4'),
('orient', 'i2')])
# identify contigs we want to select
subset = []
for item in dir_contents:
pattern = re.compile(r'.*_(\d*)\.gbk$')
match = pattern.match(item)
if match:
ctg_num = match.group(1)
if mode == "exclude":
try:
if int(ctg_num) in genome[mode]:
msg = "("+ctg_num+")"
print msg,
run_ref.log(msg)
else:
subset.append(ctg_num)
except KeyError:
msg = "WARNING: no ignored segments list, including all"
print msg
msg = ctg_num
print msg,
subset.append(ctg_num)
run_ref.log(msg)
elif mode == "select":
try:
if int(ctg_num) in genome[mode]:
msg = ctg_num
print msg,
run_ref.log(msg)
subset.append(ctg_num)
else:
msg = "("+ctg_num+")"
print msg,
run_ref.log(msg)
except KeyError:
msg = "WARNING: no selected segments list, including all"
print msg
msg = ctg_num
print msg,
subset.append(ctg_num)
run_ref.log(msg)
# at this point we should have a subset of contigs selected
for ctg_num in subset:
logstring = "".join(["\t", ctg_num])
run_ref.log(logstring)
# set inputs
mauve_file = mauve_dir+ctg_num+".mauve"
bb_file = mauve_file+".backbone"
try:
# parse Mauve output
coords = mauver_load2_k0(bb_file, prox_D, mtype)
# determine which segment to use as anchor
anchor_seg = get_anchor_loc(coords)
anchors_array = np.insert(anchors_array, 0,
(ctg_num,
anchor_seg['start'],
anchor_seg['end'],
anchor_seg['orient']))
except IOError:
msg = "\tERROR: Mauve alignment not found\n\t"
print msg
run_ref.log(msg)
except Exception:
msg = "\tERROR: Iteration failure\n\t"
print msg
run_ref.log(msg)
# abort if there is no valid contig to proceed with
try:
assert len(anchors_array) > 1 # always 1 left from stub
except AssertionError:
msg = "\tWARNING: Contig list empty\n\t"
print msg
run_ref.log(msg)
else:
# order contigs by anchor location
anchors_array = np.sort(anchors_array, order='start')
# load contig records from the genbank files in the matches directory
ctg_list = []
for ctg_anchor in anchors_array:
ctg_num = ctg_anchor['ctg']
if ctg_num > 0:
contig_gbk = ctgs_dir+g_name+"_"+str(ctg_num)+".gbk"
record = load_genbank(contig_gbk)
if ctg_anchor['orient'] == -1: # flip record
record = record.reverse_complement(id=True, name=True,
annotations=True, description=True)
ctg_list.append(record)
else: # workaround for having 0 value leftover from stub
pass # having it might come in handy in later dev
# output scaffold files
write_fasta(scaff_fas, ctg_list)
scaff_record = SeqRecord('', id='temp')
scaff_bumper = SeqRecord(separator, id='join')
for record in ctg_list:
feat_start = len(scaff_record.seq)
scaff_record += record
feat_stop = len(scaff_record.seq)
scaff_record += scaff_bumper
feat_loc = FeatureLocation(feat_start, feat_stop)
pattern = re.compile(r'.*_(\d*)$')
match = pattern.match(record.id)
try: ctg_num = match.group(1)
except Exception: ctg_num = 'N'
feature = SeqFeature(location=feat_loc,
type='contig',
qualifiers={'id': ctg_num})
scaff_record.features.append(feature)
scaff_record.description = g_name+" scaffold from "+ref_n
try:
scaff_record.id = g_name
write_genbank(scaff_gbk, scaff_record[:-100]) # rm last bumper
except ValueError:
scaff_record.id = g_name[:10]
write_genbank(scaff_gbk, scaff_record[:-100]) # rm last bumper
print ""
def glomp_blast_out(dataset, ref_nick):
"""Consolidate Blast output files."""
# Identify the genome
nickname = dataset['nickname']
# Determine the input file root
root_dir = dirs['blast_out_dir']+nickname+"/"+ref_nick+"/"
file_root = root_dir+nickname
# Signal process start
print "-- Consolidating B_out for", nickname, "against", ref_nick, "--"
print datetime.now()
# Cycle through bin types
series_index = 0
averages = [] # for comparing series later
binned_pos = []
for bin_type in bin_types:
index = 1
bin_arrays =[]
while os.path.isfile(file_root+bin_type+"_"+str(index)+"_blast.out"):
infile = file_root+bin_type+"_"+str(index)+"_blast.out"
rec_array = read_array(infile, blast_dtypes)
if len(rec_array) > 0:
bin_arrays.append(rec_array)
index +=1
print "\t\t"+str(len(bin_arrays)), "arrays for", \
nickname+bin_type, "series"
if len(bin_arrays) > 0:
series = numpy.hstack(bin_arrays)
else:
series = []
print "\t\t"+str(len(series)), "total records in", \
nickname+bin_type, "series"
# Save to file
cons_outfile = file_root+bin_type+"_cons_out.npy"
numpy.save(cons_outfile, series)
# Evaluate match positions on reference
positions = []
match_read = []
for row in series:
# collect match read info while we're at it
# use regex to extract query index
query_pattern = re.compile(r'\w*_(\d*)')
query_match = query_pattern.match(row[0])
query_index = int(query_match.group(1))
match_read.append(query_index)
# use regex to extract ref coords
ref_pattern = re.compile(r'\w*_\d*_(\d*)')
ref_match = ref_pattern.match(row[1])
ref_pos = int(ref_match.group(1))
pos_scaled = ref_pos/cpm['size'] # adjust to db segment length
positions.append(pos_scaled)
# uniquify the match read array
unique_matches = numpy.unique(match_read)
print "\t"+str(len(unique_matches)), "unique matches for", bin_type
# write to file for future use
match_dir_root = dirs['match_dir']+nickname+"/"+ref_nick+"/"
ensure_dir(match_dir_root)
match_outfile = match_dir_root+nickname+bin_type+"_match.npy"
numpy.save(match_outfile, unique_matches)
# now count ocurrences per position
pos_np = numpy.array(positions)
binned = numpy.bincount(pos_np)
binned_pos.append(binned)
pos_count_average = numpy.average(binned)
averages.append((pos_count_average, series_index))
series_index +=1
# compare series
averages.sort()
averages.reverse()
order_indices = []
for pair in averages:
order_indices.append(pair[1])
# identify reference
ref_name = [reference['full_name'] for reference in references if
reference['nickname'] is ref_nick]
# prep directory & file
fig_root = dirs['reports_dir']+"match_figs/"
fig_file = fig_root+nickname+"_"+ref_nick+".png"
ensure_dir(fig_root)
# generate a figure
pylot.autoscale(enable=True, axis='both', tight=True)
pylot.xlabel('Position on the chromosome (/'+str(cpm['size'])+')')
pylot.ylabel('Number of matches (includes multiples)')
pylot.title(nickname+' matches to '+ref_name)
pylot.grid(True)
for index in order_indices:
label_root = nickname+bin_types[index]
label_str = label_root+" ("+str(numpy.sum(binned_pos[index]))+")"
pylot.plot(binned_pos[index], label=label_str)
pylot.legend(loc=1)
pylot.savefig(fig_file, dpi=None, facecolor='w', edgecolor='w',
orientation='portrait', papertype=None, format=None)
pylot.clf()
print "\t"+str(series_index), "series consolidated and parsed"
print "-- Done, see plot --"
print datetime.now()
return "OK"
def filter_contigs(run_ref, run_id, genomes, norm_matches, seg_size, threshold, r_root_dir, run_dirs, fixed_dirs, timestamp):
"""Filter contigs."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
fas_root = fixed_dirs['fas_contigs_dir']
report_root = run_root+run_dirs['reports']+ref_n+"/"
ensure_dir([report_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Filter contigs @", timestamp, "\n"])
run_ref.log(logstring)
# process
# evaluate segment specificity using negative controls
neg_controls = [genome['name'] for genome in genomes if ('ctrl' in genome.keys() and genome['ctrl'] == 'neg')]
neg_dat = [norm_matches[g_name]['ctg_scores'] for g_name in neg_controls]
neg_RA = np.vstack(neg_dat)
neg_mean = nanmean(neg_RA, axis=0)
# process the genomes we're testing
test_genomes = [genome['name'] for genome in genomes if not ('ctrl' in genome.keys())]
for g_name in test_genomes:
print "\t", g_name,
ctg_hits = norm_matches[g_name]['ctg_scores']
ctg_stats = {}
#process individual contigs
counter = 0
for ctg_RA in ctg_hits:
# identify this contig by name
ctg_name = norm_matches[g_name]['ctg_names'][counter]
counter += 1
# subtract background signal from match scores
recal_ctg_RA = np.subtract(ctg_RA, neg_mean)
recal_ctg_RA = recal_ctg_RA.clip(min=0)
# compute total similarity score
s_score = np.sum(recal_ctg_RA)
# compute clustering score (primitive)
streak = False
c_score = 0
for hit in recal_ctg_RA:
if hit == 0:
if streak == True:
c_score += -1
streak = False
else:
c_score += 0
elif hit > 0:
if streak == True:
c_score += 2
else:
c_score += 1
streak = True
# compute backbone vs. cargo burden
ctg_rec = load_fasta(fas_root+g_name+"/"+ctg_name+".fas")
bbone = np.sum(np.ma.make_mask(recal_ctg_RA))*seg_size
if bbone > len(ctg_rec):
bbone = len(ctg_rec) # workaround for last segment being always a little short
cargo = len(ctg_rec) - bbone
# make inverted array mask (used for redundancy detection)
ctg_mask = np.ma.getmaskarray(np.ma.masked_equal(recal_ctg_RA,0))
# consolidate contig information
ctg_stats[ctg_name] = {'s_score': s_score,
'c_score': c_score,
'vector': recal_ctg_RA,
'inv_mask':ctg_mask,
'bbone': bbone,
'cargo': cargo}
# detect redundant contigs
### use np.ma.mask_or(m1, m2)
### if any elements returns false there is a redundancy between two contigs
### if so evaluate which has better c_score and s_score
# compute overall stats for the genome
gs_score = sum([ctg_stats[contig]['s_score'] for contig in ctg_stats])
gc_score = sum([ctg_stats[contig]['c_score'] for contig in ctg_stats])
g_bbone = sum([ctg_stats[contig]['bbone'] for contig in ctg_stats])
g_cargo = sum([ctg_stats[contig]['cargo'] for contig in ctg_stats])
print gs_score, gc_score, g_bbone, g_cargo,
#
if gs_score > threshold:
## run plotters again
## pass the genome on to the next step (others will be dropped)
print "MATCH"
else:
print "(-)"
def align_cstrct2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs,
genomes, max_size, chop_mode, mtype, mauve_exec):
"""Align constructs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ref_ctg_file = run_ref.file
mauve_root = run_root+run_dirs['mauve_out_dir']+ref_n+"/constructs/"
segments_root = run_root+run_dirs['aln_seg_dir']+ref_n+"/constructs/"
scaff_root = run_root+run_dirs['scaffolds_dir']+ref_n+"/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Align scaffold constructs to reference @",
timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs
g_name = genome['name']
scaff_gbk = scaff_root+g_name+"_"+ref_n+"_scaffold.gbk"
file_list = (ref_ctg_file, scaff_gbk)
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# set outputs
mauve_dir = mauve_root+g_name+"/"
aln_segs_dir = segments_root+g_name+"/"
ensure_dir([mauve_dir, aln_segs_dir])
mauve_outfile = mauve_dir+g_name+"_"+ref_n+".mauve"
segfile = aln_segs_dir+g_name+"_"+ref_n+"_segs.txt"
# abort if the reference file is not found
try: open(ref_ctg_file, 'r')
except IOError:
msg = "ERROR: Reference file not found"
print msg
run_ref.log(msg)
raise
# abort if there is no scaffold construct
try: open(scaff_gbk, 'r')
except IOError:
msg = "WARNING: No scaffold construct to align"
print msg
run_ref.log(msg)
else:
# prep segments file
open(segfile, 'w').write('')
# purge any pre-existing sslist file
sslist_file = scaff_gbk+".sslist"
if os.path.isfile(sslist_file):
try: os.remove(sslist_file)
except Exception: raise
# do Mauve alignment
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile+".backbone", 0, mtype)
print len(coords), '->',
logstring = "".join(["\t", str(len(coords))])
run_ref.log(logstring)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode, mtype)
print len(chop_array), 'segments <', max_size, 'bp',
logstring = "".join(["\t", str(len(chop_array))])
run_ref.log(logstring)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_genbank(scaff_gbk)
id = iter_align(chop_array, ref_rec, query_rec,
aln_segs_dir, segfile)
print "@", id, "% id. overall"
logstring = "".join(["\t", str(id)])
run_ref.log(logstring)
except IOError:
msg = "\nERROR: Mauve alignment failed"
run_ref.log(msg)
print msg
def glompX_blast_out(genomes, run_ref, blast_mode, r_root_dir, run_dirs,
run_id, fixed_dirs, blast_dtypes, references,
min_nt_match, min_nt_score, min_nt_idp, min_aa_match,
min_aa_score, min_aa_idp, capture_span, timestamp):
"""Collect Blast results and extract match contigs."""
# load inputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
match_root = run_root+run_dirs['match_out_dir']+ref_n+"/"
capture_root = run_root+run_dirs['capture_dir']+ref_n+"/"
print " ", ref_n
# log
logstring = "".join(["\n\n# Collect Blast results @", timestamp, "\n\n"])
run_ref.log(logstring)
# collect results
ref_hits = {}
control_scores = []
run_ref.log("Segs/Gs\t")
run_ref.log("\t".join([genome['name'] for genome in genomes]))
for seg in run_ref.segs:
seg_n = seg['name']
print "\t", seg_n, "...",
run_ref.log("".join(["\n", seg_n]))
blast_dir = run_root+run_dirs['blast_out_dir']+ref_n+"/"+seg_n+"/"
capture_dir = capture_root+"/"+seg_n+"/"
ensure_dir([blast_dir, capture_dir])
ref_flag = True
for genome in genomes:
g_name = genome['name']
print "|",
# process
if g_name not in ref_hits.keys():
ref_hits[g_name] = {}
matches_dir = match_root+g_name+"/"
ensure_dir([matches_dir])
blast_infile = blast_dir+g_name+"_out.txt"
genome_ctg_dir = fixed_dirs['fas_contigs_dir']+g_name+"/"
rec_array = read_array(blast_infile, blast_dtypes)
if len(rec_array) > 0: # take qualified hits
p_cnt = 0
n_cnt = 0
if g_name in [ref['name'] for ref in references]:
copyfile(genome_ctg_dir+g_name+"_1.fas",
matches_dir+g_name+".fas")
if ref_flag:
# positive control TODO: better solution
control_scores.append(rec_array[0][11])
ref_flag = False
for line in rec_array:
idp = line[2]
q_start, q_stop = line[8], line[9]
score = line[11]
length = abs(q_stop-q_start)
# check the blast mode to use the right thresholds
if blast_mode == 'n' or blast_mode == 'tx':
min_match = min_nt_match
min_score = min_nt_score
min_idp = min_nt_idp
elif blast_mode == 'tn':
min_match = min_aa_match
min_score = min_aa_score
min_idp = min_aa_idp
else: # default to nucleotide mode
min_match = min_nt_match
min_score = min_nt_score
min_idp = min_nt_idp
if length>min_match and score>min_score and idp>min_idp:
print "+",
p_cnt +=1
contig_id = line[1]
if contig_id not in ref_hits[g_name].keys():
ref_hits[g_name][contig_id] = {seg_n: score}
else:
ref_hits[g_name][contig_id][seg_n] = score
pattern = re.compile(r'('+contig_id+')\.fas')
for item in listdir(genome_ctg_dir):
match = re.match(pattern, item)
if match:
fas_file = matches_dir+match.group(1)+".fas"
if not path.exists(fas_file):
copyfile(genome_ctg_dir+item, fas_file)
# context capture
capture_flag = False
while True:
try:
if int(seg_n) in run_ref.capture:
capture_flag = True
else:
break
except ValueError:
if seg_n in run_ref.capture:
capture_flag = True
else:
break
else:
break
if capture_flag:
# load the sequence
contig_file = matches_dir+contig_id+".fas"
contig_rec = load_fasta(contig_file)
# check orientation
if q_start < q_stop:
c_start = q_start-capture_span
c_stop = q_stop+capture_span
else:
c_start = q_stop-capture_span
c_stop = q_start+capture_span
print c_start, c_stop
# check limits
if c_start < 0:
c_start = 1
if c_stop > len(contig_rec.seq):
c_stop = len(contig_rec.seq)
# proceed
cxt_file = capture_dir+g_name+"_"+contig_id+".fas"
cxt_rec = SeqRecord(id=contig_id+"_"
+str(c_start)+"_"
+str(c_stop),
seq=contig_rec.seq
[c_start:c_stop])
write_fasta(cxt_file, cxt_rec)
else:
print "-",
n_cnt +=1
if n_cnt > 0:
logstring = "".join(["\t", str(p_cnt), " (",
str(n_cnt), ")"])
else:
logstring = "".join(["\t", str(p_cnt)])
run_ref.log(logstring)
else:
print "-",
run_ref.log("".join(["\t", "0"]))
print ""
return ref_hits, control_scores
def prep_maps(run_ref, run_id, timestamp, g_select, r_root_dir, run_dirs,
genomes, fixed_dirs, segtype, min_size, fct_flags,
fct_colors, idpt):
"""Set up generation of various maps."""
# set inputs and outputs
ref_ctg_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ref_gbk = run_ref.gbk
cst_root = run_root+run_dirs['scaffolds_dir']+ref_ctg_n+"/"
segments_root = run_root+run_dirs['aln_seg_dir']+ref_ctg_n+"/"
ctg_segs_root = segments_root+"contigs/"
cst_segs_root = segments_root+"constructs/"
maps_root = run_root+run_dirs['maps_dir']+ref_ctg_n+"/"
ctg_aln_maps_root = maps_root+"contig_alns/"
cst_ann_maps_root = maps_root+"constructs_annot/"
cst_aln_maps_root = maps_root+"constructs_aln/"
ensure_dir([cst_root, ctg_segs_root, cst_segs_root, maps_root,
ctg_aln_maps_root, cst_ann_maps_root, cst_aln_maps_root])
print " ", ref_ctg_n, "...",
# log
logstring = "".join(["\n\n# Generate maps @", timestamp, "\n\n"])
run_ref.log(logstring)
# map of reference with segment details
map_ref_segs(run_ref, run_id, r_root_dir, run_dirs, min_size,
fct_flags, fct_colors, idpt)
# log
logstring = "ref_map"
run_ref.log(logstring)
print logstring
# cycle through genomes
for genome in genomes:
# set inputs and outputs
g_name = genome['name']
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
while True:
try:
if g_name in g_select: pass
else: break
except TypeError:
pass
print "\t", g_name, "...",
scaff_gbk = cst_root+g_name+"_"+ref_ctg_n+"_scaffold.gbk"
ctg_aln_maps_dir = ctg_aln_maps_root+g_name+"/"
ensure_dir([ctg_aln_maps_dir])
# maps of contigs aligned to reference
logstring = "ctg_aln"
print logstring,
logstring = "".join(["\t", logstring])
run_ref.log(logstring)
map_ctg_alns(run_ref, ref_gbk, genome, ctg_segs_root,
ctg_aln_maps_dir, fixed_dirs, segtype, min_size,
fct_flags, fct_colors, idpt)
# map of scaffold construct
logstring = "cst_ant"
print logstring,
logstring = "".join(["\t", logstring])
run_ref.log(logstring)
map_cst_annot(run_ref, genome, scaff_gbk, cst_ann_maps_root,
fct_flags, fct_colors)
# map of construct aligned to reference
logstring = "cst_aln"
print logstring
logstring = "".join(["\t", logstring])
run_ref.log(logstring)
map_cst_aln(run_ref, ref_gbk, genome, scaff_gbk, cst_segs_root,
cst_aln_maps_root, segtype, min_size, fct_flags,
fct_colors, idpt)
break
def align_ctg2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs,
genomes, mauve_exec, max_size, chop_mode, mtype):
"""Align contigs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ref_ctg_file = run_ref.file
mauve_root = run_root+run_dirs['mauve_out_dir']+ref_n+"/contigs/"
segments_root = run_root+run_dirs['aln_seg_dir']+ref_n+"/contigs/"
q_ctgs_root = run_root+run_dirs['match_out_dir']+ref_n+"/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Align contigs to ref @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs and outputs
g_name = genome['name']
ctgs_fas_dir = q_ctgs_root+g_name+"/"
mauve_dir = mauve_root+g_name+"/"
aln_segs_root = segments_root+g_name+"/"
ensure_dir([mauve_dir])
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# list genbank files in matches directory
dir_contents = listdir(ctgs_fas_dir)
for item in dir_contents:
pattern = re.compile(r'.*_(\d*)\.fas$')
match = pattern.match(item)
if match:
ctg_num = match.group(1)
print ctg_num,
logstring = "".join(["\t", ctg_num])
run_ref.log(logstring)
# set inputs and outputs
q_contig = ctgs_fas_dir+item
file_list = (ref_ctg_file, q_contig)
mauve_outfile = mauve_dir+ctg_num+".mauve"
aln_segs_dir = aln_segs_root+ctg_num+"/"
ensure_dir([aln_segs_dir])
segfile = aln_segs_dir+ctg_num+"_"+ref_n+"_segs.txt"
open(segfile, 'w').write('')
# do Mauve alignment
try:
open(ref_ctg_file, 'r')
open(q_contig, 'r')
except IOError:
msg = "\nERROR: File missing, cannot align\n\t\t\t"
run_ref.log(msg)
print msg
else:
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile+".backbone",
0, mtype)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode,
mtype)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_fasta(q_contig)
iter_align(chop_array, ref_rec, query_rec,
aln_segs_dir, segfile)
except IOError:
msg = "\nERROR: Mauve alignment failed\n\t\t\t"
run_ref.log(msg)
print msg
except Exception:
msg = "\nERROR: Iteration failed\n\t\t\t"
run_ref.log(msg)
print msg
print ""
