def flat_directories(self):
"""The directories in this directory non-recursively, and sorted"""
for root, dirs, files in os.walk(self.path):
result = [DirectoryPath(os.path.join(root, d)) for d in dirs]
break
result.sort(key=lambda x: natural_sort(x.path))
return result
def load(self):
"""A delayed kind of __init__ that is not called right away to avoid
crowding the RAM of the python interpreter when you just import gefes"""
# Load #
self.loaded = True
# Automatic paths #
self.p = AutoPaths(self.base_dir, self.all_paths)
# Make an alias to the json #
self.json_path.link_to(self.p.info_json, safe=True)
# Make all the samples object that this project possesses #
if self.info.get('auto_parse_samples'):
search_dir = self.info['samples_base_dir']
search_dir = os.path.expanduser(search_dir)
paths = glob.glob(search_dir + '*.fastq*')
if not paths: paths = glob.glob(search_dir + '*.fasta*')
if not paths: raise Exception("Found no FASTA or FASTQ path in %s" % search_dir)
if all([re.search("_R[12]_", p) for p in paths]): pairs = join_paired_filepaths(paths)
else: pairs = sort_string_by_pairs(paths)
self.samples = [Sample(self, p[0], p[1], num=i) for i,p in enumerate(pairs)]
self.samples.sort(key=lambda x: natural_sort(x.name))
else:
self.samples = [Sample(self, info=info) for info in self.info['samples']]
# The samples of a project are it's children in a way #
self.children = self.samples
# Call the mother function #
return Aggregate.load(self)
def __init__(self, json_path, input_dir, output_dir):
# Attributes #
self.json_path = FilePath(json_path)
self.input_dir = input_dir
self.output_dir = output_dir
# Name #
self.name = self.json_path.prefix
# Auto paths #
self.base_dir = output_dir + 'experiment/'
self.samples_dir = output_dir + 'samples/'
self.p = AutoPaths(self.base_dir, self.all_paths)
# Get sample json file #
with open(json_path) as handle: self.info = json.load(handle)
# Load all samples #
self.samples = []
for name, desc in self.info.items():
if desc['type'] == 'humic': self.samples.append(Humic(name, desc, self))
if desc['type'] == 'fresh': self.samples.append(Fresh(name, desc, self))
self.samples.sort(key = lambda x: natural_sort(x.name))
self.children = self.samples
# Samples #
self.humic_samples = [s for s in self.samples if s.type == 'humic']
self.fresh_samples = [s for s in self.samples if s.type == 'fresh']
# Other #
self.demultiplexer = Demultiplexer(self)
#self.taxa_analysis = TaxonomyAnalysis(self)
#self.meta_analysis = MetabolismAnalysis(self)
# Runner #
self.runner = ExperimentRunner(self)
def flat_directories(self):
"""The directories in this directory non-recursively, and sorted"""
for root, dirs, files in os.walk(self.path):
result = [DirectoryPath(os.path.join(root, d)) for d in dirs]
break
result.sort(key=lambda x: natural_sort(x.path))
return result
def flat_files(self):
"""The files in this directory non-recursively, and sorted.
#TODO: check for permission denied in directory."""
result = []
for root, dirs, files in os.walk(self.path):
result = [FilePath(os.path.join(root, f)) for f in files]
break
result.sort(key=lambda x: natural_sort(x.path))
return result
def __init__(self, samples):
# Save samples #
self.samples = [s for s in samples if s.sff_files_info]
# Get all unique SFF files #
self.sff_paths = set([f['path'] for s in self.samples for f in s.sff_files_info])
# Make objects of them #
self.sff_files = [MultiplexedSFF(path) for path in self.sff_paths]
self.sff_files.sort(key = lambda x: natural_sort(x.name))
# Two way linked graph #
for sample in self.samples: sample.sff_links = OrderedDict()
for sff in self.sff_files: sff.sample_links = OrderedDict()
# Draw links #
for sample in self.samples:
for f in sample.sff_files_info:
barcode, path = f['mid'], f['path']
sff_obj = [sff for sff in self.sff_files if sff.path==path][0]
sample.sff_links[barcode] = sff_obj
sff_obj.sample_links[barcode] = sample
def __init__(self, experiment):
# Save parent #
self.parent, self.experiment = experiment, experiment
# Only for humic samples #
self.samples = self.parent.humic_samples
# Get all unique SFF files #
self.sff_paths = set([f['path'] for s in self.samples for f in s.multiplexed_files])
# Make objects of them #
self.sff_files = [Multiplexed(path) for path in self.sff_paths]
self.sff_files.sort(key = lambda x: natural_sort(x.name))
# Two way linked graph #
for sample in self.samples: sample.sff_links = {}
for sff in self.sff_files: sff.sample_links = {}
# Draw links #
for sample in self.samples:
for f in sample.multiplexed_files:
barcode, path = f['mid'], f['path']
sff_obj = [sff for sff in self.sff_files if sff.path==path][0]
sample.sff_links[barcode] = sff_obj
sff_obj.sample_links[barcode] = sample
