def parse_directory(self):
"""Traverse a CASAVA 1.8+ generated directory structure and return a dictionary
"""
raise NotImplementedError("This method is not yet implemented")
projects = []
# Create a Flowcell object
fc = Flowcell()
fc.filename = self.samplesheet_file
fc.fc_id()
unaligned_dir = self.get_sequence_dir()
basecall_stats_dir = self.get_basecall_stats()
project_dir_pattern = os.path.join(unaligned_dir,"Project_*")
for project_dir in glob.glob(project_dir_pattern):
project_samples = []
sample_dir_pattern = os.path.join(project_dir,"Sample_*")
for sample_dir in glob.glob(sample_dir_pattern):
fastq_file_pattern = os.path.join(sample_dir,"*.fastq.gz")
samplesheet_pattern = os.path.join(sample_dir,"*.csv")
fastq_files = [os.path.basename(file) for file in glob.glob(fastq_file_pattern)]
samplesheet = glob.glob(samplesheet_pattern)
assert len(samplesheet) == 1, "ERROR: Could not unambiguously locate samplesheet in %s" % sample_dir
sample_name = sample_dir.replace(sample_dir_pattern[0:-1],'')
project_samples.append({'sample_dir': os.path.relpath(sample_dir,project_dir), 'sample_name': sample_name, 'files': fastq_files, 'samplesheet': os.path.basename(samplesheet[0])})
project_name = project_dir.replace(project_dir_pattern[0:-1],'')
projects.append({'project_dir': os.path.relpath(project_dir,unaligned_dir), 'project_name': project_name, 'samples': project_samples})
return {'fc_dir': fc_dir, 'fc_name': fc_name, 'fc_date': fc_date, 'data_dir': os.path.relpath(unaligned_dir,fc_dir), 'basecall_stats_dir': basecall_stats_dir, 'projects': projects}
def parse_directory(self):
"""Traverse a CASAVA 1.8+ generated directory structure and return a dictionary
"""
raise NotImplementedError("This method is not yet implemented")
projects = []
# Create a Flowcell object
fc = Flowcell()
fc.filename = self.samplesheet_file
fc.fc_id()
unaligned_dir = self.get_sequence_dir()
basecall_stats_dir = self.get_basecall_stats()
project_dir_pattern = os.path.join(unaligned_dir, "Project_*")
for project_dir in glob.glob(project_dir_pattern):
project_samples = []
sample_dir_pattern = os.path.join(project_dir, "Sample_*")
for sample_dir in glob.glob(sample_dir_pattern):
fastq_file_pattern = os.path.join(sample_dir, "*.fastq.gz")
samplesheet_pattern = os.path.join(sample_dir, "*.csv")
fastq_files = [
os.path.basename(file)
for file in glob.glob(fastq_file_pattern)
]
samplesheet = glob.glob(samplesheet_pattern)
assert len(
samplesheet
) == 1, "ERROR: Could not unambiguously locate samplesheet in %s" % sample_dir
sample_name = sample_dir.replace(sample_dir_pattern[0:-1], '')
project_samples.append({
'sample_dir':
os.path.relpath(sample_dir, project_dir),
'sample_name':
sample_name,
'files':
fastq_files,
'samplesheet':
os.path.basename(samplesheet[0])
})
project_name = project_dir.replace(project_dir_pattern[0:-1], '')
projects.append({
'project_dir':
os.path.relpath(project_dir, unaligned_dir),
'project_name':
project_name,
'samples':
project_samples
})
return {
'fc_dir': fc_dir,
'fc_name': fc_name,
'fc_date': fc_date,
'data_dir': os.path.relpath(unaligned_dir, fc_dir),
'basecall_stats_dir': basecall_stats_dir,
'projects': projects
}
def _from_pre_casava_structure(self):
if not self._check_pargs(["project", "flowcell"]):
return
fc = Flowcell()
fc.load([os.path.join(x, self.pargs.flowcell) for x in [self.config.get("archive", "root"), self.config.get("production", "root")]])
indir = os.path.join(self.config.get("production", "root"), self.pargs.flowcell)
if not fc:
self.log.warn("No run information available for {}".format(self.pargs.flowcell))
return
fc_new = fc.subset("sample_prj", self.pargs.project)
fc_new.collect_files(indir)
return fc_new
def _from_pre_casava_structure(self):
if not self._check_pargs(["project", "flowcell"]):
return
fc = Flowcell()
fc.load([os.path.join(x, self.pargs.flowcell) for x in [self.config.get("archive", "root"), self.config.get("production", "root")]])
indir = os.path.join(self.config.get("production", "root"), self.pargs.flowcell)
if not fc:
self.log.warn("No run information available for {}".format(self.pargs.flowcell))
return
fc_new = fc.subset("sample_prj", self.pargs.project)
fc_new.collect_files(indir)
return fc_new
def _from_casava_structure(self):
"""Get information from casava structure"""
if not self._check_pargs(["project"]):
return
fc_list = []
pattern = "-bcbb-config.yaml$"
def bcbb_yaml_filter(f):
return re.search(pattern, f) != None
samples = filtered_walk(os.path.join(self._meta.root_path, self._meta.path_id), bcbb_yaml_filter)
for s in samples:
fc = Flowcell(s)
fc_new = fc.subset("sample_prj", self.pargs.project)
fc_new.collect_files(os.path.dirname(s))
fc_list.append(fc_new)
return fc_list
def _from_casava_structure(self):
"""Get information from casava structure"""
if not self._check_pargs(["project"]):
return
fc_list = []
pattern = "-bcbb-config.yaml$"
def bcbb_yaml_filter(f):
return re.search(pattern, f) != None
samples = filtered_walk(os.path.join(self._meta.root_path, self._meta.path_id), bcbb_yaml_filter)
for s in samples:
fc = Flowcell(s)
fc_new = fc.subset("sample_prj", self.pargs.project)
fc_new.collect_files(os.path.dirname(s))
fc_list.append(fc_new)
return fc_list
def samplesheet_csv_to_yaml(fn):
"""Convert SampleSheet.csv to bcbb-config.yaml file.
:param fn: input file
"""
fc = Flowcell(infile=fn)
bc_id = 1
for s in fc.samples:
sequence = fc.get_entry(s, "sequence")
name = fc.get_entry(s, "name")
# Currently only look for casava-based pattern
pat = os.path.join(os.path.dirname(fn), "{}_{}*fastq*".format(name, sequence))
seqfiles = glob.glob(pat)
if seqfiles:
seqfiles.sort()
fc.set_entry(s, "files", seqfiles)
fc.set_entry(s, "barcode_id", bc_id)
if bc_id == 1:
flowcell_id = fc.get_entry(s, "flowcell_id").split("-")[1]
bc_id = bc_id + 1
fc.fc_date = datetime.datetime.now().strftime("%y%m%d")
fc.fc_name = flowcell_id
outfile = os.path.join(os.path.dirname(fn), "{}-bcbb-config.yaml".format(name))
with open(outfile, "w") as fh:
fh.write(fc.as_yaml())
def samplesheet_csv_to_yaml(fn):
"""Convert SampleSheet.csv to bcbb-config.yaml file.
:param fn: input file
"""
fc = Flowcell(infile=fn)
bc_id = 1
for s in fc.samples:
sequence = fc.get_entry(s, "sequence")
name = fc.get_entry(s, "name")
# Currently only look for casava-based pattern
pat = os.path.join(os.path.dirname(fn), "{}_{}*fastq*".format(name, sequence))
seqfiles = glob.glob(pat)
if seqfiles:
seqfiles.sort()
fc.set_entry(s, "files", seqfiles)
fc.set_entry(s, "barcode_id", bc_id)
if bc_id == 1:
flowcell_id = fc.get_entry(s, "flowcell_id").split("-")[1]
bc_id = bc_id + 1
fc.fc_date = datetime.datetime.now().strftime("%y%m%d")
fc.fc_name = flowcell_id
outfile = os.path.join(os.path.dirname(fn), "{}-bcbb-config.yaml".format(name))
with open(outfile, "w") as fh:
fh.write(fc.as_yaml())