def _collect_casava_qc(self):
qc_objects = []
runinfo_csv = os.path.join(os.path.join(self._meta.root_path, self.pargs.flowcell), "{}.csv".format(fc_id(self.pargs.flowcell)))
if not os.path.exists(runinfo_csv):
LOG.warn("No such file {}: trying fallback SampleSheet.csv".format(runinfo_csv))
runinfo_csv = os.path.join(os.path.join(self._meta.root_path, self.pargs.flowcell), "SampleSheet.csv")
try:
with open(runinfo_csv) as fh:
runinfo_reader = csv.reader(fh)
runinfo = [x for x in runinfo_reader]
except IOError as e:
self.app.log.warn(str(e))
raise e
fcdir = os.path.join(os.path.abspath(self._meta.root_path), self.pargs.flowcell)
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
## Check modification time
demux_stats = None
if modified_within_days(fcdir, self.pargs.mtime):
fc_kw = dict(fc_date = fc_date, fc_name=fc_name)
parser = FlowcellRunMetricsParser(fcdir)
fcobj = FlowcellRunMetricsDocument(fc_date, fc_name)
fcobj["RunInfo"] = parser.parseRunInfo(**fc_kw)
fcobj["RunParameters"] = parser.parseRunParameters(**fc_kw)
fcobj["illumina"] = parser.parse_illumina_metrics(fullRTA=False, **fc_kw)
fcobj["bc_metrics"] = parser.parse_bc_metrics(**fc_kw)
fcobj["undemultiplexed_barcodes"] = parser.parse_undemultiplexed_barcode_metrics(**fc_kw)
fcobj["illumina"].update({"Demultiplex_Stats" : parser.parse_demultiplex_stats_htm(**fc_kw)})
fcobj["samplesheet_csv"] = parser.parse_samplesheet_csv(runinfo_csv=runinfo_csv, **fc_kw)
demux_stats = fcobj["illumina"]["Demultiplex_Stats"]
qc_objects.append(fcobj)
qc_objects = self._parse_samplesheet(runinfo, qc_objects, fc_date, fc_name, fcdir, demultiplex_stats=demux_stats)
return qc_objects
def _collect_casava_qc(self):
qc_objects = []
runinfo_csv = os.path.join(os.path.join(self._meta.root_path, self.pargs.flowcell), "{}.csv".format(fc_id(self.pargs.flowcell)))
if not os.path.exists(runinfo_csv):
LOG.warn("No such file {}: trying fallback SampleSheet.csv".format(runinfo_csv))
runinfo_csv = os.path.join(os.path.join(self._meta.root_path, self.pargs.flowcell), "SampleSheet.csv")
try:
with open(runinfo_csv) as fh:
runinfo_reader = csv.reader(fh)
runinfo = [x for x in runinfo_reader]
except IOError as e:
self.app.log.warn(str(e))
raise e
fcdir = os.path.join(os.path.abspath(self._meta.root_path), self.pargs.flowcell)
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
## Check modification time
if modified_within_days(fcdir, self.pargs.mtime):
fc_kw = dict(fc_date = fc_date, fc_name=fc_name)
parser = FlowcellRunMetricsParser(fcdir)
fcobj = FlowcellRunMetricsDocument(fc_date, fc_name)
fcobj["RunInfo"] = parser.parseRunInfo(**fc_kw)
fcobj["RunParameters"] = parser.parseRunParameters(**fc_kw)
fcobj["illumina"] = parser.parse_illumina_metrics(fullRTA=False, **fc_kw)
fcobj["bc_metrics"] = parser.parse_bc_metrics(**fc_kw)
fcobj["undemultiplexed_barcodes"] = parser.parse_undemultiplexed_barcode_metrics(**fc_kw)
fcobj["illumina"].update({"Demultiplex_Stats" : parser.parse_demultiplex_stats_htm(**fc_kw)})
fcobj["samplesheet_csv"] = parser.parse_samplesheet_csv(runinfo_csv=runinfo_csv, **fc_kw)
qc_objects.append(fcobj)
qc_objects = self._parse_samplesheet(runinfo, qc_objects, fc_date, fc_name, fcdir, demultiplex_stats=fcobj["illumina"]["Demultiplex_Stats"])
return qc_objects
def upload_qc(self):
if not self._check_pargs(['flowcell']):
return
url = self.pargs.url if self.pargs.url else self.app.config.get(
"db", "url")
if not url:
self.app.log.warn("Please provide a valid url: got {}".format(url))
return
if not validate_fc_directory_format(self.pargs.flowcell):
self.app.log.warn(
"Path '{}' does not conform to bcbio flowcell directory format; aborting"
.format(self.pargs.flowcell))
return
runinfo_csv = os.path.join(os.path.abspath(self.pargs.flowcell),
"{}.csv".format(fc_id(self.pargs.flowcell)))
runinfo_yaml = os.path.join(os.path.abspath(self.pargs.flowcell),
"run_info.yaml")
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
if int(fc_date) < 120815:
self.log.info(
"Assuming pre-casava based file structure for {}".format(
fc_id(self.pargs.flowcell)))
qc_objects = self._collect_pre_casava_qc()
else:
self.log.info("Assuming casava based file structure for {}".format(
fc_id(self.pargs.flowcell)))
qc_objects = self._collect_casava_qc()
if len(qc_objects) == 0:
self.log.info("No out-of-date qc objects for {}".format(
fc_id(self.pargs.flowcell)))
return
else:
self.log.info("Retrieved {} updated qc objects".format(
len(qc_objects)))
s_con = SampleRunMetricsConnection(dbname=self.app.config.get(
"db", "samples"),
**vars(self.app.pargs))
fc_con = FlowcellRunMetricsConnection(dbname=self.app.config.get(
"db", "flowcells"),
**vars(self.app.pargs))
p_con = ProjectSummaryConnection(dbname=self.app.config.get(
"db", "projects"),
**vars(self.app.pargs))
for obj in qc_objects:
if self.app.pargs.debug:
self.log.debug("{}: {}".format(str(obj), obj["_id"]))
if isinstance(obj, FlowcellRunMetricsDocument):
dry("Saving object {}".format(repr(obj)), fc_con.save(obj))
if isinstance(obj, SampleRunMetricsDocument):
project_sample = p_con.get_project_sample(
obj.get("sample_prj", None), obj.get("barcode_name", None),
self.pargs.extensive_matching)
if project_sample:
obj["project_sample_name"] = project_sample['sample_name']
dry("Saving object {}".format(repr(obj)), s_con.save(obj))
def _collect_pre_casava_qc(self):
qc_objects = []
as_yaml = False
runinfo_csv = os.path.join(
os.path.join(self._meta.root_path, self.pargs.flowcell),
"{}.csv".format(fc_id(self.pargs.flowcell)))
if not os.path.exists(runinfo_csv):
LOG.warn("No such file {}: trying fallback SampleSheet.csv".format(
runinfo_csv))
runinfo_csv = os.path.join(
os.path.join(self._meta.root_path, self.pargs.flowcell),
"SampleSheet.csv")
runinfo_yaml = os.path.join(os.path.abspath(self.pargs.flowcell),
"run_info.yaml")
try:
if os.path.exists(runinfo_csv):
with open(runinfo_csv) as fh:
runinfo_reader = csv.reader(fh)
runinfo = [x for x in runinfo_reader]
else:
as_yaml = True
with open(runinfo_yaml) as fh:
runinfo = yaml.load(fh)
except IOError as e:
self.app.log.warn(str(e))
raise e
fcdir = os.path.abspath(self.pargs.flowcell)
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
## Check modification time
if modified_within_days(fcdir, self.pargs.mtime):
fc_kw = dict(fc_date=fc_date, fc_name=fc_name)
parser = FlowcellRunMetricsParser(fcdir)
fcobj = FlowcellRunMetricsDocument(**fc_kw)
fcobj["RunInfo"] = parser.parseRunInfo(**fc_kw)
fcobj["RunParameters"] = parser.parseRunParameters(**fc_kw)
fcobj["illumina"] = parser.parse_illumina_metrics(fullRTA=False,
**fc_kw)
fcobj["bc_metrics"] = parser.parse_bc_metrics(**fc_kw)
fcobj["filter_metrics"] = parser.parse_filter_metrics(**fc_kw)
fcobj["samplesheet_csv"] = parser.parse_samplesheet_csv(
runinfo_csv=runinfo_csv, **fc_kw)
fcobj["run_info_yaml"] = parser.parse_run_info_yaml(**fc_kw)
qc_objects.append(fcobj)
else:
return qc_objects
qc_objects = self._parse_samplesheet(runinfo,
qc_objects,
fc_date,
fc_name,
fcdir,
as_yaml=as_yaml)
return qc_objects
def upload_qc(self):
if not self._check_pargs(["flowcell"]):
return
url = self.pargs.url if self.pargs.url else self.app.config.get("db", "url")
if not url:
self.app.log.warn("Please provide a valid url: got {}".format(url))
return
if not validate_fc_directory_format(self.pargs.flowcell):
self.app.log.warn(
"Path '{}' does not conform to bcbio flowcell directory format; aborting".format(self.pargs.flowcell)
)
return
runinfo_csv = os.path.join(os.path.abspath(self.pargs.flowcell), "{}.csv".format(fc_id(self.pargs.flowcell)))
runinfo_yaml = os.path.join(os.path.abspath(self.pargs.flowcell), "run_info.yaml")
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
if int(fc_date) < 120815:
self.log.info("Assuming pre-casava based file structure for {}".format(fc_id(self.pargs.flowcell)))
qc_objects = self._collect_pre_casava_qc()
else:
self.log.info("Assuming casava based file structure for {}".format(fc_id(self.pargs.flowcell)))
qc_objects = self._collect_casava_qc()
if len(qc_objects) == 0:
self.log.info("No out-of-date qc objects for {}".format(fc_id(self.pargs.flowcell)))
return
else:
self.log.info("Retrieved {} updated qc objects".format(len(qc_objects)))
s_con = SampleRunMetricsConnection(dbname=self.app.config.get("db", "samples"), **vars(self.app.pargs))
fc_con = FlowcellRunMetricsConnection(dbname=self.app.config.get("db", "flowcells"), **vars(self.app.pargs))
p_con = ProjectSummaryConnection(dbname=self.app.config.get("db", "projects"), **vars(self.app.pargs))
for obj in qc_objects:
if self.app.pargs.debug:
self.log.debug("{}: {}".format(str(obj), obj["_id"]))
if isinstance(obj, FlowcellRunMetricsDocument):
dry("Saving object {}".format(repr(obj)), fc_con.save(obj))
if isinstance(obj, SampleRunMetricsDocument):
project_sample = p_con.get_project_sample(
obj.get("sample_prj", None), obj.get("barcode_name", None), self.pargs.extensive_matching
)
if project_sample:
obj["project_sample_name"] = project_sample["sample_name"]
dry("Saving object {}".format(repr(obj)), s_con.save(obj))
def _collect_pre_casava_qc(self):
qc_objects = []
as_yaml = False
runinfo_csv = os.path.join(
os.path.join(self._meta.root_path, self.pargs.flowcell), "{}.csv".format(fc_id(self.pargs.flowcell))
)
if not os.path.exists(runinfo_csv):
LOG.warn("No such file {}: trying fallback SampleSheet.csv".format(runinfo_csv))
runinfo_csv = os.path.join(os.path.join(self._meta.root_path, self.pargs.flowcell), "SampleSheet.csv")
runinfo_yaml = os.path.join(os.path.abspath(self.pargs.flowcell), "run_info.yaml")
try:
if os.path.exists(runinfo_csv):
with open(runinfo_csv) as fh:
runinfo_reader = csv.reader(fh)
runinfo = [x for x in runinfo_reader]
else:
as_yaml = True
with open(runinfo_yaml) as fh:
runinfo = yaml.load(fh)
except IOError as e:
self.app.log.warn(str(e))
raise e
fcdir = os.path.abspath(self.pargs.flowcell)
(fc_date, fc_name) = fc_parts(self.pargs.flowcell)
## Check modification time
if modified_within_days(fcdir, self.pargs.mtime):
fc_kw = dict(fc_date=fc_date, fc_name=fc_name)
parser = FlowcellRunMetricsParser(fcdir)
fcobj = FlowcellRunMetricsDocument(**fc_kw)
fcobj["RunInfo"] = parser.parseRunInfo(**fc_kw)
fcobj["RunParameters"] = parser.parseRunParameters(**fc_kw)
fcobj["illumina"] = parser.parse_illumina_metrics(fullRTA=False, **fc_kw)
fcobj["bc_metrics"] = parser.parse_bc_metrics(**fc_kw)
fcobj["filter_metrics"] = parser.parse_filter_metrics(**fc_kw)
fcobj["samplesheet_csv"] = parser.parse_samplesheet_csv(runinfo_csv=runinfo_csv, **fc_kw)
fcobj["run_info_yaml"] = parser.parse_run_info_yaml(**fc_kw)
qc_objects.append(fcobj)
else:
return qc_objects
qc_objects = self._parse_samplesheet(runinfo, qc_objects, fc_date, fc_name, fcdir, as_yaml=as_yaml)
return qc_objects
def _make_casava_archive_files(fc, ssname, prefix, startiter = 1, nseqout=1000):
fc_dir = os.path.join(ARCHIVE, fc)
if not os.path.exists(fc_dir):
safe_makedir(fc_dir)
with open(os.path.join(fc_dir, "{}.csv".format(ssname)), "w") as fh:
fh.write(SAMPLESHEETS[ssname])
with open(os.path.join(fc_dir, "RunInfo.xml"), "w") as fh:
fh.write(RUNINFO.render(**{'flowcell':os.path.basename(fc), 'fc_id':fc_id(fc), 'date':fc_parts(fc)[0], 'instrument':fc.split("_")[1]}))
with open(os.path.join(fc_dir, "runParameters.xml"), "w") as fh:
fh.write(RUNPARAMETERS.render(**{'flowcell':os.path.basename(fc), 'fc_id':fc_id(fc), 'date':fc_parts(fc)[0], 'instrument':fc.split("_")[1]}))
outf1 = []
outf2 = []
basecall_stats_dir = os.path.join(fc_dir, "Unaligned", "Basecall_Stats_{}".format(ssname))
if not os.path.exists(basecall_stats_dir):
safe_makedir(basecall_stats_dir)
for d in [os.path.join(basecall_stats_dir, x) for x in ["css", "Plots"]]:
if not os.path.exists(d):
safe_makedir(d)
for row in SAMPLESHEETS[ssname].split("\n"):
vals = row.split(",")
if vals[0] == "FCID":
header = row
continue
if len(vals) == 0:
continue
outdir = os.path.join(fc_dir, "Unaligned", "Project_{}".format(vals[5]), "Sample_{}".format(vals[2]))
if not os.path.exists(outdir):
safe_makedir(outdir)
with open(os.path.join(outdir, "SampleSheet.csv"), "w") as fh:
LOG.info("Writing to {}".format(os.path.join(outdir, "SampleSheet.csv")))
fh.write("{}\n".format(header))
fh.write("{}\n".format(row))
r1 = os.path.join(outdir, "{}_{}_L00{}_R1_001.fastq.gz".format(vals[2], vals[4], vals[1]))
r2 = os.path.join(outdir, "{}_{}_L00{}_R2_001.fastq.gz".format(vals[2], vals[4], vals[1]))
if os.path.exists(r1):
LOG.info("{} already exists: if you want to rerun file generation remove {}".format(r1, r1))
return
outf1.append(r1)
outf2.append(r2)
## Write sequences
with open("{}_1.fastq".format(prefix), "r") as fh:
_write_sample_fastq(fh, outf1, startiter=startiter, nseqout=nseqout)
with open("{}_2.fastq".format(prefix), "r") as fh:
_write_sample_fastq(fh, outf2, startiter=startiter, nseqout=nseqout)