else:
if not "--calibration-training=" in prebasecallerArgs:
prebasecallerArgs = prebasecallerArgs + " --calibration-training=100000"
if not "--flow-signals-type" in prebasecallerArgs:
prebasecallerArgs = prebasecallerArgs + " --flow-signals-type scaled-residual"
basecaller.basecalling(
env['SIGPROC_RESULTS'], prebasecallerArgs,
env['libraryKey'], env['tfKey'], env['runID'],
env['reverse_primer_dict'],
os.path.join(env['BASECALLER_RESULTS'], 'recalibration'),
env['barcodeId'], env['barcodeInfo'], env['library'],
env['notes'], env['site_name'], env['platform'],
env['instrumentName'], env['chipType'])
basecaller_recalibration_datasets = blockprocessing.get_datasets_basecaller(
os.path.join(env['BASECALLER_RESULTS'], 'recalibration'))
if env['doBaseRecal'] == "panel_recal":
basecaller_recalibration_datasets = basecaller_recalibration_datasets[
'IonControl']
# file containing dimension info (offsets, rows, cols) and flow info for stratification
try:
c = open(
os.path.join(env['BASECALLER_RESULTS'],
"recalibration", 'BaseCaller.json'), 'r')
chipflow = json.load(c)
c.close()
except:
traceback.print_exc()
raise
try:
printtime("INFO: merging rawtf.basecaller.bam")
block_bam_list = [os.path.join(adir, env['BASECALLER_RESULTS'], 'rawtf.basecaller.bam') for adir in dirs]
block_bam_list = [block_bam_filename for block_bam_filename in block_bam_list if os.path.exists(block_bam_filename)]
composite_bam_filename = os.path.join(env['BASECALLER_RESULTS'], 'rawtf.basecaller.bam')
if block_bam_list:
blockprocessing.merge_bam_files(block_bam_list,composite_bam_filename,composite_bai_filepath="",mark_duplicates=False,method='picard')
except:
print traceback.format_exc()
printtime("ERROR: merging rawtf.basecaller.bam unsuccessful")
if do_unfiltered_processing:
basecaller_datasets = blockprocessing.get_datasets_basecaller(env['BASECALLER_RESULTS'])
try:
os.mkdir(os.path.join(env['BASECALLER_RESULTS'],'unfiltered.untrimmed'))
basecaller.merge_datasets_basecaller_json(
dirs,
os.path.join(env['BASECALLER_RESULTS'],"unfiltered.untrimmed"))
blockprocessing.merge_bams(
dirs,
os.path.join(env['BASECALLER_RESULTS'],"unfiltered.untrimmed"),
basecaller_datasets,
'picard')
except:
print traceback.format_exc()
prebasecallerArgs,
env['libraryKey'],
env['tfKey'],
env['runID'],
env['reverse_primer_dict'],
os.path.join(env['BASECALLER_RESULTS'], 'recalibration'),
env['barcodeId'],
env['barcodeInfo'],
env['library'],
env['notes'],
env['site_name'],
env['platform'],
env['instrumentName'],
env['chipType'])
basecaller_recalibration_datasets = blockprocessing.get_datasets_basecaller(os.path.join(env['BASECALLER_RESULTS'],'recalibration'))
if env['doBaseRecal'] == "panel_recal":
basecaller_recalibration_datasets = basecaller_recalibration_datasets['IonControl']
# file containing dimension info (offsets, rows, cols) and flow info for stratification
try:
c = open(os.path.join(env['BASECALLER_RESULTS'], "recalibration", 'BaseCaller.json'),'r')
chipflow = json.load(c)
c.close()
except:
traceback.print_exc()
raise
# Recalibrate
for dataset in basecaller_recalibration_datasets["datasets"]:
"rawtf.basecaller.bam")
if block_bam_list:
blockprocessing.merge_bam_files(
block_bam_list,
composite_bam_filename,
composite_bai_filepath="",
mark_duplicates=False,
method="picard",
)
except Exception:
print(traceback.format_exc())
printtime("ERROR: merging rawtf.basecaller.bam unsuccessful")
if do_unfiltered_processing:
basecaller_datasets = blockprocessing.get_datasets_basecaller(
env["BASECALLER_RESULTS"])
try:
os.mkdir(
os.path.join(env["BASECALLER_RESULTS"],
"unfiltered.untrimmed"))
basecaller.merge_datasets_basecaller_json(
dirs,
os.path.join(env["BASECALLER_RESULTS"],
"unfiltered.untrimmed"),
)
blockprocessing.merge_bams(
dirs,
os.path.join(env["BASECALLER_RESULTS"],
env["instrumentName"],
env["chipInfo"],
)
# Reuse phase estimates in main base calling task
additional_basecallerArgs += " --phase-estimation-file " + os.path.join(
env["BASECALLER_RESULTS"], "recalibration",
"BaseCaller.json")
#
# Part 2) Create aligned training BAMs as input for Calibration module
# Or skip alignment if we want to use blind calibration
#
calibration_input_bams = ""
basecaller_recalibration_datasets = blockprocessing.get_datasets_basecaller(
os.path.join(env["BASECALLER_RESULTS"], "recalibration"))
if calibration_mode == "panel_recal":
basecaller_recalibration_datasets = basecaller_recalibration_datasets[
"IonControl"]
for dataset in basecaller_recalibration_datasets["datasets"]:
if not dataset.get("read_count", 0) > 0:
continue
if "nomatch" in dataset["basecaller_bam"]:
continue
basecaller_bam = os.path.join(
env["BASECALLER_RESULTS"],
"recalibration",
dataset["basecaller_bam"],
